Category: Customer Experience

Posted on

People want a “Human Touch,” so will AI in CX succeed?

LLMs

According to the Chinese calendar, 2025 was the year of the snake (I’ll let you put a metaphor to that).  2026 is the year of the flaming horse, symbolizing intense energy, passion, rapid change, and independence.  No doubt we have lots of rapid change going on technologically, and it’s easy to argue that AI is asserting independence.  Try calling or chatting with customer service in nearly any business, and chances are it starts with a conversation with AI.  But how is that working out for you?

The level of hype in the marketplace over AI continues to climb, with no peak in sight.  Yet interestingly, in a recent Pegasystems worldwide survey of 4700 consumers, 77% said they achieve better outcomes when dealing with real humans.  Real humans prefer real humans.  Go figure.  And perhaps that’s one thing we will hold onto as AI evolves.  The implications are profound given this paradox between technological advancement and human interaction.

Ironically, as technologies improve, in the not too distant future these same consumers may at times not even realize they’re actually conversing with a machine. And stranger still, once things blur further, the whole notion of a human touch may be redefined as partially or fully delivered by machines. 

Meanwhile, most complex CRM journeys still require, at one stage or another, interfacing with carbon-based life forms.  As businesses design the transition and inexorably march toward using technology for increased efficiency and more digital engagements, how can they ensure consumers get what they want – relevant, rewarding, and timely brand exchanges?

On the surface, these two seemingly opposed forces appear to necessitate tradeoffs where one approach wins over the other.  A superior solution, however, may involve deploying hybrid AI in selective and meaningful ways.  Pure technological AI (Automated Intelligence) might replace humans in some cases, but using AI to augment human functioning may prove equally valuable, where its role amplifies human capabilities and minimizes human limitations.

AI Automated Intelligence

What is a “Human Touch?”

Consider these aspects of an agents’ human touch (where an agent can be a human or machine):

  • The agent sounds human & conveys empathy (in contrast, Alexa doesn’t sound human yet).
  • The agent listens (senses), comprehends, and suggests (responds) reasonable courses of action based on the contextual understanding of the current conversation.
  • The agent displays good judgement, resourcefulness, and common sense in real-time problem solving; achieving status as a strategic business partner.
  • The agent builds a level of trust with the customer; achieving status as a trusted advisor.
  • The agent can personalize, making the customer feel like a special individual.
  • The agent builds a relationship with the customer by telling stories related to the conversation, drawing from experience, and using that to relate to the individual.
  • The agent recalls important details of previous conversations.

All these may sound difficult for machines to mimic, however advances in artificial intelligence (especially LLMs – Large Language Models) are beginning to bridge the gap.  When asked, most consumers are sanguine about its use when implemented effectively, and if it acts in a more human-like manner, they may not care who is servicing them provided they consistently get what they want.  So what is it that humans want?

Humans want:

  • Short or no wait times
  • Accessible service (when they need it and wherever they are)
  • Fast service once the process starts.  A sense of urgency to solve problems
  • Coordinated, coherent, and seamless service (ability to switch channels, and not have to start over)
  • Personalized products, services, and support
  • Accurate & fast answers to their questions
  • A warm and friendly experience that is welcoming, enjoyable and leaves them feeling special

In short, a memorable and extraordinary experience that is story worthy.

Pitfalls to avoid when applying AI in CX

In planning for success, it’s critical to consider what to avoid, minimizing unnecessary mistakes.  To improve machine or human performance, learn from the mistakes of others.  Here are some pitfalls to sidestep:

  • Tendency to over automate before you carefully assess the affect total automation may have on customer experience.  It may be improving the bottom line (in the short run), but is it ultimately improving CX, or making it worse?
  • Placing a premium on being cute with technology or customer information, while not focusing on whether the outcome is actually ideal.  Scrutinize the value of the technology.  Improper use can backfire, such as wishing a customer happy birthday when they never gave you permission to gather and use their birthdate.
  • Loosing sight of the root cause and fixing it.  Why did the customer ask for support in the first place?
  • Designing programs based on the law of averages versus factoring in individual customer preferences & valuation. Remember, some important customers may require human interaction, and it might be economically justified to provide just that.
  • Falling victim to automation bias – Becoming sloppy, complacent, insensitive, and dulled, because machines take care of so many customer service tasks – and when humans are called on to provide service, they can’t – due to being rusty, incompetent, or rendered totally incapable.

5 Ways to employ AI in CX

  • Automate the no brainers.  For example, use automation & intelligence to lookup routine customer information, answer frequently asked questions (FAQs), get order status, or even process a payment.
  • Use automation & intelligence to classify and route emails, calls, and other requests.  They’ll get faster to the right people that can ultimately close out the case.
  • Augment staff with intelligence, such as using a deep knowledge base, with AI filtering and learning capabilities, to rank likely answers to FAQs, and provide those to staff so they can quickly answer questions, while still providing a human touch.
  • Take interest in what matters to the customer and knowing things about them pertinent to the relationship.  Use AI to help store and recall critical material at the right moment, and let humans decide how and when to weave that into conversation for a natural flow.
  • Ensure warm handoffs between self-service technology and the humans who might have to complete the servicing.  For example, when a customer engages in self-service, but then escalates, guarantee a comprehensive and seamless transition of the self-service transaction to human agents.

Follow advances in AI technologies that apply a human touch:

  • Experiment with weaving in LLM technology – this coincides with striving for an overall system that is friendly, inviting, warm, and conversational.  Simultaneously guard against AI’s detrimental potential, such as being discriminatory.
  • Test using prescriptive intelligence techniques that incrementally improve relationships.
  • Use real-time event processing technologies that leverage in the moment contextual data.
  • Select one platform to be the always-on brain, and use it as the corporate customer memory bank, insight generator, and engagement hub – the dependable friend or relative that always remembers.
  • Carefully construct overall experiences using automation, intelligence, and human beings together delivering an optimal result.

We’re all developing relationships with machines and already extremely dependent on them. We talk to devices, sometimes raising our voices, use them as our assistants, and even listen to and laugh at their jokes.  In many cases, they already appear to be our companions.

AI is here to stay.  Make no mistake – it’s going to replace some jobs, and take over many tasks, just as the industrial revolution did.  But new jobs will form, and those that embrace it will replace those that don’t.  However, where social skills are paramount, such as in handling complex customer service and social work, forecasters expect the impact to be less.

As machines and their makers relish monopolizing and optimizing commercial interactions for convenience and efficiency, the true winners will be those that realize we reward businesses that deliver delightful interactions.  Great brands will be those that devise seamless experiences that weave together machines and humans, tuned to each individual’s preference for the synthetic and humanoid mix.

As CX experts involved in shaping these AI and human experiences, design them wisely with these considerations in mind.

Post blog note:

Full transparency: I wrote the first version of this blog in 2017 on Customer Think.   I updated it on my blog site in 2024, and now here again.   The key tenants have remained the same for nearly 10 years. 

Posted on

AI in Customer Experience: A 2025 Update

AI in CX: The Comet Has Returned—And This Time, It Exploded

Back in January 2017, I wrote an article – AI in CX – likening artificial intelligence (AI) to a comet—repeatedly flaring into view with dazzling promise, only to fade into the cold darkness of unmet expectations. At the time, AI had delivered some real value, mostly in the form of automation and predictive analytics, but much of it still felt superficial. The tools were impressive, but not yet transformative. I concluded that while AI had made progress, the real revolution was still ahead.

That revolution arrived in November 2022.

When OpenAI released ChatGPT to the public, it marked a seismic shift in how businesses and consumers perceived and interacted with AI. For the first time, AI wasn’t just a backend tool for marketers or a silent assistant in your smartphone. It was front and center—conversational, creative, and shockingly capable. The AI comet didn’t just return; it crash landed.

From Automation to Augmentation

In 2017, AI in CX was largely about automation and using traditional statistical models for predictions. We saw gains in intelligent routing, basic chatbots, and predictive recommendations. These were (and still are) valuable, but they didn’t fundamentally change the ability to create content at scale for 1 to 1 personalization and to achieve better customer experience.

Fast forward to 2025, and we’re in a different world. AI is no longer just powering decision making at run-time, improving customer experience in digital and physical channels, and powering content production at design time. Large language models (LLMs) like GPT-4 and its successors are now embedded in customer service platforms, sales enablement tools, marketing engines, and in marketing content operations solutions. They don’t just deflect simple queries—they resolve complex issues through dynamic conversations, generate personalized content, and even detect customer sentiment in real time.

Consider the modern contact center. AI agents now handle the majority of customer interactions, not with rigid scripts, but with fluid, empathetic dialogue. They understand nuance, context, and intent. When escalation is needed, they summarize the issue for human agents, reducing resolution time and improving satisfaction. This isn’t superficial intelligence—it’s deeply integrated, high-value augmentation.

The Rise of Personalization at Scale

One of the most profound shifts since 2017 is the rise of personalization at scale – meaning having something unique and differentiated to say and show to each customer – and figuring that out and delivering it in real time. Back then, personalization meant inserting a customer’s name into an email or recommending products based on past purchases. Today, AI crafts entire customer journeys tailored to individual preferences, behaviors, and even emotional states. AI also detects sentiment, context, and intent – synthesizes these insights, and factors them into decisions – again all in real time.

LLMs analyze vast amounts of unstructured data—emails, chat logs, social media posts—to build dynamic customer profiles. These profiles inform every touchpoint, from marketing messages to service interactions. The result? Customers feel seen, understood, and valued.

This level of personalization was once the holy grail of CX – a theory in books. Now, it’s attainable in practice.

Conversational Interfaces: The New Front Door

In 2017, I pointed to early chatbot experiments and voice assistants as promising but limited. They often frustrated more than they helped. Today, conversational interfaces are the new front door to many brands.

Whether it’s a voice assistant embedded in a car, a chatbot on a website, or a virtual agent in a mobile app, customers increasingly prefer to “talk” to businesses. And thanks to LLMs, those conversations feel natural, helpful, and even enjoyable.

These interfaces are also multimodal—capable of understanding text, voice, images, and even video. A customer can snap a photo of a broken product, describe the issue in natural language, and receive a resolution—all without speaking to a human.

Trust, Transparency, and the Human Touch

Despite these advances, the human element remains critical. In fact, as AI becomes more capable, the need for trust and transparency grows. Customers want to know when they’re interacting with AI. When that experience is frustrating – taking too long – they want to be quickly connected to a human – and they want context to be remembered and factored in. They want explanations for decisions – especially AI-driven ones.

Leading brands are embracing this by designing hybrid experiences—AI handles the routine, while humans focus on empathy, quality control, creativity, and complex problem-solving. This synergy is where the real magic happens. Great brands get this, and have expert humans monitoring AI, and quickly course correcting it – always with the consumer’s experience in mind.

The Organizational Challenge

As I noted in 2017, the biggest barrier to AI adoption isn’t the technology—it’s the organization. That’s still true today. Many companies struggle to integrate AI into their workflows, data systems, and culture. They chase shiny tools without rethinking their processes, understanding the use cases, or upskilling their teams.

The winners in this new era are those who treat AI not as a bolt-on, but as a core capability. They invest in data governance, cross-functional collaboration, and continuous learning. They empower employees to work alongside AI, not fear it.

What’s Next?

So, where do we go from here – now that all of the AI children are growing up?

One of the next frontiers is emotional intelligence. AI is getting better at detecting and responding to human emotions. Sentiment analysis is evolving into affective computing—systems that can sense frustration, joy, or confusion and adapt accordingly. This will unlock new levels of empathy in digital interactions.

We’ll also see more proactive AI. Instead of waiting for customers to reach out, AI will anticipate needs and offer help before problems arise. Think of it as predictive CX—where the best service is the one you never needed to ask for.

And we’ll continue to see AI as a content production assistant, helping brands put more of the right content on the inventory shelf, and then using AI to help just-in-time assemble the right piece of content for the right person, and quickly test and learn which ones resonate.

And finally, we’ll see AI become more ethical and accountable. As regulations evolve and public scrutiny increases, businesses will need to ensure their AI systems are fair, explainable, and aligned with human values.

Conclusion: The Comet Has Become a Constant

In 2017, I ended with a prediction: the AI comet would return. It did—and this time, it’s not fading away into an AI winter – it’s here to stay. AI is no longer a novelty or a niche. It’s a foundational technology reshaping how we live, work, and connect.

For customer experience professionals, the message is clear: embrace the new normal. Invest in real intelligence, not superficial gimmicks. Focus on outcomes, not just outputs. And above all, remember that the goal of AI isn’t to replace humans—it’s to help us be more human to more people – at scale.

The comet has crash landed. Let’s build something extraordinary with the rare materials that came with it.

Posted on

Agentic CRM: Magic Wand or more Smoke and Mirrors?

In the CRM world, the cycle of reinvention is relentless. Old ideas are rarely retired — they’re rebranded, rewrapped, and reintroduced with a flourish of marketing fanfare. It’s a bit like watching a magician pull the same rabbit out of a slightly shinier hat. This year’s CRM buzzword bingo has a new square: “Agentic.” And now, the latest rabbit? Agentic CRM.

It’s the promise of AI agents that don’t just assist — they do, decide, and act. Independently. Autonomously. Flawlessly, if you believe the press releases. But before we all start handing over the keys to our customer journeys, let’s pause and ask: is this truly the dawn of a new era, or just another round of CRM and Marketing Automation hocus pocus?

Enter the 800-pound Agents: AgentForce, Agent Orchestrator, and the Copilots

Salesforce, never one to miss a branding opportunity, has unveiled “AgentForce.” The name alone suggests a squadron of tireless digital workers, ready to execute tasks and rule without oversight. Sounds impressive—until you dig into the details. What exactly are these agents doing?  Who is managing them and what decisions are they making?  And more importantly, how well are they really performing their jobs?

From what’s been shared, AgentForce appears to be a mash-up of existing Einstein capabilities, some Flow automation, and a sprinkle of LLM-powered task execution. But is it truly autonomous? Or is it just a glorified macro with a chatbot interface?

Adobe, not to be outdone, this year launched “Agent Orchestrator.” The name evokes a symphony of AI agents working in harmony. But peek behind the curtain:  there’s no conductor to be found, and it’s unclear whether we’re hearing a symphony or just a few instruments tuning up. Adobe promises that Agent Orchestrator will coordinate multiple AI agents to handle complex marketing workflows. But what’s GA? What’s still in beta? What’s just a slide deck?

Microsoft has fully embraced the “agentic” hype wave, rolling out an aspirational broad ecosystem of agentic AI tooling into 365 and Copilot Studio under the umbrella of what it calls the “open agentic web” and the “autonomous enterprise.” 

True to its vision of a technology factory and fabric to support the agentic future, Microsoft also rolled out Azure AI Foundry – For developing and managing enterprise-grade AI agents, and the Model Context Protocol (MCP) Servers – New infrastructure in Dynamics 365 to support agent-based automation across CRM and ERP systems 

Microsoft’s agentic AI rollout is kind of like announcing a self-driving car that still needs you to steer, brake, and occasionally push it uphill—but don’t worry, the dashboard is AI-powered.

The Problem with the Premature Parade

Let’s be clear: the idea of agentic AI is compelling. Who wouldn’t want a digital assistant that not only recommends the next best action but actually takes it after it’s proven to be trustworthy, reliable, and acting on your best interests? But the reality is, we’re not there yet. These systems are nascent, untested at scale, and often lack the guardrails needed to ensure accuracy, compliance, and brand safety.

Yet here we are, watching vendors parade these capabilities as if they’re plug-and-play. As if they’re ready for prime time. As if they won’t require months of configuration, training, and oversight to avoid embarrassing—or worse, damaging—mistakes.

Déjà Vu All Over Again

If this all feels familiar, it should. Remember the CDP smoke and mirrors and that hype that started last decade?  They were supposed to unify customer data and unlock 360-degree insights. Instead, we got a proliferation of overlapping products, vague definitions, unscalable point solutions, and a lot of shelfware.

Salesforce alone now has at least three flavors of CDP, depending on which cloud you’re in, and the one in core cloud – now called the Data Cloud – has undergone an identity parade of 5 name changes since 2020.  Was Data Cloud in the witness protection program?

Now, with Agentic CRM, we’re seeing the same pattern. Big promises. Vague demos. Glossy names. And very little clarity on what’s road-tested and what’s just a concept.

Buyer Beware: Focus on Outcomes, Not Acronyms

So, what should buyers do in the face of this agentic arms race?

  • Demand documentation. Not just marketing decks—real product manuals, configuration guides, and training materials.
  • Ask for references. Not just logos on a slide—actual customers using the product in production.
  • Look for outcomes. What business value has been delivered? What KPIs have improved? What manual work has actually been eliminated?
  • Read the reviews. Sites like G2, TrustRadius, and Gartner Peer Insights can offer a reality check that’s often missing from vendor events.

In Summary: Don’t Buy the Cape Until You See the Hero Fly

Agentic AI may one day transform CX and CRM. But today, it’s mostly a promise—wrapped in a new name, dressed in transformational and revolutionary garb, and launched with the same old fanfare. Before you bet your customer experience on it, make sure you’re not just buying a cape and hoping for a superhero.

Because in Customer Relationship Management, as in magic, the trick is often in the misdirection.

Endnote: I let GenAI create the image I included with this blog post – and coerce it as I tried to get its spelling and number of fingers correct, after several attempts, and it trying to convince me it had fixed the issues, I went with its product – perhaps to make a point.

Posted on

Vince Value Vault: Maximizing Your Shopping Experience

Welcome to the first post in a new series called “Vince’s Value Vault!” This series is dedicated to helping you, the savvy consumer, navigate the sometimes-tricky world of shopping and dealing with big companies.

Themes will include tips & tricks for getting deals as a consumer, spending money wisely, and defending the consumer’s position in a world that is increasingly dominated by AI, automation, and mega companies.  Does a consumer really have any power left?   We’ll see.  Unfortunately, the trend lately is it’s diminishing, so we’re all going to need as much help as we can get.

With 60+ years’ experience (I like to say I’m 20 years old with 40 years of experience), I started to learn at the age of 4 from my father — who was the quintessential savvy shopper and deal finder.  Early on, I received a childhood PHD in frugality, learning from a man who had to provide for eight kids on a librarian’s salary.

After my childhood apprenticeship, I embarked on a 40+ year career in marketing technology – consulting mainly with large organizations on how to build marketing systems.  As a corporate consultant by day, and a consumer by night, I’ve gained perspective on both sides of the equation.

Today, we are kicking things off with a list of a baker’s dozen (13) essential tips and tricks to ensure you get the best value for your money.  In future posts, we’ll dig in further and keep real-life tips and examples coming!

Upfront advice:

  • Do Your Research – Before making any purchase, it’s crucial to do your homework. Compare prices across various retailers, read customer reviews, and check for any online promos or discounts. Websites like PriceGrabber and Google Shopping can help you compare prices easily.
  • Study the nuances of the market you are buying from – What that means is there is no such thing as one market.  There geographic markets – local markets and global markets.  There are markets served by big companies and ones served by smaller ones.  Work to better understand the market you are shopping in, which will inform you about deal parameters in that market.  For example, occasionally, I shop at Farmer’s markets, and I’ve noticed many of them have price fixing.  All the vendors are colluding (maybe implicitly) selling at essentially the same price.  This means you won’t find deals there, since competition isn’t a factor.

Now, the Baker’s Dozen:

  1. Balance time spent finding savings with the savings potential – In other words, spending even just 15 min to save less than $1 isn’t worth it. That said, everyone has their own threshold. Mine is generally about $1 per min. So, if I can save $5 and it take me 5 min, I’ll do it.
  2. Never make large purchases on an impulse – Instead, create a spreadsheet with a simple set of columns to lay out the decision and options, and do some quick online shopping.  This will help you narrow down the exact product you are looking for, as well as to get a good price for it.
  3. You will always be trading off price for quality – So understand this and realize that often paying more upfront doesn’t mean paying more in the long run, but it just depends on how long you really want to keep using that product without having to invest in fixing or replacing it.
  4. It’s rarely worth buying an extended warranty – Companies make huge margins on these. If you simply consider the cost, often a large fraction of the product’s total cost, you are better off taking the risk and paying for another product if it fails early.  Often, there is a base warranty anyway.  If this is a very expensive product, like a car, and this is important to you, make this an important aspect of your purchase decision – that is – the length of the base manufacture’s warranty, and if it includes parts & labor
  5. Utilize Cashback and Rewards Programs – Many credit cards and online platforms offer cashback and rewards points for purchases. Make sure to sign up for these programs to earn money back on the items you are already buying. Websites such as Rakuten and Honey provide cashback on a wide range of purchases.
  6. Sign Up for Newsletters – Brands often send out special discounts and promotions to their email subscribers. By signing up for newsletters from your favorite stores, you can receive exclusive deals directly in your inbox.
  7. Use Price Tracking Tools – Tools like CamelCamelCamel and Honey’s Droplist feature can help you track the price history of products on platforms like Amazon. These tools alert you when prices drop, ensuring you get the best deal possible.
  8. Shop During Sales Events – Take advantage of major sales events like Black Friday, Cyber Monday, and annual clearance sales. These events often feature significant discounts on a wide range of products.
  9. Negotiate with Customer Service – Don’t be afraid to negotiate with customer service representatives, especially when dealing with big-ticket items or services like cable and internet. A polite request for a discount or negotiation on price can often yield surprising results.
  10. Use Coupons and Promo Codes – Always search for coupons and promo codes before completing a purchase. Websites like RetailMeNot and Coupons.com offer a plethora of discount codes for various retailers.
  11. Consider Open-Box and Refurbished Items – Open-box and refurbished items can offer significant savings while still providing high-quality products. Retailers such as Best Buy and Amazon have dedicated sections for these items, often with warranties included.
  12. Take Advantage of Free Trials and Samples – Many companies offer free trials and samples of their products. This is a great way to try before you buy, ensuring you only spend money on items you truly love and need.
  13. Read Return Policies – Always familiarize yourself with a retailer’s return policy before making a purchase. Understanding the return and exchange terms can save you from potential headaches and ensure that you can get your money back if needed.

By following these tips and tricks, you can make smarter shopping decisions and get the most value out of your purchases. Stay tuned for more insights and advice in future posts of Vince’s Value Vault

Posted on

Balancing AI and Human Touch: The Future of Personalized Customer Engagement

AI and automation have the potential to transform customer engagement by providing the ultimate efficient, personalized, and accessible service.  But can technology alone provide the human touch sometimes required?  And do consumers want this from a machine?  

When brands strive for hyper-personalization and automation, they should carefully choose which interactions to personalize and when that personalization is purely mechanical.  And they should be transparent when doing so.  Otherwise, they will feel begrudged and disappointed.  Or worse, like a Turing test gone wrong, feel hoodwinked, disturbed, and betrayed.   

Let’s take a very simple example.  How does it feel when you receive an automated happy birthday or happy holidays message?  

Reflecting on that question, personally I’m unfulfilled when I receive one.  There is nothing personalized or human about the production or delivery of the message.  It simply compares the current month and date with a birthdate, takes a name from a database, and using an email template mass produces the message.  There isn’t a caring human behind it deciding to take time out of their day to reach out and provide well wishes. 

Or take this a step further.  What If this message were more carefully crafted and personalized, to appear to be from a human, would that be better? 

In fact, this is where AI technology is going, and some companies are already taking this too far.  Those small steps too quickly become impulsive giant leaps into a world forcing customers into interactions with machines and AI that don’t always end well.  Take this example scenario where someone needing mental support is interacting with AI, but then the conversation ends abruptly:

https://www.rowbotai.com/industries/health-and-wellness (scroll to #7 “Miserable” – it’s 1 min 28 sec)

Did the technology simply fail to come up with a follow-up response? Did the database or connection go down?  Was the technology programmed to recover and reach back out to the client, or better yet, escalate this to a human?  He still desperately needed help.

Pondering these questions is not suggesting firms ignore the potential benefits of using AI and automation in the right circumstances.  Instead, it points to the importance of understanding the appropriate times and methods for utilizing advanced technologies, as well as knowing when to engage human expertise, and how to ensure a smooth and acceptable handoff and transition.   

What consumers want

As businesses design the transition and inexorably march toward using technology for increased efficiency and more digital engagements, the goal should be to give consumers what they want – relevant, rewarding, and timely brand exchanges.  But what level of personalization and humanization should a company strive to achieve in each interaction?

That depends on the situation and the customer’s intent.  For example, AI can be deployed selectively and tastefully.  In some cases, it’s fine that no human is involved (and it should be glaringly obvious that’s the case).  In others, humans should be in the loop, where AI is used to amplify human capabilities and minimize human limitations.

A “Human Touch”

Consider these aspects of an agents’ human touch – and imagine here that an agent can mean a human or machine:

  • Sounds human & conveys unconditional empathy.
  • Listens, comprehends, and suggests reasonable courses of action based on the contextual understanding of the current conversation.
  • Displays good judgement, respect, resourcefulness, and common sense in real-time problem solving, achieving status as a strategic business partner.
  • Builds a level of trust with the customer, achieving status as a trusted advisor.
  • Personalizes the experience so that the customer feels special.
  • Relates to the customer by telling stories meaningful to the conversation.
  • Recalls important details of previous conversations.

All these may sound difficult for machines to mimic, however advances in artificial intelligence and machine learning are beginning to bridge even these gaps.  When asked, about 40% of consumers believe AI has the potential to improve customer service (in that same consumer study, only 26% did not believe it could), suggesting that once it does, and if it acts in a more human-like manner, they may not care who is servicing them provided they consistently get what they want.[i]

So, what is it that humans want?

  • Short or no wait times – Time is precious, and customers appreciate immediate responses.
  • Accessible service – Availability across all platforms and at any time is crucial.
  • Fast service – Once engaged, customers expect a prompt resolution.
  • Coordinated service – Seamless transition between service channels without repetition is key.
  • Accurate & fast answers – Quick and correct responses build trust.
  • Tailored and relevant recommendations – Personalized and non-intrusive product, service, and support recommendations that make sense are welcomed.
  • Warm and cordial experiences – Friendly and memorable service fosters loyalty.

Pitfalls to avoid when applying AI in customer engagement:

In planning for success, it’s essential to consider what to avoid, minimizing unnecessary mistakes. To improve machine or human performance, learn from the mistakes of others. Here are some dangers to sidestep:

  • Tendency to over automate before you carefully assess the impact total automation may have on customer experience. It may be improving the bottom line (in the short run), however is it ultimately improving customer satisfaction, or making it worse?
  • Placing a premium on playing parlor tricks with technology or customer information, while not focusing on whether the outcome is ideal.  Scrutinize the value of the use case. Improper use can backfire, such as wishing a customer happy birthday when they never gave you permission to gather and use their birthdate.
  • Losing sight of the root cause and fixing it. Why did the customer ask for support in the first place?
  • Designing programs based on the law of averages versus factoring in individual customer preferences & valuation.  Remember, some customers may require human interaction, and it might be economically justified to provide just that.
  • Falling victim to automation bias – Becoming sloppy, complacent, insensitive, and dulled, because machines take care of so many customer service tasks – and when humans are called on to provide service, they can’t – due to being rusty, incompetent, or rendered totally incapable.

5 ways to employ AI in customer engagement:

  • Automate the no brainers. For example, use automation & intelligence to lookup routine customer information, answer frequently asked questions (FAQs), get order status, or even process a payment.
  • Use automation & intelligence to classify and route emails, calls, and other requests. They’ll get faster to the right people that can ultimately close out the case.
  • Augment staff with intelligence, such as using a GPT knowledge base, with filtering and learning capabilities, to rank likely answers to FAQs, and provide those to staff so they can quickly answer questions, while still providing a human touch.
  • Take interest in what matters to the customer and knowing things about them pertinent to the relationship. Use AI to help store and recall critical material at the right moment, and let humans decide how and when to weave that into conversation for a natural flow.
  • Ensure warm handoffs between self-service technology and the humans who might have to complete the servicing. For example, when a customer engages in self-service, but then escalates, guarantee a comprehensive and seamless transition of the self-service transaction to human agents.

Advances in AI technologies that apply a human touch:

  • Experiment with chat bots, focusing on which interactions can be fully handled by machines, and which need to be either immediately routed to a human, or escalated to a human once its apparent the chat bot has reached its limitations.  This coincides with striving for an overall system that is friendly, helpful, and convenient to do business with.
  • Guard against AI’s detrimental potential, such as being tone deaf and discriminatory.
  • Test using prescriptive intelligence techniques that incrementally improve relationships.
  • Use real-time event processing technologies, voice AI, and journey analytics to gather contextual behavior, detecting and reacting to in-the-moment customer struggle and intent.  For example, paths that cause customers to repeatedly drop out or abandon an objective, such as failing to finish an application.  Or customers who raise their voice or rage-click a button, indicating high levels of frustration.
  • Select one next-best-experience engine and connect it to all channels.  Its role: act as the corporate always-on brain – a 24 x 7 x 365 customer memory bank, insight generator, and engagement hub that knows when to automate and when to escalate to humans.
  • Employ usability testing and customer surveys to monitor customer journeys and experiences.  Fine tune journeys so they combine the right mix of automation, convenience, relevant recommendations, and human touch to deliver optimal results.

Conclusion

We’re all developing relationships with machines and already extremely dependent on them. We talk to our devices, use them as assistants and navigators, and laugh at their jokes.  Ten years ago, the movie “Her” seemed like far-fetched science fiction, yet today there are apps like Replika where people form emotional relationships with AI.

AI is here to stay. Make no mistake – it’s going to automate more manual tasks and change the nature of many jobs and our consumer experiences, just as the industrial revolution did over 100 years ago.   An economic outlook published in 2017 by PWC predicted that by 2030 automation would replace as many as 40% of current jobs, such as transportation, manufacturing, and trade. [ii]  And that was 5 years before the generative AI revolution. 

Even so, where social skills are paramount, such as in customer service and social work, those same forecasters expected the impact to be radically less.  It’s hard to say precisely how this plays out.  No doubt, a significant share of jobs are at risk of automation, and there will be completely human-less customer journeys.  And although AI will create new jobs, and in some cases better experiences, the transition will be disruptive for many people, who will need to adapt and re-skill.

By avoiding common pitfalls and strategically employing AI, businesses can create a customer engagement model that is both technologically advanced and warmly human. The goal is not to replace human interaction but to enhance it with AI’s capabilities, ensuring that the customer’s journey is as convenient as it is delightful.  

Technology continues to change lives and create opportunities for businesses. Those that learn to use it effectively at scale, with a proper balance of automation, AI, and human touch in customer interactions, will be more relevant to customers, and win more loyal long-term relationships with them.

[i] What Consumers Really Think About AI: A Global Study, https://www.pega.com/ai-survey, 2022

[ii] PWC, Economic Outlook, https://www.pwc.co.uk/economic-services/ukeo/pwcukeo-slides-final-march-2017-v2.pdf, 2017

Posted on

Trusting Personalization without Sacrificing Privacy

In today’s digital age, consumers are increasingly seeking more personalized products and services, and a customized experience.  And although personalization is nothing new, it has evolved radically in the last 30 years with large businesses able to use AI and technology to cater to these demands.  Just look at the latest frenzy, GPT, to get a sense for how technology is accelerating its impact on marketing, customer service, and personalization.

Balancing Personalization & Privacy

Already a wide variety of personalization techniques are employed by businesses to tailor to individual preferences and improve the customer experience.  It’s been 30 years in the making, starting with early adopters using simple techniques such as addressing a direct mail letter with first name rather than “Current Resident,” to today’s savvy enterprises using massive databases of consumer behavior and advanced analytics to provide hyper-conditional content, individualized promotions, and concierge-like digital services.

But a counter dynamic is also at play.  As personalized products and services have become more prevalent, consumers have also awakened to how their data is collected and even misused.  Because of this some are less likely to share information, push for more legislation, more frequently opt out, and even ask for their data to be deleted.  Clear battle lines have been drawn between hyper personalization and privacy.

This presents a dilemma for businesses, as they attempt to balance providing personalization in a responsible and controlled manner.  Consumers are fickle.  They want great experiences, but they also expect that any data they turn over is secure and is used in compliance with their wishes.  In this context, it’s crucial for companies to strike the right balance between personalization and privacy protection. 

The Value of Personalization

When asked, consumers repeatedly respond that they want more personalization (in many cases greater than 75%), especially younger people. [i]   When they see customized ads fitting their preferences and behaviors, they’re more likely to engage and then purchase. This benefits both the consumer, who gets offered interesting products or services, and the company, which increases its revenue and profits.  Moreover, personalized recommendations can create a sense of loyalty and trust between the consumer and the company, leading to repeat business, long-term customer retention, and positive word-of-mouth marketing.

Personalization has become a principal factor in consumer decision-making. A study by Epsilon found that 80% of consumers are more likely to do business with a company if it offers a personalized experience. [ii]  Personalization has many benefits for consumers, such as saving time and increasing convenience. For instance, personalized recommendations on e-commerce websites can help consumers find and then ask for products that they may not have otherwise considered. Similarly, personalized apps can help users achieve their goals by providing services like tailored financial plans and wellness insights & activities.

Some consumers may view the collection of their personal data as a fair trade-off to get more personalization.  They accept that the exchange of their data is worth getting good recommendations, more perks, and a better overall experience.  And these consumers assume and inherently trust that businesses will use their data responsibly and will take the necessary steps to protect their privacy. 

But not everyone thinks that way about turning over their data and trusting businesses.

The Need for Privacy

There is another camp of consumers that prioritize privacy and may view personalization as a threat to their rights. They hold that their personal data is being exploited without their consent and that businesses are profiting from their information.  And they are very vocal about this, and are influencing their friends, followers, and even law makers.

These consumers are skeptical of businesses’ ability to protect their data and worry that their information could be used against them in the future.  As a result, they have pushed for more protections and use various mechanisms (ad blockers, opting out, surfing incognito) to avoid sharing personal information. 

They are very hesitant to share what they consider extremely personal information, such as their location, browsing history, or purchase behavior. And they are not a small faction. In fact, a Pew Research Center study found that 79% of Americans are concerned about the way their personal data is being used by companies beyond what they intended.[iii]  Years ago, the Cambridge Analytica scandal highlighted these fears, when the data of millions of Facebook users was harvested without their consent and weaponized for political purposes.

Consumers believe they have a fundamental right to privacy and are increasingly demanding more control over their personal data.  In recent years, there have been several high-profile data breaches and scandals involving the misuse of personal data. These incidents have increased consumer awareness about the importance of privacy, and they are becoming more vocal about their concerns. Consumers want to know how their data is being used, who has access to it, and how it is being protected.

Consumers are also becoming more aware of their digital footprint and the potential consequences of sharing personal information online. They’re concerned about identity theft, fraud, and other forms of cybercrime. In addition, they worry that their personal information could be used against them, such as by insurance companies or potential employers. As a result, consumers are becoming even more cautious about sharing personal information online, and as such only do it with businesses they trust.

The Role of Trust & Regulation

Trust is a crucial factor in the relationship between consumers and businesses. Consumers are more likely to share personal information with businesses that have not breached that trust and that prioritize their privacy. Even so, trust is fragile, and businesses must work hard to never violate it and always maintain it.

Businesses can build trust with consumers by being transparent about their data collection practices and providing clear explanations of how they use consumer data. They should obtain explicit consent from consumers before collecting any personal information and should provide consumers with the option to opt-out of data collection. And they must ensure that data is protected from data breaches and cyberattacks.

Privacy concerns can also impact consumer behavior in a more general sense. A review of the literature by Taylor and Francis Online found that online privacy concerns can lead to reduced trust in online transactions and lower engagement with online platforms.[iv]  This can have significant consequences for companies that rely on online channels for marketing, sales, and deepening relationships with existing customers.  What’s more, privacy concerns have a ripple effect across businesses and industries, as consumers become more skeptical with each new incident.

Best Practices

Here is a baker’s dozen of best practices to effectively balance personalization and privacy.

For Personalization:

  • Rollout individualized personalization – Individualized personalization uses preferences and behaviors of individuals (not segments they belong to) to custom tailor products, servicing, and messaging.  It affords numerous benefits, including increased convenience, engagement, and ultimately overall satisfaction.  And done right, it will build long-term trust.  When personalizing for pre-login purposes (such as for unknown browsers or mobile devices) pay careful attention to whether adtech data collection vendors are using tricks, like cname cloaking, to mask a 3rd party domain still receiving data when the end consumer may not approve of this.  Using a trick like this might erode trust and ultimately backfire.
  • Build cross-functional personalization teams – Personalization requires input and collaboration from multiple departments within an organization, including marketing, IT, data & analytics teams, and customer support. Building cross-functional teams can help companies break down silos and ensure that all stakeholders are aligned on personalization goals and strategies.
  • Adopt agile methodologies for feedback and testing – Agile methodologies enable companies to iterate quickly and respond to changing customer needs and preferences. Adopting agile methodologies can help companies test and refine personalization strategies and ensure that they are delivering the right content and experiences to the right customers at the right time.  Pick an agile personalization platform.  It should be capable of always-on variation testing (runs champion / challenger experiments automatically) and able to deploy necessary adjustments to programs in a day or less not weeks.

For Privacy:

  • Appoint a Chief Privacy Officer – Appointing a Chief Privacy Officer (CPO) is a critical organizational change that can help companies balance personalization and privacy. The CPO is responsible for ensuring that the company’s data privacy policies and practices align with industry standards and regulatory requirements, while also driving data-driven innovation and personalization.
  • Establish a clear data governance framework – Companies must have a clear governance framework for data management that outlines data privacy policies, data protection practices, and compliance requirements. This framework should be regularly reviewed and updated to ensure alignment with changing regulations and best practices.
  • Use privacy nudges – Nudges can be an effective way to help consumers make more informed choices about their data privacy, while still allowing for customization.  Privacy nudges are interventions designed to influence behavior without restricting freedom of choice.  For example, a company could use a privacy nudge to encourage consumers to read the privacy policy before accepting it. A study by Balebako et al. (2015) found that privacy nudges can be effective in improving privacy outcomes for consumers on social media platforms. [v]
  • Conduct regular privacy impact assessments – Privacy impact assessments (PIAs) can help companies identify privacy risks and implement appropriate controls to mitigate these risks. Conducting regular PIAs can help companies stay ahead of changing privacy regulations and address potential privacy concerns before they become major issues.
  • Foster a culture of privacy first – A culture of privacy first starts at the top of the organization, with senior leadership setting an example and emphasizing the importance of privacy in all aspects of the business. Companies can also provide privacy training and awareness programs to all employees, to ensure that everyone understands the importance of privacy and how to protect it.

For making the right technology choices:

  • Invest in a consent management platform – These platforms allow companies to manage user consent and data collection preferences, enabling users to choose what data is collected and how it is used. Consent management platforms can also help companies comply with regulations such as the GDPR and CCPA, which require informed and explicit user consent for data collection and processing. Invest in a customer data platforms (CDPs) only if first-party data and device identity management is scattered or missing.   CDPs are a centralized system combining customer data from multiple sources (such as transaction & behavioral data), device identity, and consent. Companies must be transparent about their data collection and usage practices and provide clear and concise information about how data is being used. This can be accomplished through user-friendly privacy policies, clear consent mechanisms, and open communication channels.  
  • Leverage the right artificial intelligence (AI) – AI can help companies analyze vast amounts of customer data and identify patterns and trends that can inform personalization strategies. Companies can use the right AI for the right job do better. 
    • For example, firms using Bayesian models to deliver personalized next-best-action recommendations tailored to each individual customer (that adapt in real-time as preferences shift) report up to 6x lift in response rates. [vi] 
    • They also enjoy an added advantage that these models are transparent and can be pre-checked for bias, responsibly deployed, and explained.  Compare that to this statement in a recent research paper on using GPT: “We do not intend for the model to be used for harmful purposes but realize the risks and hope that further work is aimed at combating abuse of generative models.” [vii]
    • Use other responsible techniques, such as federated learning and homomorphic encryption, which enable machine learning models to be trained on user data without accessing or exposing individual user data.
  • Use anonymization and pseudonymization techniques – These are methods of data de-identification that can be used to ensure that sensitive personal information is protected while still allowing for effective analysis to improve personalization. Anonymization removes all identifying information from a dataset, while pseudonymization replaces identifying information with a pseudonym, allowing for the data to be re-identified if necessary.  If data is shared externally with other parties, use clean rooms.
  • Consider differential privacy algorithms – Differential privacy is a technique that adds noise to a dataset to protect individual privacy, while still allowing for effective data analysis and personalization. Differential privacy algorithms can be used to provide personalized recommendations, while still ensuring that individual user data remains private.
  • Employ secure data storage and transfer protocols – Companies must ensure that user data is stored and transferred securely, to prevent unauthorized access or data breaches. Technologies such as encryption, secure sockets layer (SSL), and transport layer security (TLS) should always be used secure data storage and transfer.

Conclusion

Being customer-centric is about adopting a mindset that improving customer experience is paramount to business success. Personalization is an essential strategy for businesses to improve the customer experience. However, as personalization becomes more prevalent, consumers are also becoming more concerned about their privacy.

As the debate between personalized products and services and data privacy rages on, companies must navigate a delicate balance between meeting consumer demands for customization and respecting their privacy. Failure to do both right could have profound consequences, including loss of trust and customer loyalty, as well as legal and regulatory penalties. However, those companies that can offer personalized products and services while also prioritizing transparency and data privacy will be well-positioned to succeed in the digital age.

The future belongs to those companies that can harness the power of data-driven personalization in a responsible and transparent way, while respecting the privacy and autonomy of their customers. In short, the choice is clear: companies can either embrace data privacy as a core value and use it to build lasting relationships with their customers, or risk being left behind in an increasingly competitive market.

[i] Capco. (2021). Insights for Investments to Modernize Digital Banking. https://www.capco.com/Intelligence/Capco-Intelligence/Insights%20for%20Investments%20to%20Modernize%20Digital%20Banking

[ii] Epsilon. (2018). The power of me: The impact of personalization on marketing performance. https://www.epsilon.com/-/media/files/epsilon/whitepapers/emea/the-power-of-me.pdf

[iii] Pew Research Center. (2019). Americans and privacy: Concerned, confused, and feeling lack of control over their personal information. https://www.pewresearch.org/internet/2019/11/15/americans-and-privacy-concerned-confused-and-feeling-lack-of-control-over-their-personal-information/

[iv] International Journal of Human-Computer Interaction. (2020).  Online Privacy Breaches, Offline Consequences: Construction and Validation of the Concerns with the Protection of Informational Privacy Scale. https://www.tandfonline.com/doi/full/10.1080/10447318.2020.1794626

[v] Balebako, R., Danish, R. K., Hong, J. I., & Cranor, L. F. (2015). Privacy nudges for social media: An exploratory Facebook study. Proceedings of the 33rd Annual ACM Conference on Human Factors in Computing Systems, https://dl.acm.org/doi/abs/10.1145/2487788.2488038

[vi] Pegasystems. (2022).  Coutts: Delivering world-class banking experiences at scale.  https://www.pega.com/customers/coutts-customer-decision-hub

[vii] Semantic Scholar. (2023). Structure and Content-Guided Video Synthesis with Diffusion Models. https://www.semanticscholar.org/paper/Structure-and-Content-Guided-Video-Synthesis-with-Esser-Chiu/07be0ec1f45e21a1032616535d0290ee6bfe0f6b

Posted on

The Hyper-Personalization Paradox: being relevant without crossing the CREEPY LINE

Brands are using AI to drive hyper-personalization, but can it also help them avoid being hyper creepy?

hyper-personalization

Source: https://www.adclarity.com/2015/04/digital-marketing-2015-hyper-personalization-display-ads/

Apparently, I have 8 seconds to grab your attention, so here goes.  What if I personalized every aspect of this blog for you?  That is, I knew so much about you – your reading behavior, the writing style you prefer, subjects you love – took all of it into account, and assembled these words and pictures just for you?  Would you find that creepy or cool?

At our conference in Las Vegas recently, I was a guest on Sam Charrington’s, podcast series “This Week In Machine Learning and AI.”  In that episode, we discussed a similar hyper-personalization scenario, where an automotive company used intimate knowledge about a consumer and her connected car to custom-tailor each marketing and service treatment[i].  And half-way through (at 23:07), Sam observed that although “consumers appreciate personalized experiences,” it can go too far and “sometimes come across as creepy.”

And suddenly, we both realized something.  Customer experience experts haven’t used AI to govern this.  In other words, CX pros personalize without recognizing if their personalization levels are approaching creepiness.

Which led to this question: can creepiness be quantified?  And if so, with that knowledge, could a company effectively use it?  With the right tooling, could they safely test and simulate how far personalization should go, carefully delivering each customer a tailored experience with the right level of relevance and value, without crossing into their creepy space?  Simply put, hyper-personalizing without being hyper-personal — the personalization paradox.

You’re marketing is creeping me out

Creepy land is that forbidden zone where consumers call out businesses for using personal data and revealing insights that are a bit too private.  And though consumers increasingly want personalized experiences (according to a recent Epsilon study[ii], 90 percent of consumers find it appealing), ironically, they will happily make examples of brands that invade their personal space.

No brand wants a creepy reputation as it implies:

  1. Stalking, snooping, or spying; collecting personal data and invading privacy
  2. Revealing something private, no matter how valuable the insight
  3. Not having customers’ best interests in mind
  4. Ill-intent, even when there isn’t intention to do harm

With big data galore, a culture of a data sharing, and pressure to mass personalize to remain competitive, you need ways to safely and systematically explore the creepy line’s location without ever crossing it.  Understanding what customers expect and why they love a product (or don’t) is crucial to great personalization.  Avoiding a creepy moniker means effectively steering clear of areas that are, frankly, none of your business.  And if the customer says it’s none of your business, it’s none of your business.

Today, the digital world abounds with copious quantities of demographic, psychographic, and behavioral data.  There’s a sea of it, because for decades companies have wired up clients and monitored them like lab rats.  And with more IoT tech and data coming every day, firms increasingly misuse it, giving customers more reasons to demand privacy.  The problem is the definition of what’s private and sensitive can be different for each person.  Hence the dilemma: under personalize and risk being labeled clueless, not cool, and worse miss out on revenue; over personalize and risk breaking trust and doing irreparable damage to your reputation.

Sorry we’re creepy. We apologize for any inconvenience 

Customer engagement professionals need new and scalable ways to survey buyers, collect preferences and permissions, sense their intent and moments of need, and personalize appropriately.  So, they need ways to test where that creepy boundary is.  That line is fluid and ever shifting and finding the right level of personalized insights and recommendations without crossing into risky territory is never without some uncertainty.

Where that line lurks changes with time because initially customers may be leery of something, then later adapt to it.  It also changes because privacy legislation changes, individual consumers have distinct levels of sensitivity, and varying levels of awareness. It can even differ by geography.  For instance, a 2016 study of 2000 consumers in Europe found that 75 percent were uncomfortable with facial recognition software used to target them with personalized offers (consumers in the US were much less sensitive)[iii]

Data-driven marketers have evolved their practices (Figure 1) using data to acquire more customer knowledge which in turn powers more personalization.  Over time, more marketers have evolved their practices, from the general advertising Mad Men approaches of the 1960’s to the super-personalized, AI-Powered approaches possible today.  It also highlights how that pushes them closer to the creepy space.

hyper-personalization evolution

Figure 1: Evolution of Data-Driven Marketing

Here’s the bottom line: if a given customers perception is it’s creepy, it’s creepy.  And depending on who slaps that label on, and whether their rights that have been violated, firms may face legal battles, fines, and reputation damage leading to significant commercial impact.  For instance, potential fines for GDPR privacy law violators can reach 4 percent of a firm’s revenue (up to a maximum of €20 million).

And none of that is music to a businessperson’s ear.

Creeping toward creepy

In 2014, Pinterest managed to spam a major segment of customers when they sent emails to unengaged women congratulating them on their upcoming weddings.  And Shutterfly made an even bigger spam faux pas that same year, congratulating women on the birth of babies they didn’t have.

In Figure 1, these events fall into the SPAM circle because marketers placed people into the wrong macro segments, and the resulting emails were both irrelevant and hilariously erroneous.   Clumsy customer experiences indeed, but not creepy-smart marketing.

Here are some other examples of Mad Men SPAM marketing:

  • You market wedding offers after a wedding – low sensitivity
  • You market wedding offers after a cancelled wedding – high sensitivity

On the other hand, the risk of being labeled a creepy marketer increases when knowledge of customers goes up, insights increase, yet marketers fail to understand an individual’s sensitivity to certain marketing actions.

For each marketing treatment, you need to determine if it will be creepy to everyone or only some:

creepy meter

Figure 2a:  Creepy Meter detecting creepy treatments

If it’s clearly creepy to everyone, during the pre-market approval process you should reject it.  But, if its potentially cool to some, and creepy to others, then provided you can discriminate at runtime using eligibility rules, you can approve its use for those who will find it cool.

To do this, get a readout on consumers’ sensitivity to hyper-personalization.  Build a model that learns this, and use this score to select, by individual, the levels of personalization they’re eligible to receive.

creepy index

Figure 2b:  The Creepy Sensitivity Index readout on each consumer

Here are a few examples of events, corresponding covert marketing approaches, and creepy readings:

 Event Covert Marketing (but not illegal) Creepy Meter Approve?
Hospital admittance / serious health issues detected Mortuary makes discount offers Extremely creepy Reject
Conversation recorded (without clear permission to use for marketing) Ads for products related to keywords in the conversation (e.g., pet toy video recently, which illustrates the point yet is likely a hoax) Very creepy Reject
Facial recognition or location detection Upon a patron entering a branch or store, their profile & preferences are relayed to a salesperson Borderline creepy Conditional
Consumer traveling; recent activity and calendar scanned Push notifications offering travel recommendations based on triangulating travel intent and destination Borderline creepy Conditional
Consumer browsing a web page with product offers Website background, images, language, offers, and other page fragments hyper-personalized Borderline creepy Conditional

Table 1: Examples of potentially creepy marketing

Leading-edge 1:1 marketers are constantly listening for keywords, tracking interaction device, time & location, codifying behavior, sensing mood, recording preferences, and using that knowledge to hyper-personalize with content variations in the millions.  The risk, however, is meandering into that forbidden creepy zone (even if it’s legal), so discerning this by customer by treatment is vital.

Suggestions

As you move into deeper levels of hyper-personalization, do so deliberately and methodically, fully grasping the implications before rolling out.  Consider taking this approach:

  • Collect only data that matters to your ability to personalize specific experiences – that your customer will value. For example, if you sell insurance, you don’t need to understand pet preferences unless you’re selling pet insurance.
  • Start with simple / minimal risk personalization strategies. These should easily pass the creepy test.  For instance, if you can tune you web experience to shopper color preferences, do it.  No one will find that creepy.
  • Gradually apply regional and demographic personalization strategies.
  • Use AI to crawl your products and content to extract taxonomies, attributes, cross-classifications, and descriptions. This will help better match customer intent and preferences to products that will match needs.
  • Use AI to match the right products to clients (making relevant recommendations) and doing so in a personalized way that enhances their experience
  • Use sampling to test hyper-personalization treatments, selecting a wide variety of customers.  Essentially, you get a stratified sample of creepiness raters.
  • In general, avoid even borderline covert marketing unless you have a firm handle on any backlash that might result if customers discover it. In a recent survey, most consumers (81%) think firms are obligated to disclose they’re using AI – and how they’re using it.[iv]
  • Be sensitive to consumers’ preferences for public recognition.  Some might love it if you great them by first name and show appreciation for their loyalty in public.  A few, however, may be mortified.

Hyper-personalization requires great data, great technology, and great sensitivity.  With GDPR now in effect, most businesses are proactively disclosing their data collection practices and privacy policies.  As consumers, we’re consenting to and accepting new privacy policies more than ever before, and in some cases, we’re even reading and understanding them.  Less clear, however, is exactly how that data is used, combined with other data, and when it might show up as an insight, recommendation, or hyper-personalization – and again, which of us might be freaked out by this personalization.

AI is driving personalization to new levels.  There’s no stopping that.  It automatically figures out what works and what doesn’t.  Techniques, such as Bayesian algorithms, quickly learn which offers work, when, and in which channels.  Others, like collaborative filtering, find which products pair best, that in turn drives cross-sell and bundling strategies.  Design of experiments and monitoring devices measure the impact and enable fine tuning.

What’s missing, however, are tools to sense consumers’ sensitivity to personalization, so overt practices are optimized with the right people, and so covert methods are prevented from ever reaching production, or if they are approved for use, are carefully applied.

The study shown in Figure 3 provided some proof that overt personalization pays off.   Yet the very definition of overt blurs as AI improves, content becomes hyper-conditional, and levels of personalization get more complex.  Thus, you’ll need more sophisticated ways to gauge levels of personalization relative to creepiness, and the sensitivity levels of different people.

personalization

Figure 3: Overt vs covert personalization performance[v]

Conclusion

Great marketers push beyond perceived barriers by understanding customers, knowing products, and then elegantly combining creativity and technology to provide valuable recommendations and experiences to customers.  Ironically, when done right in the eyes of the receiving consumer, they don’t appear to be selling anything; instead simply providing a service.

With website personalization, one-to-one content, natural language generation, image recognition, and countless other AI tools, businesses inexorably march toward hyper-personalization.  Make sure you manage it, so you’re always cool and never creepy.

Endnotes:

[i] https://www1.pega.com/insights/resources/pegaworld-2018-pegas-ai-innovation-lab-sneak-peek-and-your-vote-counts-video, June 2018

[ii] http://pressroom.epsilon.com/new-epsilon-research-indicates-80-of-consumers-are-more-likely-to-make-a-purchase-when-brands-offer-personalized-experiences/, January 2018

[iii] https://www.forbes.com/sites/fionabriggs/2016/07/04/fingerprint-scanning-is-cool-but-facial-recognition-creepy-new-richrelevance-survey-shows/2/#493b953f3d68, July 2016

[iv] https://www.richrelevance.com/blog/2018/06/20/creepy-cool-2018-richrelevance-study-finds-80-consumers-demand-artificial-intelligence-ai-transparency/, June 2018

[v] https://www.sciencedirect.com/science/article/pii/S0022435914000669#abs0005, March 2015

 

Posted on

Consumers kill for digital convenience: Can AI help your business?

We’ve all seen countless images of the proverbial empowered consumer.  That mythical creature seeking convenience and instant gratification.  It’s a conjured-up image of a time-strapped digital native that juggles five devices and 15 tasks, interacting simultaneously on a host of channels, using their super-human consumerism to wield terrifying powers capable of paralyzing unworthy brands.

AI in business

Hyped-up as they are, these visuals still serve a healthy purpose.  They remind us just how far digital bars have been raised, and that should cause pause and beg the question, “as businesses, are we measuring up?”

Collectively, the answer is we’re not.  In fact, consumer satisfaction studies repeatedly confirm it.  Simply search on, “consumer study poor digital experience” and voila – hundreds of examples.  One study conducted by Software Advice found over 90 percent of consumers had one or more deal-breaker digital experiences when seeking customer support on mobile[i].  So, in an age with so much technology at our fingertips, why are we falling short?  What can we do to fix this?

Too often, we fall short because we focus on the wrong problems in the wrong order.  To correct this, it’s important to first consider a modern consumer’s mindset and what they’re demanding.  With greater resolve, they’re chasing after nirvana, in a quest for brands that deliver products, services and experiences that are:

  • Valuable / relevant
  • Consistent / high-quality
  • Enjoyable / attractive / personalized
  • Familiar / trusted
  • Secure / lower risk
  • Compatible with values / social beliefs
  • Convenient / simple / timely

Enterprises, however, can’t perfect all seven of these deadly-important areas simultaneously.  So, the trick is finding what matters most, and then using AI and automation technologies to help.

AI in business won’t magically transform a company with fundamental structural flaws, such as poorly designed products, no unique selling proposition, or cost containment issues. These take great human leadership, creativity, and collaboration to fix.  And it won’t manage the job of building and maintaining corporate culture.  But in other cases, AI applied pragmatically to streamline processes and eradicate friction can make an enormous difference.

What’s proven to be a winning recipe in business is paying attention to customer-centric details.  Brands hyper-focused on customer experience build a lasting reputation and increase in value.  Look at Apple, Uber, Airbnb, Amazon, and even Booking.com.  All built on the backs of nailing digital experience, often with a mobile-first mentality.  Yet, with seven major areas and hundreds of experience details to consider, where should you start?

Is convenience king?

Out of the above seven criteria, convenience may be the most important in terms of driving long-term value, and the one CX professionals can influence the most.  Perfecting convenience can separate winners from losers; sellers from re-stockers. Consider this quote from a CEB study[ii]:

“Brands that help consumers simplify the purchase journey have customers who are 86 percent more likely to purchase their products and 115 percent more likely to recommend their brand to others.”

AI in business

And convenience contributes to and builds up other factors, such as being viewed as valuable, familiar, and trusted.  It may be one of the chief drivers of loyalty.  It can even trump something like price.  For example, wireless carriers have learned consumers prefer unlimited communication plans because they’re convenient and simple, even though they may cost more[iii].  Consumers make impulsive and emotional purchase decisions when enough of the factors align, and tend to justify things afterwards.  Since consumers’ assessment of convenience is qualitative, figuring out how to elicit positive emotional responses regarding convenience is crucial.

In a consumer’s mind, the label of convenience translates into a business being viewed as:

  • Useful and suitable
  • Easy to buy from, use, and transact with
  • Requiring less overall effort
  • Simple to understand / responsive to issues
  • A time saver

Each is a judgment call by an individual, but with critical mass and time, these opinions converge to a collective market consensus (the wisdom of the crowd).  They manifest themselves in the form of review scores, ratings, and tribally-shared social advice.  It’s this reputation that drives commercial allegiance.

Largely, consumers make emotional decisions when they choose one product over another.  Sometimes they want combinations that are seemingly impossible to get:

  • A readymade desert that tastes great and is nutritional
  • A car that is inexpensive, fast, great looking, economical, and durable
  • A delicious pizza that comes in a few minutes, is made by an environmentally-conscience brand, and oh…costs less than $10

It’s no wonder brands struggle to satisfy whimsical consumer desires, but fickleness aside, they cry out for brands to simply simplify things.  Ironically, they work longer and harder to live in a world that supplies them with exploding choices for everything but precocious little time to weigh options, which in turn drives them to crave simplicity in decision making. They demand trusted information that is easily accessible.  They want user-friendly ways to weigh options, and help navigating processes.  In a 2016 survey on travel shopping preferences, consumers picked ease of use as the top reason they booked using an online travel site.[iv]

AI knows there’s no second chance to make a first impression

Consumers want convenience, but which actions will achieve maximal impact?   Before answering this, keep in mind a marketing 101 maxim: perception is nine tenths reality.  And perception is often built-up on first impressions.  Further, when an initial impression goes wrong, it takes multiple positive interactions to repair it.  As such, consider using AI as tooling in helping elevate levels of perceived (and real) overall convenience in critical first-impression customer journeys such as:

  • Getting a quote
  • Completing an application
  • Navigating a sign-up or onboarding process
  • Completing an initial purchase
  • Setting up online payments

And during service scenarios such as:

  • Order status checking
  • Returns
  • Claims
  • Lost card replacement process
  • Scheduling an appointment
  • Finding a doctor

How does AI support these?  If we agree that AI is a mixture of automation and intelligence technologies, AI can help streamline the process for consumers getting answers such as the status of an order, return, or claim.  Further, consumers can even ask these systems to schedule a store or branch appointment, find the most convenient time and location, and then add the appointment to their calendar.

AI-powered chat bots (and other self-service portals) can provide 24 x 7 first-line support for answers to questions like:

  • How to transfer funds
  • Make an online payment,
  • Get account and policy status

In many cases, without any human intervention, bots can answer questions, close out an inquiry, and even assist with completing a transaction.  In situations requiring human agents, AI-based systems can orchestrate seamless hand-offs of data and case details, allowing humans to pick up precisely where machines left off.

Make no mistake, AI skills are already going far beyond performing simple tasks.  Today, AI engines can give nuanced advice, surface unique insights, and provide proactive recommendations.  The most sophisticated systems even factor in customer context, such as location, weather, mood, and motivation before arbitrating on the next-best-action.

In banking, for instance, AI can help track savings and spending habits, and send threshold alerts. To illustrate, suppose a consumer has a recurring transfer from checking to savings each month.  AI can monitor account balances and send an alert when upcoming bill payments are forecasted to drain a checking account beyond non-fee thresholds.

In healthcare, there’s Dr. AI from HealthTap, who can engage in conversation aimed at providing triage and care advice, using a locally-stored health profile, a network of over 100,000 doctors, and Bayesian learning AI to serve up the next-best-advice.

What’s the right set of technologies for your stack?

Well, there’s good news and bad news.  First the bad news – there is no one right answer, and with thousands of vendors (6,829 in this marketing landscape), open-source packages, and resulting combinations of solution stacks possible, there’s no evidence anyone has found the absolute best combination, or ever will.

Now the good news – you have a ton of alternatives, with many combinations likely to work, but finding a stable and winning blend is tricky.  Some tools, on the surface, look easy to use but aren’t.  Others won’t live up (functionally) to their marketing hype.  The best advice is to form a solid basis with at most one or two platforms covering essential infrastructure (that you can’t afford to switch in-out), and make sure these platforms allow for plug and play with adjacent pieces likely to have shorter useful lives.

For example, find vendors with durable connectors for wrangling data into an actionable customer profile, a real-time hub that acts as a central brain to arbitrate customer decisions, and integrated customer analytics.  These components are foundational, and must be centralized so they operate in a channel agnostic fashion.  New channels may spring up, and others diminish in importance, but a decision engine which feeds on key behavior data, arbitrates decisions, and renders appropriate next-best-actions is a necessary constant.

Final thoughts

There’s a real irony forming with AI in business.  We’re building and teaching computers to be more human, while as humans we’re being led and conditioned by our busy lives and workplaces to be more machine-like.  The problem is computers are no humans, and humans are poor computers.

Step back and consider what’s best for the consumer.  Providing great first impressions, as well as seamless and gratifying ongoing experiences, requires well-functioning and well-behaving humans and machines working in concert.  Consumers want products and services they’re proud to recommend because they make life easier and more enjoyable. When things go wrong, they expect flexible help and fast solutions.  When self-service isn’t working, they demand cases smoothly transition to well-informed, caring, and compassionate humans.  Brands must skillfully, judiciously, and mindfully weave together computer systems with humans as they design for convenience in all the complexities of customer journeys.

Delivering convenience must be a paramount goal, so reflect on the unique characteristics of the individuals you serve and the nuances of their voyages.  Dry run how each will navigate your services:  some will be older and less familiar with technology; some will be capable of juggling five devices on five channels; sometimes technology will fail and require fallback processes.

Ultimately, your convenience reputation will be defined by a diverse set of consumers steering through a wide variety of conditions and processes.  Use AI and humans to start off on the right foot, deliver consistently under normal operating conditions, and to proficiently handle the inevitable miscues.

[i] https://www.softwareadvice.com/resources/improve-cx-with-mobile-support/

[ii] https://news.cebglobal.com/press-releases?item=128138

[iii] https://www.theverge.com/2017/2/17/14647870/us-carrier-unlimited-plans-competition-tmobile-verizon-att-sprint

[iv] http://www.traveltripper.com/blog/why-do-travelers-prefer-booking-with-otas/

 

Posted on

Incremental contextual marketing for the distracted consumer

Consumers today are distracted and so I’ve coined a new term describing how to reach them: incremental contextual marketing.   The idea came to me by when considering the mental diagnosis of ADHD, and then drawing parallels to phenomena observed in modern consumer behavior.  In this blog, I explore this and then unveil some thoughts on how to best approach it with marketing techniques.  And I’m going to try to do all that in a short blog, because…well…read on.

Blame it my ADHD

“I feel like I’m from a different planet.”  Years ago, this statement might have been uttered by a person diagnosed with ADHD – Attention-deficit/hyperactivity disorder.   Yet, it’s how I feel today when I compare myself to the average modern day consumer (is there such a thing?), with their impulse decision making, short attention span, and lack of focus.  Does everyone have ADHD?  Of course not – that’s a real mental disorder, and not one to poke fun at.  But call me crazy because when I consider an important purchase, investigate a brand, or generally try to write about something or solve a problem, I take a deliberate approach, use critical thinking skills, do in-depth research, triangulate sources, and plan ahead.  One thing is certain – I know I’m from a different era.

Brain with ADHD

Consider this definition of ADHD from the Mayo clinic website:

“ADHD is a chronic condition that includes a combination of problems, such as difficulty sustaining attention, hyperactivity and impulsive behavior.”[i]

Or this about those with ADHD by William Dodson, MD from ADDitudemag.com:

“The ADHD world is curvilinear. Past, present, and future are never separate and distinct. Everything is now. ADDers live in a permanent present and have a hard time learning from the past or looking into the future…”[ii]

Sound a little like today’s in the moment consumer with a gold fish like attention span, who expects instant gratification and exhibits spontaneous buying behavior?   It’s no secret; the world we live in promotes and fosters this behavior.

  • Consumers are generally impatient and hyperactive. They get upset if they are in any line or have to wait for anything for more than a few seconds.  They are upset if businesses don’t remember them.
  • Consumers can’t focus. They are constantly interrupted or busy with smartphones, multiple devices, packed schedules, and keeping connected with extensive networks.   The average user checks a phone 150 times per day[iii].
  • Consumers live in the moment. They quickly forget their own loyalty history (or lack thereof), and often inflate their true propensity to do something in the future.  They expect their present context to be understood, catered to, and appreciated.

Is there a cure?

Although you won’t change this conduct, you can still accomplish your goals by employing a best practice I’ll term incremental contextual marketing.   As a marketer, accept that there is very little patience, increasingly shorter message shelf lives, and a consumer that is less and less receptive to any outbound marketing, or any message for that matter that is too long or ordinary.   Make no mistake, you can still build brand awareness and garner loyalty – however realize two things:

  • It can take years to build and establish loyalty, yet with one bad experience it can evaporate in seconds, since a consumer’s mindset is, “What have you done for me lately?”
  • You need to message in waves; in small compelling quickly digestible chunks available in any channel.

On the second point, and this is key, even if you have a deeper, more complex, longer term idea that you need to explain or impression you want to make, break it up, and message it over a period of time.   Incremental contextual marketing means both repeating and/or building upon your message, always using current consumer context to reinforce, sustain, adjust, and evolve it, with the end result being a better educated consumer, more likely to remember and to be impressed.   This approach works for marketing a product, educating a consumer on your services, or surveying them for preferences.   The stream of tactics used can be repetitive or additive depending on the goal.

For example, if your primary goal is brand awareness, a repetitive message is fine, but don’t repeat it in the same slots, at the same times, using the same channels.  Mix those up.  Make your points proactively to consumers in their preferred outbound channels, with the same messages available to them when they are in-channel.

If your goal is educating the consumer on retirement strategies, run a series of messages that build on each other, with each subsequent message summarizing key points from the latter, and then appending new ideas.  Use current context to adjust the message, if for instance, the consumer has clearly become interested in college savings plans.  The trick is for each message to be randomly incremental (without a predictable rhythm but still a sustained undertaking) and conversational – strategically placed in varying media at a calculated (but somewhat random) cadence so as to optimize its movement from short-term to long-term memory.  Use storytelling and make it relatable in its tone.   I learned the effectiveness of this method a few years back when I embarked on learning French as an adult – That said, I think you can be much more effective that I was in becoming fluent by not waiting too long before you start.

As a firm, you need an approach, a systemic mindset, and technology stack that can react and adapt to this dynamic environment.  Ironically, some of the same technologies that may be reinforcing this behavior can be used to combat it.  Together, the total solution has to be customer focused, concise, real-time (responding in milliseconds since the consumer won’t wait), consistent, contextual, and unified.  It must be unwavering in its ability to identify consumers with pinpoint accuracy, summon a photographic memory of their interests and preferences, factor in their current context and condition, and then act with predictive intelligence, rendering easily consumed yet compelling content, all while still delivering actions with a personalized / human touch.

We live in a fast paced, short story world, with millions of micro blogs (less than 500 words), billions of videos (less than 3 minutes), and trillions of twitter posts (less than 141 characters).  Accept what you can’t change and use the right methods with the time crunched consumer, craftily using bits of incremental air time to make your points.

You made it to the end (a blog with nearly 1000 words…congratulations) and so perhaps you are like me.  For the rest from earth, however, I’ll need more posts.

Comments and alternative views are always welcomed.

[i] http://www.mayoclinic.org/diseases-conditions/adhd/basics/definition/con-20023647

[ii] Dodson, William.  http://www.additudemag.com/adhd/article/10497.html

[iii] Meeker, Mary and Wu, Liang.  Internet Trends D11 Conference.  Kleiner Perkins Caufield Byers (2013): 52