In Part 1, I explored how and why to test for the right mix of channels. In Part 2, I covered channel activity coordination and consistency. In this post, I delve into using technology to achieve effective channel integration for optimal marketing decision making.
What is effective channel integration?
Channel integration is really all about intelligent data sharing for just-in-time decision making in the appropriate channel. Marketing across channels in a way that is advantageous to both you and your customers involves efficiently sharing descriptive attributes about them that can be leveraged to make the best recommendations and personalized experiences when they are in channel, or when you are proactively communicating with them.
To illustrate, think about what you expect a business to be able to do when you go to their website. Should they know your location? Should they be attentive to how you arrived at the site? Should they be aware of your recent purchases or activity in another channel? What if today you achieved a new tier of loyalty? I think we would all say that not only should they know these things (assuming proper permissions are given), but they should take them into account in the actions they decide to take in this channel interaction.
Yet too often, firms are not effective with getting the right data to the systems charged with calculating the next best actions for the channel in question. Why is this? From what I’ve observed, it boils down to these reasons:
- Channel ownership in terms of data and decision rule making is still done in silos.
- Channel owners are not given incentives to make changes for the greater good, but instead have measurements in place that encourage channel myopia. Without executive leadership for change, status quo remains.
- If the time to get the right data and make a decision based on it is deemed to take too long, the tradeoffs to make this work are quickly dismissed as too costly or risky.
- The analytical science, methodology and process involved to get to the right set of customer attributes for a given channel are not well understood, and there isn’t enough priority placed on solving for this.
- The candidate pool of potentially predictive data that should be tested is not readily available, and the process to make it accessible and distill it down appears to involve too much effort, time and cost.
Why is channel integration important?
Without effective channel coordination and integration, as a business you will make less relevant and timely marketing actions, and risk failing behind your competition resulting in less satisfied customers, lost customers, and declining market share.
Evidence suggests that customers expect companies to understand them, at least in terms of being aware of past interactions and information voluntarily surrendered[i]. Moreover, they are increasingly cross-channel shoppers, using more than one channel in their buying process. If consumers sense firms “don’t have a clue”, it’s not uncommon for them to feel it’s their social responsibility to broadcast digitally via reviews, social media, or blog posts. This can have an adverse impact on a company’s brand image. Think of it as a negative promoter score – or “net demoter score” that firms should aspire to improve.
Conversely, consumers often react favorably to activities and promotions that are relevant to them, and in the short run will reward firms with higher response rates (it’s not uncommon to achieve 3x or higher), yet more importantly long-term loyalty improves because of increased levels of trust, convenience, and meeting requirements, leading to higher lifetime value and more referrals.
Consider the dilemma businesses face today and then ask if better channel integration as described helps crack this. More than ever, consumers are impatient on two sometimes opposing fronts – they expect the right answer and they expect it fast.
- It’s been said that today speed trumps quality, but actually consumers expect both, and if they don’t get both from you, they will shop around – and consumer switching, especially for younger segments, is on the rise[ii]
- What does it mean to be making decisions in real-time – e.g., providing speedy answers? I submit that what is most important is providing decisions in the right amount of time to meet true customer expectations and needs. So if a customer is clearly seeking deeper content on a particular product, offering them alternative product recommendations in sub-seconds may be less useful to them (and to their likelihood to buy anything from you) versus providing them detailed content on that product minutes or even hours later. This is what is meant by JIT (Just in Time) decision making.
What systems are required to achieve effective channel integration?
Channel integration as described entails a number of systematic capabilities such as:
- Maintaining a universally accessible channel preference center or data store that houses:
- Consumer stated preferences
- Channel effectiveness indicators organized by events and actions. For instance, if a consumer states a preference for direct mail yet data suggests they respond better to new product offers in emails, preserve and share that
- Channel effectiveness indicators organized by timing. Again, if its known that certain consumers have higher click thru rates on emails sent on Wednesdays between 11:30a and 1pm eastern time, codify it
- Fast access to customer behavior and demographic attributes that can be fed to a given channel decision engine
- Fast access to model scores and the ability to request recalculations as necessary for an interaction in a given channel
- Fast access to personalized content and promotions that can be suitably combined with a prescribed channel action
- Ability for one channel system to trigger transactional actions on another channel, such as when a customer is on a website, and the next best action is to send a personalized SMS
- Facility to orchestrate a series of prescribed actions across channels that are set into motion by a genesis event or action on a given channel. The resulting flow diagram enabled by such a system might look something like this (there are many such systems in the market today):
Comments and alternative views are always welcomed.
[i] My Buys – 6th Annual Personalization Consumer Survey; January 2014
[ii] National Consumer Agency – Market Research Findings: Consumer Switching Behaviour; September 2013