What’s the first thing you think of when I say the word “Nirvana?” I guess it depends on your age, but for me it’s the early 90’s grunge band that had teenagers everywhere smelling “Like Teen Spirit.” But lately, the term “nirvana”, has been used in a different context. You see – omnichannel is nirvana for most financial institutions. A utopian place where every credit union wants to reside, but none can attain.
Every bank or credit union I speak to talks about how they want to be (or even claim that they are) a true omnichannel institution. I feel it is my civic duty (not to mention my job) to bring them back down to earth by giving them the definition of omnichannel. An omnichannel organization has gone beyond single-channel, multichannel and cross-channel to deliver and orchestrate offers across all member touch points and channels seamlessly. In their terms, this means that if I accept an offer on my online banking mobile app, then the ATM I visit just a minute later will react accordingly after I’ve inserted my card. A few minutes later, when I walk into the branch and apply for a mortgage, they know all of the previous actions I’ve taken. For most financial institutions, they may claim omnichannel, but the reality is that they aren’t quite there.
One of the core capabilities for omnichannel organizations is being able to perform analytical optimization. You must have the capability to select the optimal message from a pool of messages that could be presented to a member, based on supporting data such as time, channel, device and stage in the member journey or life cycle. This can be difficult to do, especially if you are using business-rules-based or priority-based optimization instead of analytical optimization. The former typically rely on SQL or event-triggered responses to present an if/then conditional logic, but this doesn’t always work when more than one variable is the input to an optimization problem.
So how do credit unions decide what an individual member needs at an exact moment on a specific channel? Many credit unions set out to answer that question using marketing optimization. The goal of marketing optimization is to analytically decide which campaign, communications or messages should go to what individual using the entire member population as the pool. It also can determine the frequency and channel preference the member has, all while staying inside the confines of any organizational rules, constraints, policies or budgets that may exist. Often this decision is needed in real time – while the institution is interacting with the member via a call center, in the branch or at the ATM.
These are five best practices that I have seen project teams learn while heading toward their “Omnichannel Optimization Nirvana”:
Set requirements and define success early. Ask: What do you consider a requirement? Hardware, software, resources on the project, teams and channels that should contribute to the running of the optimization scenarios – define all of those things up front. Define and plan for integrations and APIs to channel. Then ask: What are you optimizing for? Fewer pieces of direct mail sent, fewer emails sent per day, higher profit or ROI from certain channels? Define success for the implementation and deployment of the solutions as well; it’s important to know what you want to achieve before you start shooting in the dark.
Don’t let data stand in the way. Make sure wherever you are sourcing your data from is on board with the project. If it’s just member data, or risk data, or even fraud data, ensure departments inside the organization (like IT) are on board with what you are asking of them. Often, they have to be educated on how the optimization solution works as well, so consider doing that during the beginning stages of the project.
Have the right number of cooks in the kitchen. This one is important, especially at financial institutions where there may be many departments vying for ownership of the solution. Make sure that the digital marketing, optimization, analytics, measurement and IT departments are involved up front in determining roles and responsibilities.
Fail early and often, because you can. With the optimization solution, you can attempt many scenarios and tweak input variables as needed until the scenario is correct and the sensitivity analysis is on point. Know that you can try these varying scenarios over and over again, iterating on them, before they are pushed live into market.
Don’t limit yourself to marketing. As the name suggests, marketing is often the first place brands begin to use marketing optimization. What you will find, however, is that there are many other uses for the solution – including collections, risk, sales, service, etc. While in the planning stages, consider how the solution might evolve over the years.