I’ve said it over and over again, “the sole purpose of collecting data is to take action on it and you need to take those actions as close to real-time as possible.” While we’d all agree with that statement, it’s getting to the point that becomes the challenge. I’ve written several newsletters over the past year focused on different ways to integrate data - Email Marketing, VOC, Call Center, etc. - but have yet to touch on one of the more valuable integration points for your business: CRM.
I write these newsletters because I’ve watched many organizations become too fixated on creating an “omni-channel view” while missing very easy ways to activate data today based on real business use-cases. Business use-cases should be the driver for all technological design. Simply put, if there isn’t a business use-case then there isn’t a reason to move forward with the project plan.
CRM data opens the door to creating that "omni-channel view" without actually trying to build a giant, clunky database that sits internally on a server cluster somewhere. However, when you think about content personalization, you risk making the wrong content decision for an individual if you don’t have all the cards. In that regard, what “omni-channel” should mean to you is creating a data activation framework that allows you to hold as many cards as possible about that individual or persona that you can actually use to make decisions.
There are three steps to consider as you build out your CRM integration:
Visitor ID Framework
KPI/Use Case Development
Data Activation Framework
Visitor ID Framework
With the multi-channel nature of your business comes the ability to identify a visitor in many ways. Whether it’s a cookie, a username, an email address, etc., at the end of the day you have a one-to-many relationship between an individual and the identifiers for that individual. Depending on the channel, one ID may hold more weight over another. For CRM, that ID tends to be an account ID, transaction ID, or email address.
Take the time to map out all of the IDs you have at your disposal across your channels at the onset of your integration project. For many of our clients, this exercise ends up being a table in Excel that outlines the ID, the source of the ID, etc.
Now that you have your IDs outlined you can start to work through the actual reason(s) you are integrating your CRM in the first place. For many organizations, this becomes a bit of a struggle because you have to move beyond the “send everything in” philosophy that you tend to hear when an organization has skipped this step.
With the goal of creating an optimized experience for your customers, here are a few CRM integration use-cases that we have put in place for clients to get you started:
The last X purchases made by that customer
The X most traveled to destinations by quarter for that customer
The last X email offers a customer has interacted with
Data Activation Framework
Once you have the ID mapping completed and the use-cases created, the next and most complex step is the “how.” To be able to respond and alter content in real-time, you need to have the CRM data made available almost immediately. To do that, you typically would combine the power of your CRM (and the APIs available within it) with a cloud-based database with very little latency. The cloud-based database would query your CRM as needed when it needs new information, while also keeping a smaller subset of that data available in real-time for immediate response. Combining that with a Strategic Data Layer and Tag Management will give you the most control and efficiency in building this framework.
In our industry, the visitor database and data activation space is heating up more and more each day. While you could build your own dataset in the cloud using Amazon or similar, we have had a lot of success using the following for clients (listed alphabetically):
Adobe Audience Manager
The biggest limiter to taking this more simplistic, yet effective, approach to integrating data ends up being the people you meet with along the way. Whenever it comes to data integration, you can definitely find at least these 4 personas in your meetings:
The person/group who wants to put every, single metric into every, single system just in case you ever need it.
The person/group who is so concerned with privacy and security that they make Ben Stiller’s character in Along Came Polly look like an adrenaline junkie.
The person/group that doesn’t trust the cloud, Google, or anything that doesn’t involve the old-school way of managing data.
The person/group that reads an article and thinks every dataset belongs in [insert trendy database here].
Education will become a major part of your job as you work to deploy the data automation framework we have outlined throughout this newsletter. You’ll have to work with the 4 personas above to gain alignment and identify the right business stakeholders to ensure you pick the use-cases that will resonate most within your organization, and thus, add the most value in the shortest amount of time. Just remember that omni-channel means that you try to incorporate some information from all channels, not all information from some channels.
For more information on building a Data Automation Framework, check out this PAGEor contact me directly!