How to Carefully Plan Your MarTech Implementation (and Save Time + Money)
While MarTech tools undoubtedly solve pain points, implementing them can create new ones.🤔
It’s fascinating how it happens. The best intentions are brought to the purchase cycle. Your team has even run an RFP and found a partner who could solve each presented use case. Heck, even the engineering team is behind the decision. Check, check, check! Implementation goes as expected, and you’re getting value out of the tool. The pain point is solved. Everyone gathers for a glass of bourbon in this fantasy MarTech world.
Fast-forward three years. Your CFO wants to recoup the budget and asks what tools you can live without. You reevaluate the tool and find that one of your other tools has a similar feature that solves the pain point to equal effect. You align on resourcing, then migrate the functionality and cut the tool. Bam.☄
And yes, you just spent three years of your budget on a tool you didn’t need in the first place. And no, you’re not alone. This is a pervasive problem that’s more common than you think.
Most evaluations fail to identify a tool’s specific features that solve particular use cases. Let’s review how to do this correctly.
Example: a FinTech brand has a use case of upselling high-yield savings accounts to customers with average monthly balances exceeding $30k and an engagement score greater than 0.75. The MarTech stack consists of Iterable, Hightouch, BigQuery, and Tray.io.
Features needed are deep segmentation, validation of marketing permissions, a data science model to generate engagement scores, the ability to send email and SMS communications to targeted customers, and a pipeline to send data securely.
Based on those four requirements, here are the features we can tap into:
✔ Deep segmentation based on attribute and event data
✔ Out of the box ML engagement model
✔ Sends targeted email and SMS
✔ Holds a record of marketing permissions
✔ Audience segmentation based on data within the warehouse
✔ Pipes data securely from the data warehouse
✔ Holds the segmentation criteria, including user attributes
✔ Holds a record of users’ marketing permissions
✔ Generates data science model for engagement score
✔ Pipes data securely from the data warehouse
There’s a lot of overlap. One use case isn’t enough to run a consolidation exercise, but it is important to note available options. Using out-of-the-box features, this use case could be solved with just Iterable, Hightouch, and BigQuery. Alternatively, use your own data science model and do most of the heavy lifting through Hightouch’s Audience feature.
The path forward will depend on the underlying data required, the level of effort to accomplish it with any given combination of tools, and any additional costs associated with activating those features that enable the solution.
It’s a lot of work, don’t get me wrong, but having this level of clarity from the outset will prevent over-tooling and ultimately increase the efficiency of your team.
Ready to unlock the full potential of your data?
Talk to Ragnarok about implementing a robust data dictionary and elevating your personalization efforts to new heights. Your customers will thank you for it.
