For the past few years I have been working with a financial services organization to define its marketing technology roadmap. This organization has no shortage of great technology assets but lacked an overall plan for how to use these technologies. The company also lacked a plan to organize and operationalize its data. As a partner, we’ve been hard at work building an extremely robust martech architecture to create a complete consumer event stream and an identity graph to stitch these events together.
Marketing had tons of ideas about how to personalize the consumer experience across many media and channels. This includes personalizing the brand’s home page and implementing site-triggered email for capturing digital events. These ideas were incubated for months but were never implemented. The technology support team insisted that the foundational architecture needed to be complete before any of these ideas could be implemented. The problem is that without being able to do tests along the way, each section of the roadmap couldn’t be tested in advance of its final design.
Many of marketing's best ideas will fail. The message and creative won’t perform or the offer isn't timely or the data isn't accurate. For this reason, martech support organizations need to embrace the idea of failing fast, with minimal technology and data. This learning will inform the longer-term roadmap and help prioritize investments.
One example is that this client was using PII from logged-in customers to personalize the home page with name or account type themes. It would be great if there was an identity service that was capturing signals in real-time to inform the CMS. Then PII data could be collected from the customer services. These projects are in the works and will unlock powerful capabilities in the long term. However, using the web data layer stored within the tag management vendor, the organization was able to write a simple encrypted cookie to persist this data. Using this data and a website testing tool, the company quickly learned which consumer data increased engagement.
One ultimate example of testing was demonstrated in the 2016 presidential election. The Trump campaign worked with adtech vendors to do relentless test and learn to determine what ads worked and didn't work within minutes of its launch. "We weren't afraid to make changes. We weren't afraid to fail. We tried to do things very cheaply, very quickly. And if it wasn't working, we would kill it quickly," he said. "It meant making quick decisions, fixing things that were broken and scaling things that worked." Jared Kushner
It’s vital for brands to have a robust, scalable martech infrastructure, but don't let that long-term vision get in the way of making incremental progress. Organizations need to find ways to do both, learn from failure early on and make progress on quick wins while continuing to lay the foundation for the long term.