There is no analytics without data management -an imperative for digital marketers.

In my experience at startups and large companies, good analytics often boils down to the availability of organized data to answer business questions. This is especially important for digital marketers, with the audience data from many channels pouring in and the need to stay on top of key metrics.

Seemingly simple questions can spin up the entire MarTech engineering team!

“If I increase my spend on display ads retargeting by 20%, for middle of the funnel prospects, what can I expect for conversion lift? Is this different by gender, geo or behavioral attributes( highly engaged vs low potentials)?

To answer this question, the analyst has to look at historical data of similar campaigns and conversions. The customer journey by time and event attributes has to be organized so that a data scientist does not have to spend weeks and months to get it into shape. It boils down to data organization. Without proper information models and organization it is impossible to understand cause and effect, and the best algorithms are not going to be of much help.

Collecting and organizing data with an eye on the key drivers of user acquisition and retention is hard. Because this is hard, the task of organizing user journey data is left to engineers/IT and soon the cart is before the horse, frustrating everyone.

So what can CMOs and marketers and analysts working in marketing and user behavior analysis do?

The foundation for analytics on user behavior is smart data management, taking as inputs raw events from user activity and converting them to analyzable datasets. Marketers should own the task of describing their data needs, add context to the data and drive analytics templates with minimal IT/Engineering help. The right set of tools, that enable marketing data scientists to be self sufficient, are critical.

For example – In order to fully understand customer journey and the implications of various marketing medium spend on eventual conversion and LTV( Life Time Value), a foundation of first party audience data is critical.

1. Can you micro-segment your audience based on first party behavioral data( how they interact with our website, store, mobile app) and combine 2nd and 3rd party data when needed? Requires modeling all events and attributes from all customer touch points on web, mobile and other offline data.
2. Can you track all referrers for each session your prospect is engaged with you, so you can model and attribute? Requires encoding all campaigns and track them for each session.
3. Can you model a customer journey with time (some segments convert in 1 session, some across 4 or 5 sessions spanning multiple days). Requires maintaining a path of all the customer activity across sessions.

Can all this be done without deploying an army of engineers, and is it repeatable or a one time effort?

In this age of the tech savvy CMO teams, its vital for marketers and marketing data scientists to hold and gain expertise in the domain modeling exercise with the right platform and unburden IT and Engineering of this task. The CMO /CIO combination works well when marketing data scientists drive the data strategy and engineers drive the tech stack, scalability and general infrastructure needs.

Views: 1027

Tags: analytics, data, management, marketing, martech, science


You need to be a member of Data Science Central to add comments!

Join Data Science Central

© 2021   TechTarget, Inc.   Powered by

Badges  |  Report an Issue  |  Privacy Policy  |  Terms of Service