A Step-by-Step Guide to Building In-house Marketing Analytics
Marta Fogel heads up Performance Marketing at 7Mind, one of the global leaders in mindfulness and meditation. After completing a MA in Media and Communications, Marta moved to Berlin to begin her journey in the start-up world. Since then, she has established a track record of scaling mobile user acquisition globally at early-stage companies. As an active ambassador for mental well-being, Marta is on a mission to demystify meditation and mindfulness and make them accessible to people worldwide.
Learn more from Mobile Hero Marta.
In a world where data is king, and data-driven processes are your competitive advantage, access to data is no longer an innovation—it’s a requirement.
Suppose you are an app marketer who recently joined a startup or a mid-sized organization. You look at old-school Excel spreadsheets as the only source of data where manual calculations still serve as the daily marketing analytics tool for performance optimization. Don’t get me wrong; there is nothing wrong with using good old Excel. Even with an entire data warehouse at your disposal, you could still find manual calculations useful for micro-analyses. Although data warehouse and visualization tools are extremely helpful for a macro overview, a marketer may overlook data inconsistencies they otherwise would not manually.
But I’m not here to discuss the pros and cons of building an in-house data analytics tool, since there are many helpful companies out there to eliminate the need for building tools internally. Our company opted for an internal solution because we did not want to depend on non-customizable functionalities for our specific needs.
Regardless of our organizational roles, we all desire the same thing: access to organized data to understand our users, optimize product functionality and avoid information silos. Whether you are a fan of manual data optimization or not, here is a step-by-step guide to help your organization get closer to a data-driven user-centric approach.
1. Set the stage through in-depth discussions
Let’s be honest. You can’t build your MarTech Stack without the dev team—the ones who have the magic wand and the ability to structure all the data. But, it is up to you and your team to request and communicate what, why, and how you need data. Your instructions guide developers through the process, so make sure your individual app events setup is clean and tailored to your business needs.
While orchestrating front-end and back-end changes, it is essential to work closely with a data architect to lay the foundation and serve as your go-to person for connecting the web and mobile applications. They will decide on the data warehouse stack, event management solution for all 3rd party marketing tools, and the integration of customer success center and CRM solution (plus any other CDP you might be using).
2. Review useful products
There are many different BI tools in the ecosystem you can choose from, though to be honest, none of them have it all.
Some of the questions a growth marketer should answer are:
- What is your budget?
- Do you need access to real-time reporting?
- What is your team size (often required for several licenses)?
- What is your data visualization tool primarily used for – acquisition, activation, or retention?
- How many data analysts do you have, and what is your data volume?
After answering these questions, narrow down the selection and choose the best actionable insights tool for your team.
3. Figure out legacy data
In parallel, you also need to set up an entire attribution solution. If a marketer does not have a consistent naming convention respected across departments, it creates unnecessary and complicated work to merge both mobile and web applications. Is it worth it for the dev team to go through the hassle of getting historic data up and running, or is it easier to leave legacy data behind and start from scratch? If you choose to start anew, try an attribution tool and a solid naming convention that are not hieroglyphics to your team. More on the naming conventions next.
4. Come up with the optimal naming convention
Ideally, marketers need a naming convention that considers both the parameters currently used and those you may not be using yet (but are likely to test in the future.)
Think about all the departments that will use it. After, sync with the respective team members on the networks—media, campaigns, optimization objectives, projects, user demographics, platform targets, placements and audience targets. Essentially, all parameters that help you determine which campaign outperforms the others and why.
It is not an easy task to create one generic and detailed tool to help maximize performance, but if you can accomplish this—it’s a real game-changer. Once info is injected into data, they are almost impossible to alter. So prior to campaign launches, always double and triple check.
5. Get the juicy insights
Once all the prep work outlined above is completed, you are good to go. Draft your questions and the dashboards you want to be visualized with the help of an organizational data analyst (or perhaps yourself, depending on the stack chosen). Your marketing analytics will always evolve along with the campaigns, so it is impossible to anticipate each scenario and get the perfect functionality in one go. It is important to stay up to date with industry news and keep up with the fast pace of the ever-evolving mobile industry.