Of course, data is one of the big trending topics. And of course the data people in a company are happy about a budget. However, a high level of frustration (emotional & financial) can occur
in a company if the simplest rules are not observed in a data project.
-
Goals, questions, goals. You have no defined goals and no questions, don't start collecting and visualizing any data, but to do something with data, just because some manager
heard something about data at some fair!
- Asking the right questions, i.e. specific questions:
-
- Counterproductive question: How can we increase conversions? Why aren't my website visitors converting?
- Productive questions: Which online campaign performed best and how can we repeat the success? Did the change xy on the website increase the conversion? Where do we lose the website
visitors in the checkout process?
- Asking the right questions on to which you cannot answer "So what?"
-
- Counterproductive questions: How many website visitors do we have? How many conversions do we have?
- Productive questions: Have we met our conversion goals? What are our top-performing products and what can we learn from them? Is our company dependent on one traffic source (= high
company risk)?
-
Data people: I have seen too many times that a data scientist is deep in his work and loses the business perspective. In the end, the results from analyses don't bring
any business value or, to be blunt about it: Management celebrates because "something was done with data" but nobody uses that data/analyses/dashboards.
-
Business people: Active support from the business side is therefore necessary to ask the right questions and define goals.
-
Management: Management must understand, use and implement the data and the KPIs.
-
Data-driven corporate culture: The understanding that you can make decisions based on data and should monitor your business must be deeply rooted in the corporate culture and
not just a to-do that you do when you have time. One of my favorite sentences from a product owner (not): "The data topic is important, but I had other things to do!" Of course, everyone
always has a lot to do, but defining the most important KPI and having them monitored once a month is something product owners must be able to do! As said above, you shouldn't create data
graveyards, but every product or project must have at least one goal and corresponding key performance indicators.
-
Firmly defined responsibilities