Data, data, data is always talked about in every business, yet is handled and analyzed so differently. The way I view or think about data in the past was always monotonous, something that the guy in the dev team would be going crazy for in the corner because he can pivot and merge and extrapolate all sorts of magical stuff.
Scary and often intimidating contents, matter of fact. For that reason, I really found myself being reserved or shy from truly getting to know the capabilities of what it brings to the table. Not even at a micro level but even at a high executive level everything, shape or form can be measured, analyzed and understood with data. Boom mic drop.
Suddenly the guy in the corner playing on excel and SQL became a lot more interesting. I think the issue was the way I perceived data. I did not know how to make sense of it. As a result I resorted to what the norm was of ‘talking like you know what you are talking about’ except all the time. In the product management world this quickly became an issue as I realized that product managers eat data as their daily diet. I was a little stressed, no actually I was somewhat stressed since now I have to become friendly with these obscure numbers, ratios or percentages. It was like consuming something you weren’t even sure what it was…
After a decent amount of observation I put together some key things that I have realized in the product manager data diet to be somewhat useful. For starters I found it really helped to think of data as a form of storytelling instead of just binary numbers. Allow me to explain;
– The type of data you track is critical
The type of data you measure is like the setting of where your story is going to take place. It foreshadows relevant information to the reader like, time period, or genre which give a reader an idea of what to expect in the story. Similarly, your data is setting up the foundation for what you want to know as a product manager in order to make sense and relate to your customers.
– The way you track data is important
Making sure you only get the juicy parts of the story is all that matters. Nobody likes a long winded, dragged out plot that is filled with many useless packets of information that you have to filter through. Get to the stuff that matters with data that tells you exactly what you want to know. Forget everything else, even if it does sound cool or make you sound smarter. In the end you just have to sift through more crap.
– Making sense of the data is how you should only be making decisions
It’s like how you perceive and understand the characters and the ending of a plot. A good book or film will leave you satisfied in the conclusion of how the ending of the story goes. It makes sense, we don’t like cliff-hangers and want to know whether in the end of the day there is a clear conclusion to what the data is saying to you. If we did all that work and in the end were not able to come to a realization of what tangible result is occurring? Then most likely, we are measuring the wrong sort of data in the first place. Naturally, we want to make informed and logical decisions. Using data is our way to navigate through the mess of uncertainty.
Truthfully, I find that analyzing data in any shape or form can tell you something. Although I believe that given the limited time and resources that you most likely have, sifting through data is probably the most valuable before you go making judgement calls.
Though making sense of the data can be a struggle, I think that having another member within your team to support you in how to get a final measurement is extremely helpful. By no means do I expect myself to be figuring out how to configure and filter 1000s of rows data. That is a job for a data scientist or an engineer that probably has a background in dealing with that sort of work — but it doesn’t harm knowing how to also.
Although, being succinct and clear about what and why you want specific data points to be measured is far more important, even more when you are working with a team. I found that given the amount of planning and thinking necessary when responsible for a product, having the data metrics outlined beforehand to what to measure is extremely helpful. You don’t have to know how you are going to get the data. That is another conversation with a data scientist or engineer as mentioned before.
Lastly, once you have the data reports sitting crisp on your table the last thing you want to have to deal with is the verbatim of complex crap that you now have to explain to someone else. Again, like a movie or book how does an author or director convey the information to you? In an easy to digestible format. By this, I mean that the visualization of data is extremely important. Anyone can slap a couple of graphs with nasty acronyms on a slide, but a person that is able to take that information and summarize into a one liner — is pure golden.
It’s like the concept of “Too Long Didn’t Read [TL;DR] that’s become a lifesaver in browsing through web articles! However, I have found myself struggling at times with this element also, mostly because to summarize means that you also need to have a thorough understanding of the product and end user — something of which takes time.
Additionally, depending on who you are presenting the data too, means that not all information is relevant. I would say think always first before summarizing on who exactly the data is going to be read by. I don’t think an executive would like to know the percentage ratios, she probably wants to know if they are positive or negative. Less is more in this case.
Finally, remember that along the way of launching the product into the wild, you will always have theories and assumptions in how the end product will interact with the end user. Though, you can plan and predict, data in my opinion is the glass ball in making more sound predictions. Therefore use data to fill in the gaps of verifying the key assumptions along the way in making your product more successful. Just how a good book includes a number of tests of virtue to the protagonist, as the product manager, I have come to realize that we too should use tests in the form of measuring data to fill in the gaps of how our end story/product will flow.
Originally Published on Sept 2018 from my Previous Blog
