The Gap in your Data Strategy (Part 1)
Two simple questions to make your business more data-driven
Every business is constantly facing new challenges, and decision makers are tasked with addressing these challenges quickly. To face these obstacles, operators and leaders are increasingly turning towards data to determine and execute their business strategies. However, business leaders are getting frustrated when the existing data that they’ve spent time and resources to collect, clean and centralize isn’t a panacea. We’re seeing key decision makers reach a chasm between the outcomes they want, and the data they have.
The response to seeing this gap that I’ve observed throughout my career is a collective throwing of hands in the air and saying, “well, looks like data is actually useless here, let’s figure this out another way.”
But this shouldn’t be the response. I argue that all you need is to ask two simple questions to bridge this gap, which will connect the goals of your business to your data strategy, and turn your business into one that is truly data-driven.
So what are these two questions we’re missing?
1. What data do we need to address this problem?
There is a misconception that data just appears — and that’s certainly not true — data must be deliberately collected. Given the dynamic ecosystems businesses operate in, it’s quite unreasonable to expect to have access to all the data at the moment your new critical problem arises. Data takes time to instrument, clean and distribute. So instead of expecting the right data to be at your beck and call and giving up on a data-driven approach when its not, we need to change our frame of mind and ask ourselves what it is we truly need from our data. Once we’ve done this, we can move onto the second critical question.
2. Is this data worth collecting?
While some data is easier to collect than others, it always takes a deliberate effort to do so. Businesses today tend to collect data based on how easy it is to gather, but we should rather take a value-based approach to determining what data we should collect. This means assessing the net value (i.e. potential value minus cost) of a data collection exercise.
I also want to stress the ordering of these questions. You should try to answer the first question without any consideration for the second. Removing constraints in your thinking will help you and your team come up with more diverse and creative solutions.We all have our own perceptions about what data would be worth, or even possible to collect, and it’s best to enter the ideation stage with everyone on your team on the same page — and that’s with any empty page — no constraints on your ideas.
I’m going to cover this topic in two parts.
In part 1 here, I’m going to focus on the high-level framework for bridging this gap in your data strategy with an illustrative example. Part 2 will focus on how you can build qualitative and quantitative arguments to get buy-in for a shift towards a value-based data collection strategy.
Bridging the gap in our data strategy — An example
Suppose I run an email list for data science related content that follows a freemium model. Anyone can sign-up and receive one article per-week for free, but they can also sign-up for a “premium” subscription and get three fresh articles in their inbox each week instead of just one.
As the owner of this email list, who’s trying to earn a living, I have an obvious problem here.
How do I convince free subscribers to start paying for my content?
Right now, I get 5% of folks that sign up as free subscribers to convert to premium each month. However, I notice that free subscribers are giving lots of positive feedback on my articles through comments, which indicates to me that they’re getting value out of their one allotted article and could be getting more value out of additional content with a premium subscription. This leads me to believe there is potential to increase my conversion rate by building a tool to better demonstrate the value of my content to free readers.
Next, I need some sort of solution for better demonstrating the value of my content to free readers.
I consult my brain trust and we decide that I could send free readers one of the three articles I write each week that would most interest them. This would help demonstrate more value to each individual reader, and consequently bump up the free-to-premium conversion rate. But now I have a new problem, how do I determine what my readers are most interested in?
Now I get to the data chasm, which I can bridge with our two questions:
1. What data do we need to address this problem?
Since data science is such a broad topic, I think that gathering data about which subsets of the data science world they’re most interested in will be crucial for solving this problem. This determination leads me to build an additional step to the sign-up flow for my email-list asking for folks to choose up to three data science topics they’re most interested in (e.g. machine learning, visualization, data strategy etc.)
But gathering this data doesn’t come without cost.
2. Is this data worth collecting?
I can hear marketers screaming from here that adding any friction to a sign-up flow is going to lower my overall sign-up rate. While this might be true, we need to keep our value-based approach in mind by asking: Is the reduction in my signup-rate worth the increase in my conversion-rate?
I think it’s important to tackle any business question with both a qualitative and quantitative argument, and I’ll demonstrate a strategy for doing just this. For now, I want to stress the process of addressing the gap in your data strategy but in part 2 of this post, I’ll delve into how you can get buy-in for this approach.
In this example, we’ve laid down the bridge we need to cross this data gap
Instead of deciding on an intuition based approach to this problem (for by instance saying, “oh machine learning is all the hype these days, lets just send free-readers articles about this!”), we chose to figure out what data we needed to solve our problem. Then we teed up a question about whether or not it’s worth it for us to collect this data (and again, I’ll cover this in part 2).
For now, I just want to stress the importance of asking these two questions.
The value of data to businesses has been demonstrated in the market. I’ll let readers decide if it’s really the new oil but, it’s hard to read the news without seeing articles about AI, or massive IPOs of data companies like Snowflake. However, not all data is created equal, and not all data just lands in your data lake.
We need to change our attitudes when we face friction in using data to address our business’s biggest problems. Right now, we might just give up when it’s not easy to take a data-driven approach. But we need to push through this roadblock by taking a value-based approach to how we think about our data. And again, that comes with asking these two questions:
1. What data do we need to address this problem?
2. Is this data worth collecting?
