Data itself has no intrinsic value.
Let me say that again for budding data entrepreneurs.
Data itself has no intrinsic value.
After my first appearance on the My First Million pod with Sam Parr and Shaan Puri, a bunch of folks reached out saying they had a dataset on X or Y (or an idea for a data business).
Their questions ultimately boiled down to “How can I make money off of this?”
My general observation after seeing what most of them had was that they were going 90 mph in the wrong direction.
Why?
Because…
…Data itself has no intrinsic value.
Just having data, even if proprietary or really hard to get, does not make it valuable. The customer doesn’t care how difficult it was for you to get or that you’re the only one with it.
Data is only valuable if it delivers value to the customer.
Below, I’ll offer a framework you can apply to assess if a dataset is valuable. If you’re just starting out or thinking about building a data business, scrutinize your idea with the ECO framework.
The 3 dimensions of the framework.
- E – Edge & Stakes for Customers
- Value Proposition: Does the dataset provide significant value to customers, such as competitive advantage or critical decision-making support?
- Ideally, you want to tap into fear or greed and this usually comes in the form of making them more money, keeping them compliant with the law, reducing expenses, etc.
- If you find yourself saying “Wouldn’t it be cool if…?” or the value proposition is exclusively about time saving, that’s usually a tell that the edge your data delivers and the stakes for the customer are not high enough.
- A lot of analytics data products fall into this low stakes camp. They are interesting to have and they might be bought in a bull market but in a downturn, they are the first things to go as they usually don’t provide that edge.
- C – Collection Feasibility – Can you collect the data?
- How easy or difficult is to collect the data? What are the costs? If licensing data from others, what happens if they cease to exist?
- A lot of data inbounds I get are from folks who have a ‘hack’ to get some data. This is very cool but they never consider what happens if that partner changes the rules, gets acquired, goes out of this business, etc. You don’t want to have the equivalent of “supply chain risk” in your data collection efforts, and if you have that risk, you should think of ways to mitigate especially if you want to build a big data company.
- I covered the 7 methods to acquire data for a data business here with examples of companies doing each.
- O – Opportunity and Market Demand – Is the opportunity big and attractive?
- Who is going to use this and what meaningful change will it drive for them?
- Are there enough of these customers? And are they growing?
- Is it high stakes enough or does it deliver enough of an edge to get paid a lot for it?
- Are the customers ‘savvy’ enough to use data?
- What alternatives, if any, are they using today?
- How often do they need it? You want frequency of need which enables recurring revenue and reduces churn.
- warning: DO NOT TRY TO CREATE NEW BEHAVIORS OF STUFF THEY SHOULD DO. Doing that means you’re making 2 sales. The first is to convince them to change their behavior and the 2nd is to use your product. Don’t make your life more difficult unneccessarily.
As you think about the opportunity, it’s worth remembering who buys data:
- Financial services- banks, insurers, hedge funds for investing, risk management, benchmarking, etc
- Sales & marketing including ad tech
- AI models
This doesn’t mean other industries don’t buy data, especially specialized data, but per bullet 4 under the Opportunity section above, they might not be data savvy.
I’d also strongly advise you to avoid trying to sell data to dying industries, i.e. media for example. I’ve heard of data or analytics products targeted at the newsrooms of media companies. I cannot think of a worse market to go after.
If folks want a quick assessment of the value of their data, I’ll review/share my thoughts on their dataset’s value if they can share a bit more in the comments below.
Leave a Reply