Matching data products with use cases is a fundamental challenge for data monetization. To better understand how data providers can effectively serve their customers’ use cases, I spoke to Brad Schneider, co-founder and CEO of Nomad Data, and Anthony Cosgrove, Harbr’s co-founder. The conversation covered use cases for data products, how privacy impacts the use of data and each company’s business model, as well as their combined experience of decades spent working with data.
Nomad Data is where qualified data buyers and providers connect, matching use cases with data products. The conversation followed Nomad’s recent launch of enterprise-grade capabilities which extend use case search across organizations, and help data providers manage all their data relationships.
Data providers use Harbr to build, grow, and scale their core data commerce offering.
Thank you both for joining me. Brad, can you tell me a little bit about why you started Nomad?
We were seeing an increase in interest from companies in using external data to run their business in a better way — making better investment decisions to better target clients, increasing efficiency in their operations, things like that.
But the problem is that the data is typically described in the language of variables, rows, and columns. Business problems, on the other hand, are defined in the language of business. And these are very two very different things. So if you’re a business person with a problem that you want to solve with data, it’s not so simple to figure out which data can address your problem. What do you even look for? How do you find the data you need? It may be that there is data that solves your problem, but it’s described in a completely different way to what you are looking for.
So we saw this need to unlock the market. Business people tell us the problem they are trying to solve. We help them find the data sets that do exactly that.
AC, you and Brad have similar backgrounds in terms of working with data in financial services. What motivated you to start Harbr?
I was building a big data platform at HSBC, which meant having to acquire significant amounts of data from internal and external sources from commercial data vendors in the market. The first issue is discovery — where do you find useful data? The next step is evaluating and assessing that data for my use case. Then I had to get that data into a shape where it could be useful.
Getting to the point where it can actually be used was extremely difficult, particularly when the data had to traverse legal and organizational boundaries. There were all sorts of legal and commercial processes that had to be completed in order for that to happen.
I could see that there was a very high value problem to solve. And if we could solve that problem, we would unlock a lot of opportunity both for commercial data vendors and for their customers who are trying to use data to solve meaningful problems because the cost of acquiring and adapting and using data is so high at the moment and it takes so long and there’s so much risk involved in it in terms of, you know, not delivering an appropriate return on investment, a lot of things that could be data driven and could be dramatically better at. So, by reducing the overall barrier to entry for all of the participants in the market, we would be able to unlock a huge amount of value in the global economy.
There is a wide range of use cases that people need data to solve. Have you noticed unifying qualities across these use cases?
There is similarity in the broad things people are looking to do with data. ‘How is this company doing?’ ‘How is this market growing?’ ‘What investment opportunities look like this?’ In some ways, it’s a bit like playing mad libs. The blanks that you fill in will be different, but the structure of the problems are similar.
What’s interesting is that if you look at the requests that data buyers are submitting on our platform, you wouldn’t be able to tell what kind of company is making that request. Is it a consulting firm? Is it an investment firm? The queries look similar; they’re just being used for different things.
I would definitely agree with that. I think there are some fairly classical use cases for data. If you look back at traditional data warehousing and business intelligence, a lot of the use cases were really around organizations trying to understand themselves. ‘What’s our sales cycle?’ ‘Who are our customers?’
Then, they wanted to put themselves in the context of the market: their competitors, their partners, their suppliers, and their customers. And this would obviously go beyond their own data footprint. So then they have to go and acquire data from other sources, hence the existence of third party data providers and professional data vendors.
Where we’re at now is that we’re seeing a second wave where organizations, having gone through digital transformation, have built out data footprints from their kind of everyday work which do go beyond their own context. They’re starting to capture data around every other entity they interact with. And if you’re a large global organization that’s potentially millions, or in some cases billions of individual organizational entities and their behaviors.
So now you have a situation where large organizations have data that’s relevant to other large organizations. People increasingly have different parts of the puzzle. And now, I think we’re starting to see that there’s a wide range of ways of solving a particular problem or answering a particular question. Different types of data can be findable. So they could both answer the same question to a greater or lesser extent, but be completely different in terms of how they help you answer that question.
Definitely. Another point here — one thing we see over and over again every time — is that there is not one answer to each question. There’s not a single solution or a single way to get there, right?
Absolutely. Besides finding this match between use case and data product, what are other obstacles to data purchases going ahead?
If you look at a traditional data marketplace, exchange, or broker, they all primarily generate revenue from the sellers, right? They charge them a commission on every sale they would make. We also don’t touch the data of the vendors. We don’t disclose their identity until they choose to do so to each individual buyer. So ultimately, we’ve tried to remove every hurdle that you could imagine, every objection that you have to overcome. We’re bringing data providers a live qualified lead from a buyer who’s ready to purchase data today. I think by making a couple of small tweaks to the model of a traditional marketplace, we’re able to align incentives really well between the buyer and seller.
Finally, AC, have you seen a shift in how people are accessing and using data?
I think we’re starting to see that instead of providing data, you’re providing some kind of processed output. So that may be taking your data and running it against the model and generating an output from the model, rather than providing the data that then allows you to create the same output. That effectively means that rather than exposing raw data, the raw data is pushed into an application or a visualization or some kind of other output. Ultimately, data consumers don’t really want to receive data and have it because of the risk around it, and providers don’t really want to give it away. What they’re looking for is the insights. How can they get to the insight that enables them to make a decision, take a particular action, or understand something that’s important to them without actually exposing — or being exposed to — raw data. I think that’s where things are heading. If you’re able to do that at scale, that’s far more convenient for data consumers. It’s not just lower risk, but it’s actually higher value and lower cost as well.
To learn more about Nomad Data, check out their website. You can also look at this handy explainer for Nomad and Harbr.