Startups just don’t have a future in selling their AI data, venture capitalist says
Startups just don’t have a future in selling their AI data, venture capitalist says
TIM LEEMASTER
Voyager Capital Managing Director James Newell
Startups hoping to capitalize on all the masses of data they collect from their AI algorithms should think again, says James Newell, managing director at Voyager Capital.
“I can almost assuredly tell them no wants their data because they can’t figure out what to do with it,” according to Newell. “Yet so many founders feel like they’re sitting on a goldmine.”
Even the actual startup that generates the data may struggle to use it effectively let alone someone who had no role in its conception, development, acquisition or processing, he adds.
“It’s the number one thing founders get wrong,” Newell says – that they can have both develop the main business that they started with and attract interest in a data cache.
The Seattle-based fund, focused on first-round investments, just announced a fifth fund at USD 100m and has already participated in five funding rounds, that along with other investors, raised USD 2 to USD 8m each.
The company is keenly focused on the firms with a business to business orientation in software as services, almost all of which are increasingly leveraging AI and machine learning. The venture firm further limits itself to Washington, Oregon and British Columbia-based startups.
A first step for Voyager when looking at a potential funding possibility, like any AI focused investment firm, is working out if the technology the company is using is really bona fide AI or not.
“We throw technical resources at them to pull apart the techniques and models they are using to see if they can stand up,” Newell says.
Somewhat surprisingly he isn’t that much interested in the technical depth of the algorithms and AI systems a startup may have developed.
“Where is the value for customers in improved decision making or automation or efficiency,” Newell asks. “It doesn’t matter how cool the technique is or how complicated your deep learning model is – that’s great – but if you’re not actually driving business value then it really doesn’t matter.”
Voyager was founded in 1997 by former consultant and Westin Hotels executive Bill McAleer.
Previous investments include Wise.io, a machine learning company which was acquired by GE in 2016 for an undisclosed sum.
Voyager plans to lead four to six investments a year with the new USD 100m fund. Each investment will total between USD 2 to USD 8m driven by a consortium of funds.
Marketing firms are a core area for Voyager as well as digital media, the infrastructure around content delivery and logistics, particularly former Amazon staff who have set out a shingle with their own shop, Newell says.
Previously Newell worked for Institutional Venture Partners, a later stage venture capital company based in Menlo Park, and was a technology banker at Bank of America Merrill Lynch.