Philip Odelfelt

The pitch fits in one sentence: the right product, on the right shelf, at the right time, at the right price. Philip Odelfelt has spent years and $27 million making that sentence true at the scale of millions of individual products. Simple to say. Almost nobody in the building materials business had done it.

Odelfelt is the co-founder and chief executive of Datavations, a New York company that reads the retail world for the manufacturers who supply it. Think of the companies that make faucets, lumber connectors, power tools, insulation, the things that fill the aisles of a home improvement store. Most of them run on lagging reports, gut feel, and spreadsheets that never quite agree. Datavations gives them real-time, SKU-level visibility into pricing, inventory, assortment, and distribution. In plain terms: it tells a manufacturer what is happening to their product on shelves they do not own.

The story of how he got here starts a long way from Madison Avenue.

The shop floor

A childhood spent around factories

Odelfelt grew up in Sweden, where his father consulted for manufacturers. That meant a young Philip absorbed something most software founders never see up close: how a factory actually runs, how the people who lead them think, what keeps them up at night. It is an unglamorous education, and it turned out to be an advantage. When he later walked into rooms full of industrial executives, he was not a technologist explaining data to skeptics. He spoke their language because he had grown up around it.

He came to the United States for college, graduating from Colgate University. His summers were not typical. He interned in IT at Deutsche Bank, worked at a hedge fund, and, in a detail that stands out on any resume, joined a recruiting program for three-letter agencies in North Africa. The through-line was curiosity about how systems work and a willingness to go where the interesting problems were.

The quant years

Predicting revenue before building the product

Before Datavations, Odelfelt was a quantitative analyst at 7Park Data, a firm owned by Vista Equity. He ranked in the top 1% of employees there. His job was to turn alternative data into financial predictions for some of the largest funds in the world, names like Citadel, Tiger, and Point72. He built a proprietary machine learning model that predicted the revenues of industrial and consumer companies with sub-1% error across five consecutive quarters. That is not a lucky quarter. That is a system that works.

One detail from those years explains a lot about the founder he became. To keep costs down while building the supporting infrastructure, he ran it on Raspberry Pis, the small hobbyist computers that cost a few dollars each. A person who could serve predictions to multibillion-dollar hedge funds and still count every dollar of compute is a specific kind of person. Frugal in the good sense. Comfortable being resourceful.

The math is not clean because rounds overlap with earlier angel capital, but the trajectory is what matters: a seed round, then a Series A more than four times its size.

The obvious next move for someone with that track record was more of the same. He had offers. Hedge funds wanted to acquire what he was building or hire him outright. He said no. He wanted to build the product himself, for the companies whose revenues he had been predicting from the outside.

The company

Datavations, and a $2.3 trillion blind spot

In 2020 he co-founded Datavations with Jacob Lucas. The thesis was straightforward and, in hindsight, obvious: the building materials and home improvement industry is enormous, worth around $2.3 trillion, and it was starved for good data. The companies inside it were making decisions about pricing, inventory, and which products to stock using information that was weeks old and stitched together by hand. A manufacturer might not know it was running low on shelf space at a major retailer until sales had already slipped.

Datavations built what Odelfelt describes as a market visibility engine. It ingests messy, fragmented omnichannel retail data and turns it into something a sales team, a supply chain team, or a pricing team can actually act on. When a competitor has a stockout, the platform can flag it as an opening. When a tariff shifts the math on a category, it can trace the impact. When an assortment is drifting away from what sells, it can say so, product by product.

In July 2025 the company raised a $17 million Series A led by Forestay Capital, with participation from Morpheus Ventures, Sage Venture Partners, and Nevcaut Ventures. The list of individual backers is its own signal: Travis May, the founder of LiveRamp and Datavant, and Rusty Reed, the former CFO of Procore. These are people who have built and scaled exactly the kind of data business Datavations is trying to become. The round brought total funding to $27 million.

"We're at a defining moment in the evolution of data use within the building materials industry," Odelfelt said at the time. The line is measured, not hyped, which is in keeping with how he talks generally. The ambition is to move an entire sector off lagging indicators and gut feel and toward decisions that are faster and more confident because they are grounded in what is actually happening.

The operator

Always fundraising, and eating beans when it counts

Ask Odelfelt about raising money and you get a philosophy rather than a playbook. "I'm always fundraising," he says, meaning the relationships that close a round are built long before the round exists. He is skeptical of volume networking. When he wants to judge whether an investor will actually be useful, he does not ask them for introductions to other investors. He asks them to introduce him to a customer. If they can, they know the space. If they cannot, that tells him something too.

He has been willing to go to unusual places for the right people. Rather than only chasing funds, he approached limited partners, the people whose money sits behind venture funds, and invited domain-expert angels onto his cap table directly. He set up quarterly advisory board meetings before anyone required him to, so that when a priced round came, the governance was already there. Preparation as a habit, not a scramble.

That line is his shorthand for the unglamorous side of running a company. When times get hard, the founder cuts costs, makes the difficult calls, and does not wait for someone else to do it. It is the same instinct that had him running hedge-fund-grade models on hobbyist computers years earlier. He is an avid reader who keeps returning to a short list of books, Zero to One, How Google Works, and The Seven Habits of Highly Effective People, and he treats continuous learning as part of the job rather than a side hobby.

Where it goes

Teaching giants to trust the data

Odelfelt's stated ambition is not really about dashboards. It is about getting industrial giants and Fortune 500 manufacturers to adopt data science and AI at all, to trust a model over a hunch, to make the leap that finance made years ago and that his old world of quant funds takes for granted. The building materials industry has been slow to that shift. He is betting that the companies who make the change first will pull ahead, and that Datavations can be the reason they do.

It is a patient bet on an unfashionable industry, made by someone who has spent his whole career finding signal in places other people overlooked. The shop floors in Sweden, the alternative data at 7Park, the LPs behind the funds. Now the shelves of a home improvement store. Same instinct, new aisle.