Teikametrics raises $10M to optimize Amazon ads

Teikametrics is a Boston-based startup that helps retailers tackle the challenges of advertising on Amazon. Today, the company is announcing that it has raised $10 million in Series A funding.

CEO Alasdair McLean-Foreman said third-party sellers represent 60 percent of the transactions on Amazon. But they don’t have any real data science capabilities, so they need help advertise their goods in a way that maximizes profitability.

“We are using big data to help sellers optimize for profitability,” McLean-Foreman said. He compared it to the work that Amazon has done “optimizing on the consumer side — all the advanced econometrics” to determine things like the price of Amazon Prime. “We’re on the other side. We’re helping sellers and brands.”

That’s a very different challenge from optimizing Facebook ads to get the most clicks. McLean-Foreman argued that it’s not even something Amazon can do properly, because, “They don’t have critical information on cost of goods sold, and they also don’t have the context of being on the supply chain side.”

(At the same time, he emphasized, “We’re aligned with Amazon, we’re pro-Amazon and we’ve built our company off the back of Amazon.”)

In contrast, Teikametrics — through its “retail optimization platform” Flywheel — allows sellers to incorporate things like transaction data, inventory data and pricing data. So when they look at the results of of their campaigns, they can see their gross profit margins and profitability after ad spend.

How appealing is this to sellers? Well Teikametrics says it’s being used by advertisers who represent 1 percent of all sales on Amazon, including brands like Razer, Power Practical and Zipline Ski. Eventually, the company plans to expand its technology beyond Amazon, to other marketplaces.

Teikametrics has been bootstrapped since its founding in 2013, at least until now. McLean-Foreman said he decided to raise outside funding because “the crown jewel is the sheer amount of data that we can model,” which means hiring “a tremendous amount of very, very high-powered machine learning folks.”

The Series A funding was led by Granite Point Capital, Jump Capital and FJ Lab.