NEW RESEARCH79% of shoppers say the same chain is noticeably better or worse store to store.Get the full report, free →

New research: why the same chain wins in one store and leaks in another.Get the report →

Causal inference analytics for multi-site retail.

Find out what your stores are capable of selling.

Our story

  • Consumer brands have measured what their advertising does for decades, with media mix models and matched-market tests. Our founder did that work at P&G and Google, and saw that multi-site operators rarely get that kind of measurement for the changes they make in their own stores.
  • Seurat began by measuring what EV chargers add to the stores that host them. It grew into reads for loyalty, deals, food, remodels and the other changes a chain makes.
  • The name comes from the painter Georges Seurat, whose pictures are made of thousands of small dots. A read works the same way: each store gets its own result, and together the results show the whole chain.

Who leads Seurat

Matt Pereira

Matt Pereira

Co-founder and CEO

At P&G, Matt oversaw the standards for the media mix models behind more than $1B of North American ad spend, then worked in brand management. He spent nearly 8 years at Google, where he led insights for consumer brands such as PepsiCo, Mars and Mondelēz and commissioned matched-market tests of their campaigns. He has a BA from Harvard and an MS in Business Analytics and AI from NYU, speaks Mandarin, and was a recent guest on the NACS Convenience Matters podcast.

"Mid-market operators deserve the same kind of analytics the biggest retailers and the Fortune 50 have relied on for years. Causal inference shows which changes paid off at each of your stores, and I can't wait to show you what it finds."

Our team has worked at

Our team has also

  • Pioneered ad personalization at Google
  • Won a Peabody Award
  • Founded and built companies in fintech, climatetech, adtech and healthtech

What we believe

  • Intuition and evidence work best together. You can know your sales drivers well and still be surprised by the data. A read shows where your instincts were right and where the results differ from what you expected.
  • Finding out what worked shouldn't take a data team. A read starts from 1 sales export from the back-office system you already run. There's no software to install and no systems to connect.
  • Knowing what didn't pay off matters as much as knowing what did. A change can help at some stores and lose money at others. You'll see which is which, so you can stop spending where it isn't working.
  • Your results are confidential. We never publish a client's results or name a client without their permission.
Ownership

An outside firm, independently owned. No parent company, no holding company.

Memberships
2026 NACS MemberConexxus Member
Contact

engage@seuratanalytics.com