The Startup That Ranks Your Complaints by What They Cost You

Every company says it listens to its customers. Almost none can prove what listening actually changed. Somewhere between the survey nobody reads and the quarterly review nobody trusts, the customer's voice goes quiet. Feedier, a software company based in Lille, in the north of France, built its business in that gap - and its pitch is unusually blunt: stop collecting feedback for its own sake, and start ranking it by what it costs you.

Feedier describes itself as an "all-in-one Customer Intelligence Platform, designed to orchestrate the entire Voice of the Customer journey, from listening to action." Stripped of the category language, the product does something specific. It pulls customer signals from wherever they already live - satisfaction surveys, online reviews, social channels, support tickets - and pipes them into one place. It then reads that pile with AI, cross-references it against a company's own business data, and hands each team a shortlist of what to fix, weighted by revenue impact. The dashboard is not the point. The decision is.

01 / THE PROBLEMDrowning in feedback, starving for decisions

The problem Feedier sells against is not a shortage of customer data. It is the opposite. Large organizations run surveys in a dozen countries, monitor review sites, log support conversations and scrape social mentions - and then leave the pile mostly unread. The signal that matters, a churn risk in one region or a broken step in one product line, sits buried under thousands of neutral responses. By the time a human analyst surfaces it, the customer has usually already left.

Feedier's answer is to treat feedback as a triage problem rather than an archive. Its AI flags weak signals - the early, quiet complaints that precede a wave - and scores each issue by how much revenue sits behind it. A loud, angry review from a small account and a mild note from a strategic one are not treated as equals. That reordering is the part customers tend to describe as the difference.

02 / HOW IT WORKSThree layers, one action list

The platform is organized into three layers, and the company is explicit that the first one is about not replacing anything you already own.

That last piece, Ask Feedier, is built to be interrogated in natural language rather than navigated by hand. It is MCP-native, meaning it is designed to be queried not only by people typing questions but by AI agents pulling answers programmatically - a bet that the next interface to customer data is a conversation, not a chart.

03 / THE PIVOTFrom reward points to boardroom reports

Feedier did not start here. In its first life the product was a gamified survey tool - it rewarded people for giving feedback, leaning on points and incentives to lift response rates. It grew from a real customer problem at the logistics firm Heppner Group, where the need was to process feedback at a scale humans could not keep up with. The founder, Baptiste Debever, bootstrapped that early version to five-figure monthly revenue inside a year before later moving on to co-found another startup, Lucis, through Y Combinator.

Then, in 2023, the company did the uncomfortable thing. It rebuilt the platform from the ground up around large language models, raised roughly €3.5M from LocalGlobe and Kima Ventures, and repositioned from a survey app toward enterprise customer intelligence. Under CEO François Forest, by 2024 the product had become an AI-native platform pulling from more than a dozen feedback sources. The gamified rewards era was over.

04 / THE EUROPEAN ANGLEThe data stays here

One choice separates Feedier from the American incumbents more than any feature: where the data goes. The platform runs its AI on Mistral AI, the French model provider, and hosts customer data in Europe. For a European compliance officer weighing a US cloud, that is not a technical footnote - it is the reason a deal clears legal review. Feedier carries ISO 27001:2022 and SOC 2 certifications and is GDPR-compliant, and it frames sovereignty as a product principle rather than a box to tick.

05 / THE MARKETWhere it fits, and against whom

Feedier plays in the Voice-of-the-Customer and experience-management market, a space dominated by large, expensive suites. Its named neighbors include Medallia, Qualtrics, InMoment and QuestBack. Rather than pitch a head-on replacement - a losing move for a startup its size - Feedier positions itself as an overlay. Keep your survey tools; put an intelligence layer on top. That framing is the wedge.

06 / THE CUSTOMERSWho's actually paying

The company reports roughly 430 customers and around $3M in annual recurring revenue, run by a team of about 27 people, more than half of them in R&D. The logos it leads with span property services, logistics and media: Savills, CEVA, RX Global, Wayne Farms and Everyday Health, among others. These are CX, operations and insights teams inside mid-market and enterprise organizations - the buyers who feel the pain of feedback-at-scale most acutely, because they generate the most of it.

In 2025, the story got a tidy footnote. The legal entity behind the product, long registered as Alkalab, adopted its own product's name and became Feedier Technologies - the company finally matching its paperwork to its brand.

07 / THE BUSINESSA quiet argument for European B2B

Feedier's business model is unglamorous and durable: B2B SaaS subscriptions, sold to enterprises that pay recurring licenses to connect their sources, run the analysis and receive automated reporting. There is no US office in the story, no nine-figure raise, no unicorn framing. What there is instead is a company at roughly $3M ARR making the case that serious enterprise software can be built from Lille, on European infrastructure, for customers who would rather their data not take a transatlantic flight.

Watch: François Forest (CEO of Feedier) on the future of CX, the impact of AI and enterprise transformation.