Webyn Raises EUR 2.5M Seed Round Led by Sharpstone Capital

We started building Webyn because we kept running into the same problem. European growth teams were using testing tools built for US-first regulatory environments and US-scale traffic volumes. GDPR compliance was an afterthought. Data was leaving the EU. The statistical engines were designed for e-commerce at Amazon's scale, not for SaaS companies with 30,000 monthly visitors and conversion goals measured in demo requests, not shopping carts.

We believed there was a better way to run website experiments specifically for European teams — one where privacy compliance was architectural, not a configuration option, and where the analysis engine was calibrated for the lower traffic volumes and longer sales cycles of B2B software companies. That conviction drove two years of product development. Today, we are closing our Seed Round with EUR 2.5M from Sharpstone Capital, and it gives us the resources to accelerate what we started.

Why Sharpstone Capital

We spoke with several European funds during our fundraising process. Sharpstone Capital stood out for two reasons that mattered to us. The first was their portfolio: they have backed B2B software companies at the Seed stage consistently across France, Germany, and the Netherlands, and they understand the specific challenges of building a sales-led enterprise tool from Paris into the wider European market. Their portfolio companies have navigated exactly the go-to-market challenges we are beginning to face.

The second reason was their perspective on privacy as a product differentiator rather than a compliance burden. Several funds we met treated GDPR-native design as a constraint that limited our addressable market — an assumption that European teams would eventually relax their requirements to use better tools from the US. Sharpstone's view was the opposite: that European data regulations were raising the floor for the entire industry, that US tools would face increasing scrutiny as enforcement intensified, and that building compliance into the core architecture rather than layering it on top was a durable competitive advantage. That perspective aligned exactly with how we had been thinking about the product from the beginning.

What the Funding Enables

The EUR 2.5M Seed Round will be deployed across three areas over the next 18 months.

The largest allocation goes to engineering. We are hiring two additional backend engineers with experience in distributed systems and real-time data pipelines, and one frontend engineer to improve the visual variant editor. Our Bayesian analysis engine is already solid for two-variant A/B tests, but we have significant work ahead to extend it to multivariate experiments and to build the sequential testing infrastructure that will allow teams to run always-on personalization alongside traditional experiments. These are hard engineering problems and we need the people to work on them properly.

The second allocation goes to product. We are investing in the integration layer — specifically, deeper native integrations with Shopify, Webflow, and WordPress to reduce the setup time from "purchased a subscription" to "first experiment running" from hours to under 30 minutes. We have learned from our earliest customers that the biggest drop-off point is not in the sales conversation or the pricing decision; it is in the technical setup phase. Reducing that friction is the highest-priority product goal for the next quarter.

The third allocation supports expansion into two new European markets where we have seen strong inbound demand but limited local presence: Germany and the Netherlands. We will be adding one customer success person in each market, someone who understands the local regulatory environment and can help customers design experiments that meet the requirements of the BfDI and the Dutch Autoriteit Persoonsgegevens respectively. European data protection regulators are not monolithic; their guidance and enforcement priorities differ, and having local expertise matters.

The State of A/B Testing in Europe

One question we get frequently is whether the European market for A/B testing tools is large enough to build a standalone company. The short answer is yes, with nuance. The total addressable market for conversion optimization tools in Europe is estimated at over EUR 400M annually, including both software subscriptions and associated professional services. The software portion alone — which is what Webyn addresses — represents a substantial opportunity at current penetration rates.

The nuance is that European adoption of systematic experimentation is behind US adoption by three to five years. Many mid-market European software companies that should be running ten or fifteen experiments per month are running zero, or one per quarter, or relying on changes made by intuition rather than data. The adoption curve is moving fast, driven by three forces: increasing competition for organic traffic making conversion rate improvements more valuable, growing sophistication of European growth teams who have worked at or been influenced by product-led US companies, and the availability of better tools — including, we hope, Webyn.

The regulatory environment, which is sometimes cited as a headwind for the experimentation market, is in our view a tailwind for Webyn specifically. Every time a European DPA publishes guidance that makes it harder for US-built tools to operate compliantly on European traffic, the case for a purpose-built European alternative strengthens. The CNIL's detailed guidance on experimentation cookies, the German data protection authorities' ongoing scrutiny of cross-Atlantic data transfers, and the Dutch regulators' active enforcement against non-compliant analytics implementations are all events that bring prospective customers to our door asking whether Webyn can solve the compliance problem. It can, because we built it to.

What We Have Learned from Early Customers

We have been live with paying customers for over a year now. The product has changed significantly since we first launched — not in its core architecture, which has been stable, but in the surface area that customers actually spend time in. A few things we learned that shaped the current product.

The experiment setup flow was too long. We originally designed a seven-step experiment creation wizard that covered every configuration option upfront. Customers abandoned it at step three and asked for help. We rebuilt it as a two-step flow — define the goal and the page, then configure everything else later — and setup completion rates jumped substantially.

Results reporting needed to answer the business question, not the statistics question. Early customers were not asking "what is the posterior probability of improvement?" They were asking "should I ship this?" We redesigned the results view around that question: the first thing you see is a clear recommendation — ship, hold, or extend — with the evidence summarized in plain language. The detailed statistics are still accessible one click away, but they are no longer the primary output.

Support for non-technical stakeholders was underestimated. The person running experiments is often a growth manager or a product manager, not an engineer. When the engineer who implemented the snippet is unavailable, the experiment workflow should not stall. We added a no-code segment builder, a simplified variant diff view that shows changes in plain language rather than HTML, and an experiment summary export that growth managers can share in Slack without context-setting about what a posterior probability means.

What Comes Next

The next twelve months are about depth, not breadth. We have customers in five European countries. We are not trying to add five more. We are trying to become the obvious choice for the customers we already serve — the tool they would not want to give up because it is embedded deeply in how they make product decisions. That means better integrations, faster results, more sophisticated analysis options, and support infrastructure that meets the expectations of teams paying for a professional tool.

The AI roadmap deserves a specific note. We use the term "AI-powered" accurately: Webyn's multi-armed bandit engine is genuinely different from traditional A/B testing in how it allocates traffic and declares results. But we are cautious about the ways "AI" is currently being applied to experimentation tools as a marketing term rather than a technical claim. We will add generative AI capabilities when they solve a real problem our customers have — hypothesis generation, variant copywriting, anomaly detection — and not before. We are not going to add a chatbot that helps you write test ideas and call it an AI experimentation platform.

We are grateful to the customers who trusted us early, the team members who have built the product, and Sharpstone Capital for the conviction and capital to take this further. If you are a European growth team that has been running experiments on tools that were not designed for your context, we would welcome a conversation.