Loading page…
AI search analytics helps marketing teams see where their brands appear and what to improve.
2,000+ customers · 55 team members · 14 months after launch
Sell the urgent job buyers already feel: give the CMO evidence about whether a new acquisition channel is working.
Fewer bars = easier, cheaper, or faster for an AI-assisted solo builder. Editorial judgments based on the case details.
Peec AI helps companies understand their visibility in ChatGPT, Perplexity, Gemini and other AI search tools. Founder Marius Meiners and cofounders Daniel and Toby arrived at the idea after abandoning several legal and regulatory technology concepts that did not produce urgent demand. They joined Antler without a settled product, interviewed people in the SEO industry, and noticed that leading practitioners were already discussing how AI answers could change discovery. Meiners built a prototype with V0 in about a day and a half. It accepted prompts, queried LLM APIs, and displayed mentioned brands and sentiment scores from 1 to 100. The team secured eight or nine letters of intent and Antler committed €100,000, then Toby built the production product in six weeks. They launched in February. Meiners said Peec reached more than 2,000 customers and $8.6 million ARR after 14 months, with 55 employees. The founders chose a lower entry price and a mid-market audience rather than focusing on enterprise contracts.
Peec’s first customer conversations came from following the people already discussing AI search, rather than asking a broad market to care about a new category. The founders watched LinkedIn and X for posts about AI visibility, then contacted those authors with a working prototype and an invitation to try it free.
The quick-reference card — acquisition channels, replication playbook, risks & traps, and our verdict — is members-only.
Join the Unicorn Club — $5/moCancel anytime, in one clickPaste into Claude Code / Codex and get a working version of this product end-to-end.
Two production-grade prompts per idea — a full build spec and an SEO growth plan — plus 10 tailored build specs a week. Members only.
Join the Unicorn Club — $5/moCancel anytime, in one clickEvery claim traces back to a listed source — open any of them and check the number yourself. The exact words are quoted below, with a timestamp that jumps to the second each figure is said.
“And now we are at a little bit over 2,000 customers and 8.6 million ARR and 55 team members.”
1:30 · hear it in the videoData credibility: Founder Marius Meiners states 2,000+ customers, 8.6 million ARR (the episode title gives it in dollars), 55 team members, and the channel shares (about 30% of revenue from word of mouth, about 20% of conversions from AI search) in a SaaS Club interview. No dashboard or audited statements are shown. The transcript alternates between 85 and 89 euros for the entry price; this entry uses the €85 figure from the host's summary.
The founders initially used letters of intent to test whether companies would buy the product, but Meiners was careful to call LOIs weak evidence because they are legally nonbinding. The first commercial proof came when early prospects tried the product and were asked after seven days whether they would pay.
Peec’s growth loop combined founder visibility, customer referrals, and the fact that AI search itself could introduce buyers to the product. Meiners said roughly 30% of revenue came through word of mouth and about 20% of conversions came through AI search.
Peec’s founders cycled through ideas that failed to generate strong market pull, and they treated those periods as short experiments rather than commitments that had to be defended. For several concepts, they spent two or three weeks trying to sell before dropping them.
Peec is an AI search visibility analytics product for companies that want to understand how their brands appear in answers from systems such as ChatGPT, Perplexity, and Gemini. The early prototype accepted a set of prompts, sent them to language model APIs, extracted answers, listed mentioned brands, and scored how a brand was described on a 1-to-100 sentiment scale. That core monitoring job remains part of the product. The broader workflow is to identify which prompts matter to a company’s buyers, track presence and descriptions across repeated queries, and help marketers see where the models source information. Meiners described Reddit as an important source for B2B SaaS brands and suggested using visibility data to guide community engagement. The buyer is generally a marketing leader or SEO practitioner who needs to report on a new acquisition channel. The product competes on faster time to insight for small and mid-market teams, rather than months of enterprise integration. Its job is measurement and prioritization; it does not itself guarantee that an AI system will cite a customer.
Peec uses subscription pricing, with an entry plan around €85 per month when the founders launched, substantially below competitors charging more than €500. The strategy was to serve a broad mid-market of smaller companies rather than a limited set of large enterprises. Meiners said the company reached $8.6 million ARR and more than 2,000 customers around 14 months after launch. Those figures are founder reported and were not independently verified in the interview. A lower-priced plan can work when onboarding is self-service, the product returns useful results quickly, and gross margins remain healthy. For a comparable company, meter plans on monitored brands, prompt volume, reporting history, and team seats, while keeping the first paid tier understandable to a one-person marketing team. Add higher plans for agencies or multi-brand organizations only when usage patterns justify them. Be cautious about pricing by API volume alone: buyers care about a decision-quality report, while model costs can vary. Review retention by cohort to learn whether monitoring becomes a durable workflow.
The prototype used V0 and language model APIs, with a basic workflow that sent prompts, extracted generated answers, and displayed brand mentions and sentiment. The production implementation took six weeks and was led by cofounder Toby, who had stronger technical depth. Our read: a similar product has variable costs from querying multiple models, maintaining prompt schedules, parsing answers, and storing historical results. The business also needs careful handling of model changes, since output formats and citations can shift without warning. Keep inference spend bounded with plan limits, caching where appropriate, and sampling policies that preserve useful comparisons. Build a quality check around prompt reproducibility and visible timestamps so customers can distinguish a trend from a one-off response. A 55-person company at the reported scale can invest in data quality, sales, support, and product coverage; a small entrant should narrow supported models and prompt volume until retention shows where expansion pays.
Copy Peec’s habit of searching for existing demand before building. The founders looked for professionals already discussing AI search, tested multiple ideas quickly, and built a prototype only after conversations showed an active topic. Their first version demonstrated the core job rather than attempting a complete marketing suite. Copy the move toward a defined buyer: a small team that needs insight quickly can adopt a focused subscription more easily than an enterprise platform requiring lengthy integrations. Change the implementation to fit a narrower market, such as one industry with a standard set of buyer prompts and a concise weekly report. Validate with payment, not only LOIs, and be precise about the limits of any AI visibility score. A solo founder can begin with a manual benchmark service and a small number of monitored prompts, then automate the repeated work. Avoid assuming that category growth alone proves a durable software business; test whether customers renew after the initial curiosity wave and whether reporting changes their marketing actions.
AI search behavior and model interfaces can change quickly, making historical comparisons difficult and the underlying measurement less stable than conventional web analytics. A product may show mentions without proving that those mentions caused traffic or revenue. Peec also operates in a competitive category where established SEO vendors and new entrants can add similar tracking features. Its relatively low entry price depends on efficient onboarding and infrastructure, while a 55-person organization carries more operating cost than a small bootstrapped tool. For a new entrant, the largest risk is building a dashboard that marketers check once but do not use to make decisions. Tie reports to concrete actions such as content updates, community participation, or sales enablement, and track renewal by use case. Explain sampling, prompt selection, model versions, and date ranges so customers understand what a score means. Do not promise ranking or citation outcomes that the product cannot control. Validate that small teams retain the tool before investing in broad model coverage or enterprise integrations.
The four founder playbooks — first customers, first dollar, the flywheel, the silent stretch — are in the Unicorn Club, on this idea and on all 1,403 others. The story, the numbers, the sources and every later section stay open to everyone.
Join the Unicorn Club — $5/moCancel anytime, in one click