Case study · Social platform · research to MVP plan · Digi117 venture

Peoplarium: from swipe fatigue to a self-serve meetup platform

What did Digi117 do, and what came out?

Digi117 took Peoplarium from a question about swipe fatigue to a costed, buildable plan in three versions: research, analysis, concept, architecture, business model and the decisions behind each cut. The result is a self-serve web platform where verified members meet in person, launching in Vancouver around January 2027 and breaking even near 4,600 monthly active members once Plus is live.

As of September 2026 · Peoplarium is a Digi117 venture in development.

  • 28research tasks in 4 parallel waves
  • 3plan versions, each cutting more
  • ~13 weeksto launch at 30 founder hours a week
  • ~CAD 20kcash before launch, own venture, not a client price
  • ~4,600monthly active members to break even
  • Trust and safety
  • AI
  • Regulation
  • Growth
  • Infrastructure
  • Pricing

The brief

Dating and friendship apps have a fatigue problem. People swipe dozens of times before one real conversation, delete the app, and come back months later. The question was whether a platform built around meeting in person, with verified people and without a swipe feed, could work in one city, and what it would take to build and run.

Peoplarium is Digi117's own venture. That makes it a clean case study: the full chain from research to price is ours, and so are the mistakes we caught along the way.

What came out

The plan went through three versions. Each one cut more.

  • v0.3 mapped the market, the law and a broad concept.
  • v1.0 consolidated the research into a host-led launch: AI-composed tables, paid hosts, a moderation team.
  • v2.0 removed everything that needed people we would run. Members plan and host; AI suggests and moderates; the platform provides tools and never takes part offline.

The v2 plan is a web app a solo founder can build and run with a fleet of AI coding agents: about 420 founder hours, about CAD 20k in cash before launch, public launch in Vancouver around 5 January 2027.

How to read this case study

Each chapter answers one question a prospective client would ask, in two sentences, then shows the evidence. Three chapters are interactive: filter the research, click through the architecture, and change the business model's inputs.

Every number is a plan, not a result. Peoplarium has not launched. Numbers are rounded; the calculator shows the model behind them.

The rules that did not move

Hard rules, in every version. No AI-generated nudity or sexual content. AI never impersonates a member or acts as a romantic companion. Consent is never delegated to AI. Minors are banned everywhere. Members decide and organize; AI suggests and moderates; the platform provides tools and never takes part offline.

The chapters

  1. Chapter 1

    Research

    How do you find out what is true?

    We ran 28 research tasks in parallel with AI research agents, in four waves, and then checked the decision-relevant claims against their primary sources. The checks caught real errors, including a tenfold cost mistake and a product described as dead that had relaunched, so every published finding carries its source, its date and its check status.

    Interactive: research filters

  2. Chapter 2

    Analysis

    What do the market and the law say?

    The swipe model is shrinking while intentional, niche and in-person formats grow, and the products that failed did so for reasons a design can avoid: unverified strangers, live video with no guardrails, AI that pretends to be you. The law in Canada, the US, the EU and the UK rewards platforms that provide tools, act fast on reports and never promise what they cannot guarantee.

  3. Chapter 3

    Concept

    What exactly are we building?

    Peoplarium is a tool, not an organizer: verified members find each other, talk and plan their own meetups, while AI suggests and moderates. The platform never runs events, pays hosts, holds money for meetups or supervises what people do offline.

  4. Chapter 4

    Architecture

    How does it work, and what does it run on?

    Peoplarium is one modular web application on Cloudflare's edge with Postgres behind it, and every photo, message and report passes through one AI moderation pipeline before anyone sees it. Nothing needs a team to run: vendors handle identity checks and the few human reviews the law requires, and the rest is configuration.

    Interactive: architecture explorer

  5. Chapter 5

    Business model

    Does it make money, and when?

    With AI coding agents doing the typing, the MVP needs about 420 founder hours and about CAD 20k in cash, and launches around 5 January 2027 at 30 hours a week. Once Plus is live it breaks even near 4,600 monthly active members, and the share of members who pay moves the result far more than the price does.

    Interactive: calculator

  6. Chapter 6

    Decisions

    What did you cut, and why?

    Across three plan versions we cut, changed or deferred 29 items: some because they are illegal or against platform terms, some because they are unsafe, some because they are outdated, and the largest group because they relied on people the platform would have to run. Saying no, with a reason, is where most of the value in the plan came from.

Ask about this, or request more analysis

Your message goes to Vadym Chernega, founder of Digi117, not a bot. Reply within one business day.

We use your email only to answer. No list, no newsletter. Want this kind of work on your own product? See what we make.