One principle first, because it decides almost everything else and it's the part most solo devs get backwards:

A catalogue that lives on someone's phone and syncs through their iCloud costs you nothing to keep running. Renting that back to them every month is what earns a one-star review reading "held my own data hostage". But a barcode lookup, an image match, a market index — those cost you real money on every call, forever. Those are the things a subscription is honest about.

Every number on this page is illustrative and deliberately conservative — they're there so you can argue with the shape, not to be believed. Apple takes 15% while you're under $1M/year (Small Business Program), which is what I've used throughout.

v1 has no marketplace, no valuations, no pro mode. So there is no transaction to skim and no "professional" tier to sell. What you have is a well-made tool for someone who owns 200 objects and cares about them — and that person is, historically, extremely willing to pay once and never think about it again.

Read that as the honest middle case: one year of evenings buys you a decent used car, not a salary. Ten times the downloads is €200k and a real business; that multiple comes from press, a category moment, or an audience you already have — not from the pricing page. The reason to ship v1 anyway is that it's the only way to find out whether people love it, and v2 is where the money actually lives.

v2 changes the economics completely, because for the first time the app sits next to a transaction — and next to two things people already pay for without blinking: insurance and knowing what something is worth.

That's a modest one-person salary, and it's the case where things go reasonably well — not the dream. Two things move it by an order of magnitude, and neither is a pricing decision: the insurance partnership converting better than 400 policies, or the business tier finding a real niche. Both are conversations with people, which is the part you said you're bad at, and also the part with the highest return per hour on this page.

You're right, and it's the most important thing on this page. The companies you named didn't win by having one model — they won by matching how their market already behaves. And they split cleanly into two families:

You said you're not against people finding each other and selling directly. That single sentence decides your whole business model: you are family 2. Everything else follows from it — and it's why the seller subscription and boosts hold up while the take rate doesn't. Never build revenue that depends on the deal closing where you can see it.

Mechanically this is one codebase and a market profile per App Store storefront: which monetisation surfaces exist, which price tier, whether rewarded video is compiled in, what the share sheet defaults to. Three profiles, chosen by storefront, not by a user setting — asking someone "do you mind ads?" is asking them to choose to be shown ads, and the answer is always no.

And one small thing to do at launch that costs nothing: log downloads by country from day one. Where the users actually come from will tell you which of these three you're really building for, and it is almost never the one you expected.

v1 is a long build and you need it to earn — so the real question isn't "is v2 a good idea", it's "has v1 proved the demand that makes v2 worth another year of evenings". Here's what to instrument. Everything below is measurable in v1, without a marketplace existing.

Four and five are the gate, and they're the reason to keep wishlists in v1 even though they earn nothing there. A wishlist is a purchase intent you can count before you've built anything to buy with, and a duplicate flag is a seller pre-registering. Put those two side by side and the decision reads itself:

On the middle case, v1 nets around €20k in year one — real money, not a year of salary. So don't plan for v1 to pay for a twelve-month v2. Two consequences, and they're both about scope:

Make v1 smaller than it currently is. Catalogue, lending, wishlist, share. Fifty-three screens is a designed product; a shippable v1 is maybe thirty of them. Every screen you cut is a week you get back, and none of the gates above need the complete app to be measured.

Slice v2 so the first slice earns. Price history and comps are read-only data — no index, no listings, no moderation — and they're the part sellers pay for. Ship that alone, priced monthly, and let it fund the index. If the numbers above come back weak, you've lost one slice, not a year.

Tool libraries, toy libraries, associative media libraries, school and club collections, repair cafés, seed libraries — and recycleries, which are a different job (they liquidate rather than lend, so they need v2's pricing before you're useful to them). What they share: things go out, things must come back, and right now that's a paper ledger, a WhatsApp group and a Facebook post.

That's the whole strategic point, and it survives the fact that an association has no money. A recyclerie with 400 customers and a Facebook page sends 400 people a rendered object card a month — warm, local, object-obsessed strangers, arriving through the received-share screen that needs no account and no download. Twenty organisations is a launch. Twenty paying seats is €7k and changes nothing.

So: give it away to associations, charge the ones with a budget, and count the members it exposes you to rather than the seats it sells. If v3 ever earns more than the consumer app, that's a signal to become a different company — see trap 4.

Six questions, one coffee. 1. Walk me through one object from arriving at the door to leaving with someone — every step, every piece of paper. 2. What do you actually lose track of: the object, the person, the price, or the state of repair? 3. Who would use this — one coordinator, or every volunteer on shift? 4. Do you have anyone's email, and are you allowed to write to them? 5. If this existed and worked perfectly, who in your organisation could sign off €20 a month — and does that person exist at all? 6. How do people currently find out what's on your shelves this week? That last answer is the one that tells you whether the distribution argument is real.

One last thing, since you said this is the part you're bad at: none of the above is a business skill. The pricing is copyable and the numbers are arithmetic. The two moves that actually matter — talking to an insurer, and asking twenty collectors what they'd pay — are both just conversations, and you can have them before you write another line of code.