The Salesforce one is the only one I'd actually pay for, and I'm not even a Salesforce admin

Ten cards today. Nine of them came from the same competitor-gap scrape, which tells you something right away: this is a batch of "someone built X, there's a gap" deductions, not raw demand. Those are fine but you have to filter hard, because a gap existing doesn't mean anyone's wallet is open.

The one that stopped me was Instant Salesforce metadata inspect and bulk export for admins, at a 77 — highest score in weeks. The pitch is dead simple: "every time I need to check a field or run a SOQL query I have to open five tabs." That's not an AI idea, that's a pain complaint, and pain complaints are the only thing I trust. The monetization makes sense too — $5–10/user/mo per seat, annual team plans, enterprise tier for audit logging. Salesforce admins work in companies that already pay $150+/user/month for the platform itself, so a $10 extension is a rounding error on an expense report. That's the sweet spot for indie pricing: invisible to procurement, painful to live without.

Would I build it? Honestly, maybe. The problem is it's a browser extension that reads Salesforce metadata, and Salesforce admins are the most locked-down user base on earth — enterprise IT will ask you about data handling, and your extension touches their org. That's a real sales tax. But the distribution path is clean: AppExchange, Salesforce subreddits, the admin Slack communities. I'd build it as a Chrome extension with a thin backend for the bulk-export jobs, charge per seat, and ignore everyone who isn't an admin. Stack: Chrome Manifest V3, the Salesforce REST API, Supabase for auth and job storage, Stripe for billing. Two weeks of work, maybe three, mostly fighting OAuth.

Second one I actually liked: Paste-and-go SQL index advisor for Postgres, 70. The line that got me was "Postgres and Supabase consoles bury slow-query debugging across panels" — that's true and I've felt it. pganalyze is the incumbent and it's enterprise-priced and heavy. A thing where you paste an EXPLAIN, get a plain-English "add this index, here's why, here's the expected win" is genuinely useful and takes a weekend to prototype. I've built worse.

But I'm skeptical of the economics. Free tier of 3 analyses a day is generous enough that most people never convert, and the $12–19/mo prosumer SQL market is the same crowd that just uses Postgres logs and vibes. Devs are cheap when the tool isn't in their deploy path. The $49 early-bird lifetime is a mistake, by the way — everyone launching does that now and it just caps your revenue at the moment you need momentum most. If I built this I'd skip the freemium and go $9/mo from day one with a 7-day trial, and target Supabase specifically since their users are already paying for tooling.

The drone flight log analyzer is the one I'd stay away from, despite the 69 and the decent $9–19/mo pricing. The compliance angle is the only real hook — commercial drone operators need flight records for insurance and Part 107 stuff — but that's a small, niche, regulation-shaped market, and every drone platform (DJI especially) is slowly absorbing this. Building a whole product on parsing DJI's proprietary log formats means you're one firmware update from a support nightmare. Hard pass. I'd rather sell into the Excel pain than the drone pain.

The spreadsheet stats analyzer is the one that sounds good in a pitch and dies in the wild. The customer who can't figure out t-tests is also the customer who won't pay $19/mo for t-tests, and the customer who would pay already has a stats package or a data person. I've seen this movie. Weak.

If I only got one shot this month it's the Salesforce extension — high pain, buyers with budget, tiny build cost. The rest are homework.