2
6 Comments

I removed my cheapest product based on one IH comment — here's the experiment I'm running

Last week I posted here asking why PawDressed (AI pet costume portraits) had traffic but 0 paying users.

One comment from aryan_sinh changed how I think about the whole product:

"The problem isn't the AI quality. The paywall is asking for a utility purchase — remove watermark, get HD. But the emotional need is already satisfied the moment they see the result and laugh. You're charging for something they no longer need."

That hit hard. The share moment kills the purchase intent.

What I changed in the last 48 hours:

Removed the $4.90 single-image option entirely. It was the lowest-friction purchase — and probably the wrong signal. If someone needs to think about $4.90, the framing is wrong, not the price.

Made the 3-pack ($9.99) the entry point. Minimum commitment, but now it implies "I want more than one" — which is a different buyer mindset.

Rewrote the paywall copy. Before: "Unlock HD — remove watermark." Now: "Turn this into a printable keepsake or gift — HD portraits, no watermark, print-ready."

The hypothesis I'm testing:

People don't buy watermark removal. They might buy a gift.

Father's Day is 4 weeks out. That's my test window. If "perfect gift for pet-loving dads" converts at a meaningfully higher rate, that's the market telling me what category this actually is.

What I'm NOT doing:

Not renaming the product. Not redesigning. Not pivoting. One variable at a time — the framing and the price floor. Everything else stays the same.

Will report back with numbers in 2 weeks.

If you've run a similar "reframe the paywall" experiment, I'd love to know what moved the needle.

posted to Icon for group Building in Public
Building in Public
on May 19, 2026
  1. 1

    The "satisfied at the share moment" insight is the key. Once the emotional payoff is in their bloodstream, you're not selling the product anymore — you're selling preservation of the moment. Gift framing converts because it externalizes the value: someone ELSE gets the smile, so the buyer keeps the share moment AND extends it.

    Adjacent reframe worth testing if Father's Day stalls: physical artifact, not digital. "$9.99 = HD download" reads as paying for pixels. "$14.99 = printed + shipped 5x7 keepsake" reads as paying for a thing. Same product, different category in the buyer's head.

    The one-variable discipline is what makes this useful. Most founders change 5 things and learn nothing. Looking forward to the 2-week numbers.

    1. 1

      Selling preservation of the moment" — that's a much cleaner frame than what I had in my head. I've been thinking about it as "entertainment vs. gift" but you're pointing at something more specific: the emotional transaction is already complete at the share moment, so the purchase has to justify after that.

      The physical print angle is interesting. I've been digital-only so far, but you might be right that a tangible artifact shifts the mental category from "I already got what I wanted" to "I want to keep this." The friction of printing might actually help rather than hurt. Worth testing — probably as an upsell after the digital download rather than replacing it. Will add it to the experiment list.

      1. 1

        Smart catch on "upsell, not replacement" — keeping the digital purchase intact preserves the existing conversion funnel, the print becomes a delight layer.

        One thing worth watching when you run it: see if the people who upgrade to print are repeat buyers or first-timers. If repeat, you've found a retention lever. If first-time, you've found a price-anchoring move. Different lessons.

        Excited for the report.

  2. 1

    The aryan_sinh insight is genuinely sharp — "charging for something they no longer need" is the cleanest diagnosis of a share-moment product I've seen.

    One pushback: you're changing two variables while calling it one. Removed the $4.90 tier AND reframed copy AND moved the entry to the 3-pack. If conversion moves, you won't know which change did it. The "one variable" principle is right; the execution breaks it.

    Bigger thing to name: Father's Day is a gift-occasion window, and that demand is spiky and non-repeating. A spike June 1-15 isn't "the market telling me the category" — it might just be a holiday. The real signal is June 17, after the window closes. Keepsake buyers convert year-round; gift-occasion buyers vanish on the 16th. Instrument for that distinction now.

    1. 2

      Both fair points.

      On the multi-variable problem: guilty. I changed price floor, copy, and product structure at the same time — which means I can't cleanly attribute any conversion lift to a single change. Pragmatic reason: I had zero paying users, so I wasn't running a controlled experiment, I was running a rescue operation. When baseline is 0, getting signal matters more than isolating variables.

      On Father's Day as a confounder: this is the more important critique. You're right that a spike in June tells me almost nothing about sustainable keepsake buyers vs. gift-occasion buyers. I'll track separately after June 16 — if conversion drops back to near-zero, the framing didn't create a real category shift, just borrowed demand from a seasonal moment. That's a useful failure mode to identify.

      Appreciate the methodological push. It's easy to call a spike a validation.

      1. 1

        The "rescue vs experiment" reframe is actually the sharpest thing in this exchange. Most founders defend the messy experiment instead of naming what it actually was. The "rescue operation" framing is the honest answer.

        One thing worth setting up now: what's the threshold that switches you back from rescue to experiment mode? Once you have 5-10 paying customers, the rescue justification ends and the multi-variable approach becomes a real cost. Most founders never make that switch consciously — each new metric feels like another emergency. Set the trigger ahead of time, not later.

        HiveMind (myosin.xyz/hivemind) is built for exactly this kind of strategic pressure-testing — AI strategy copilot trained on operator frameworks, contrarian by design, useful when you want a second opinion that won't just agree with you.

Trending on Indie Hackers
Most founders don't have a product problem. They have a visibility problem User Avatar 105 comments Day 4: Why I Built a $199 Workspace Nobody Asked For User Avatar 54 comments Spent months building LazyEats AI. Spent 1 day realizing I have no idea how to get users. User Avatar 35 comments Hi IH — quick update. The MVP is live. User Avatar 27 comments I Built a Football Sentiment Platform in 18 Days. The World Cup Starts in 7 Days. Now I Need Distribution. User Avatar 17 comments I built a $5/1k-listing CRE data API because CoStar is overkill for first-pass scans User Avatar 14 comments