We're growing a mobile app with free web tools instead of ads. Here's the playbook so far.
We're two friends building Poker Reflex, a mobile app that trains your preflop
poker decisions. Classic problem: a paid app, no ad budget, and a niche audience.
So instead of running ads, we build free web tools around the same niche and use them as the top of the funnel. The counterintuitive part: a small free tool that solves one specific problem ranks and gets shared far better than the app itself, because people share useful tools, not ads. The tool pulls in search traffic and community shares, and the site converts a slice of that into app installs.
We've shipped three so far:
- A pot odds calculator
- A range visualizer (build and save your own preflop ranges)
- An equity calculator we just launched (hand vs hand win %), which I'm oddly
proud of: the engine is validated against exact enumeration of every possible
board, so the numbers are provably correct.
All free, no signup: https://poker-reflex.com/tools
The app: https://poker-reflex.com/get?src=indiehackers
Where we're at: early but real. The tools are starting to pull organic traffic,
we get a steady trickle of installs, and a premium subscription launches in a few
weeks. Instagram converts, Reddit is hit or miss, and YouTube age-restricts every poker video (fun).
The question I keep chewing on: how do you cleanly measure tool-to-app conversion,
and is a free-tool funnel a real long-term channel or just a slow trickle? Has
anyone grown an app this way? Would love to hear what worked.
Free-tools-as-distribution is criminally underused for early-stage products. The piece most founders miss: the tool itself doesn't need to be tightly related to the paid product — it just needs to serve the SAME ICP at the SAME stage of awareness. A free Lighthouse-score widget pulls in performance-conscious devs; a free meta-tag generator pulls in SEO-conscious site owners. If your paid mobile app serves either of those personas, both tools route traffic to you for years even when you stop maintaining them.
Two operational tips from doing this at scale: (1) make each free tool a STATIC page with its own unique <h1> mapped to the highest-volume keyword you can defend (not a generic "/tools/x" route — those don't rank). (2) Add programmatic schema (FAQPage + HowTo) on every tool page; that's worth ~20% extra organic CTR in 2026 because AI Overviews source from rich snippets first.
The dwell-time signal on a working free tool is the strongest organic-rank moat I've seen in years.
This is a good fit for mobile because the free tool can answer the pre-download trust question, not just drive traffic.
For a food logging app I would not start with a generic calorie calculator, because that attracts people who want a number and then leave. I would test tools that map to the exact moment before app usage: barcode nutrition lookup, photo-to-meal estimate examples, or “what should I log when I ate X restaurant meal?” The tool should make the user think “I need this workflow again tomorrow,” not just “nice calculator.”
For measurement, I’d separate three numbers: tool completion rate, install CTA click after completion, and day-2 app open by campaign. If tool A has lower traffic but higher day-2 opens, that is probably the one teaching the right habit.
This is the sharpest take in the thread, and it reframes my own lineup. The "want a
number and then leave" trap is real: my pot odds and equity calculators are exactly
that, you grab the number and you're gone. The one that maps to your "I need this again
tomorrow" is the range visualizer, because it lets you save and edit your own ranges, so
it's a workflow you come back to, not a lookup. I hadn't drawn that line between my own
three tools until you said it.
And it lines up with the app itself: it's a daily trainer with streaks and an ELO, so the
tool that teaches "come back tomorrow" is the closest cousin to the actual product. The
number-grab tools pull traffic, the habit tool probably pulls the right people.
Your three-number split is also better than what I had planned. Completion rate and
install-CTA-click I had, but "day-2 app open by campaign" is the one that actually answers
it, because for a trainer the win isn't the install, it's whether they came back. Lower
traffic but higher day-2 opens equals the tool teaching the right habit. That's now the
metric I'll use to decide which tool to build next. Genuinely useful, thank you.
Love this. I'm doing a version of it for HealthOS — instead of ads, I wrote up the whole Toyota on-device AI story and put it on my product page. The bet is that content teaching something actually interesting pulls in the people who care about the problem, not just clickers.
The free-tool idea is smart for your niche. I've been toying with a free "voice stress test" web demo for the same reason — let people feel it before they commit to a download.
On attribution: have you tried a separate landing page per tool with App Store campaign tracking? That's my plan, so I can see exactly which channel actually drives installs.
"Content that teaches pulls the people who care, not just clickers" is exactly it. An
ad makes a claim, a tool or a real story lets someone feel the thing before they decide.
Same bet, different shape. The Toyota on-device story on your product page is a great
version of that.
And "let people feel it before they commit to a download" is the cleanest way I've heard
the free-tool logic put. A voice stress test demo is perfect for it, because the demo IS
the pitch. If it works on them, the download mostly sells itself.
On your attribution question: yes, and it's basically our setup already. Each tool is its
own page, and the download links route through a single redirect that stamps a campaign
token, so Apple's App Analytics shows installs per campaign and Play Console shows them
per install referrer. Honest caveat though: it's campaign-LEVEL counts, not per-user
paths. iOS will tell you "this campaign drove N installs" but not what that person did
before the tap. For deciding which channel actually drives installs, that's plenty. For
stitching one user's full journey, it isn't, and that's where you'd need deferred deep
links or an SDK, which I've parked until it's clearly worth the work. For your stage, the
per-tool page plus campaign tokens will already answer your real question.
This "useful tool > useful app" framing is really clicking for me — solo dev here, also paid app + zero ad budget + niche audience, so I've been thinking about the exact same funnel.
Curious about the tool selection process: did you pick "pot odds calculator" etc. based on search volume/keyword research, or just "what would our target user google when they're stuck"? And roughly how long after publishing a tool did you start seeing meaningful organic traffic — weeks or months?
Following this for sure, would love to see the conversion numbers from tool → install once you have more data.
Same boat, with one correction: ours is actually free, not paid (premium is planned,
not live yet), but the zero-budget niche-audience funnel is identical, so all of this
applies to you.
On tool selection: it started as "what does a poker player google the second they're
stuck" way more than a spreadsheet. Pot odds, ranges, and hand equity are the three
things players literally look up mid-hand or right after, so the intent was obvious
before I checked a single volume number. I did sanity-check search volume after,
mostly to decide which to build first and how to name the page. But if I'd led with
volume I'd have risked building something technically searched and useless to solve,
which is the trap. Intent first, volume to prioritize.
On timing, honest answer: too early to give you a clean number from the tools. The pot
odds calc is about two weeks old and the equity calc is days old, so they haven't had
time to rank yet. What I can tell you from the blog side is that meaningful organic was
months, not weeks. A small trickle showed up in a few weeks, but the real compounding,
pages ranking and pulling steady traffic, is a 3 to 6 month game. Anyone promising
faster without budget is selling something.
And yes, I'll post the tool to install conversion numbers once I've actually
instrumented it. A few people in this thread have convinced me that's the next job.
Good luck with yours, happy to compare notes as we both figure this out.
I think the bigger opportunity here might not be the app-to-user funnel, but the tool-to-tool flywheel.
Most founders build one free tool and hope it converts. What you have done is more interesting: each new tool increases the discoverability of the others, which compounds search traffic over time. If the equity calculator, range visualizer, and pot odds calculator all rank for different intents, you're effectively building a mini search ecosystem around poker training.
For attribution, I'd be less worried about perfect tracking and more focused on whether the tools are acquiring users cheaper than any paid channel could. If organic traffic keeps growing while install costs stay near zero, that's a pretty strong signal.
Curious: have you seen users coming back to multiple tools, or is most traffic one and done?
The tool-to-tool flywheel framing is sharper than how I'd been thinking about it,
thank you. That's basically the bet without me having named it: the tools and the
blog articles all cross-link, so every new page is another entry point that also
feeds the others. The equity calc isn't just a third tool, it's a third front door.
On attribution, you and a couple of others in this thread have pushed me the same
way: stop chasing perfect tracking, ask whether CAC beats any paid channel. With
install cost near zero and organic still climbing, the honest answer is yes, and
that's the signal that actually matters.
To your question, and it bugs me that I can't answer it cleanly yet: the tools do
link to each other so the path exists, and anecdotally blog readers do land on a
tool. But the equity calc is only two days old, so the sample for real multi-tool
sessions is too thin to call one-and-done vs sticky. Instrumenting that, tool-to-tool
movement and repeat visits, is now top of my list, because if people hop between
tools it changes how I treat the whole thing. I'll report back once I have it.
At this point it feels less like a funnel problem and more like you’re building a network of entry points where discovery is the product.
Once that clicks, conversion stops being something you chase and starts happening as a side effect of reach.
Yeah, that reframe sticks. If discovery is the product, my job stops being "optimize a
funnel" and becomes "build more things genuinely worth finding," which is a much
healthier thing to wake up to.
One discipline it forces, and I want to stay honest here: reach only becomes installs if
the bridge from each tool to the app stays honest and frictionless. Discovery as the
product, sure, but the product still has to earn the tap. That's the part I won't let
myself hand-wave. Appreciate you sharpening this, genuinely.
That feels like the right constraint.
Discovery can be the engine, but the trust in that bridge is what keeps it from turning into traffic noise. If that stays clean, the compounding effect actually becomes durable instead of just spiky growth.
Appreciate the thoughtful pushback that’s the part that usually decides whether this kind of system actually holds up in practice.
Durable over spiky, that's the whole game. This was genuinely one of the sharper
threads I've had on here, so thank you for pushing on it. I'll circle back with real
numbers once the system has had time to prove whether it holds. Until then, back to
building front doors.
That’s a solid way to frame it.
Looking forward to seeing what the numbers say once it’s had time to settle. this is one of those cases where the system will probably teach more than the discussion ever could. Good luck with it.
Appreciate it, will do. And you're right, the system will teach me more than I could
guess at right now. Good luck with yours too.
Grown a product almost entirely on free tools, so a couple concrete bits on your questions.
Measuring tool→app: the web→mobile jump is where attribution dies. A ?src= param gets you to the store and then you lose the thread. What worked for me: fire an event at the tool's "aha" moment (the calc spitting out the answer), one on the install CTA click, then a deferred deep link / attribution SDK (Branch etc) so the install actually ties back to the tool that drove it. Without that last bit you're guessing.
Slow trickle vs real channel: it's a trickle that compounds, which isn't the same as slow. The tools keep ranking and pulling traffic months later for zero extra spend, so it stacks. The thing that decided whether a tool actually fed the product for me: it had to fully solve the thing someone googled. My "lite version of the paid app" tools flopped, the ones that stood on their own converted. Your equity calc being provably correct is exactly the kind that earns links and trust, btw.
Which of the three pulls the most install-intent traffic? Usually one ends up carrying it.
This is gold, thank you. You nailed exactly where we're weak: we're on ?src=
right now, so we literally lose the thread at the store like you said. The
deferred deep link / Branch layer is the obvious missing piece, firing an event
at the "aha" moment plus the install CTA click and tying it through an
attribution SDK is going on the list this week. Without it we're guessing, agreed.
The "fully solve the googled thing vs lite version of the app" point really lands.
Our weakest tool is the one that feels closest to a teaser; the standalone
calculators pull better. Good to hear that pattern held for you too.
On which of the three carries it: honestly too early and too blind (that exact
attribution gap) to say with confidence. Early read is the pot odds calc has the
most search volume, but the equity calc is two days old. Once Branch is in I'll
actually know. Really appreciate you taking the time.
Hey everyone, I’m the other friend building Poker Reflex.
We’d honestly love to get your feedback on this. We’re putting a lot of energy into the app and these free tools, and we’re trying to figure out if this “useful tools first, app second” approach can become a real growth channel.
If you have thoughts, experience with this kind of funnel, or even brutal feedback on the tools/app, we’d be really grateful.
Thanks, and have a great day.
Free-tool funnels work long-term, but they compound slowly and most people undercount what they actually cost. The traffic looks free because there is no ad spend, but the engineering and maintenance time to keep three calculators accurate, fast, and ranking is real. Worth tracking your effective cost-per-install including dev hours, not just the zero in the ads column.
On measurement: UTM parameters on the app-store link from each tool get you click attribution, but the harder number is what happens after install. If you can tag the install source (even just a query param on a deep link or a "how did you find us" screen), you can compare 7-day retention by tool. That is the number that matters. A tool with half the traffic but double the retention is the one to double down on.
One pattern I have seen work well in SaaS (different from mobile, but the principle transfers): the best free tools teach a workflow that the paid product automates. Your range visualizer does this. Someone who builds and saves custom ranges on the web tool has already invested effort and built a mental model. The app becomes "continue where I left off" rather than "try a new thing." The calculators, by contrast, give a one-shot answer. They build zero switching cost.
If you are choosing where to invest next, I would pick the tool that creates the most user-generated data (saved ranges, tracked sessions, custom configs) over the one that answers the most searches. Data creates return visits. Return visits create installs.
This is a great idea. I often feel that a useful tool, quiz, or even a simple game can be a much stronger acquisition hook than a direct ad because it delivers value before asking for anything in return.
We just launched a music discovery app and are exploring different user acquisition channels ourselves. Your approach resonates because it creates a natural path from solving a specific problem to discovering the broader product.
I'd be curious to see how the funnel performs over time. My intuition is that if the tools continue to rank and provide genuine value, they can become a durable acquisition channel rather than just a trickle.
Keep up the great work, and thanks for sharing the playbook :)
Thank you, that genuinely means a lot. And you said the core better than my whole post did: it delivers value before asking for anything in return. That's the whole difference between a tool and an ad.
Congrats on the music discovery launch. If I were in your shoes I'd lean into how playful music is: a 30-second "find your sound" taste-match or a shareable result quiz maps to the exact pre-download moment, and the result is something people actually post, which an ad never is. Different domain, same bet.
On durability, my honest read matches your intuition: a trickle that keeps ranking and compounding stops being a trickle. The catch is the two conditions you already named, keep ranking and keep being genuinely useful. Lose either and it's back to noise. I'll circle back and post how the funnel actually performs once it's had time to settle. Good luck with yours, and thanks for the kind words :)
good
I'd be careful treating this as a measurement question too quickly.
The interesting question may not be whether tool traffic converts.
It may be what conclusion deserves confidence if it does.
Those sound similar, but they can lead to very different decisions about whether the tools are acquisition assets, standalone products, or something else entirely.
That's not a call I'd make casually from the current signals.
That's a sharper framing than mine, thank you. You're right that I jumped to "does
it convert" when the real fork is what these tools even are: an acquisition
funnel, standalone products, or both. And you're right that I can't honestly make
that call from two weeks of thin data.
If I'm honest, the equity calc could probably stand on its own, while the others
feel more funnel-shaped, which already hints they're not all the same kind of
asset.
Genuine question back: what signal would you want to see before making that call
with confidence? Retention on the tools themselves, repeat visits, something else?
I'd rather know what I'm looking for than collect numbers blindly.
The reason I'd hesitate to answer that directly is that I don't think the interesting part is any individual signal.
I think it's the decision that signal is being used to support.
That's where I'd be careful.
The same metric can look incredibly persuasive while pointing people toward completely different conclusions.
I wouldn't try to unpack that properly in a thread.
If you're curious, drop your email and I'll put together the tighter version.
Fair, and the underlying point is real: the same metric can be used to justify opposite
decisions, so the risk lives in the conclusion, not the number. Worth keeping in front of me.
On the tighter version, I'd honestly rather keep it here. Half the value of this thread has
been that it's public, other people building the same way are reading it, and a rough
version helps them too. If it's easier, IH DMs work fine. I'd just rather not drop an email
into a public thread. Genuinely curious what you'd land on, though.