adlk.ioIn the same week, Franz got featured on both Hacker News and Product Hunt. Same product, same week, both on the front page. I'd expected the two to be a bigger and a smaller version of the same thing. They weren't. They behaved completely differently, and I only noticed because I went through the traffic afterward instead of just enjoying the spike.
The headline number lies in both directions
Hacker News sent about ten times the traffic of Product Hunt. The Hacker News feature brought 6,287 visits and 6,676 pageviews. Product Hunt brought 604 visits and 659 pageviews. If you stop there the conclusion is obvious: Hacker News won, Product Hunt barely registered, put your effort into Hacker News next time.
That's the wrong conclusion, and it's the kind of wrong that has cost founders a lot of good launches. The headline number lies in both directions. Hacker News's ten-times lead is softer than it looks once you see where the traffic landed and on what device. And Product Hunt's small number is bigger than it looks once you see what it does in the weeks after. The number on the dashboard was where I started looking.
Same launch effort, opposite shapes
The two channels didn't just differ in size. They grew in opposite ways.
Product Hunt started fast. It launched on a Wednesday and peaked the same day at 351 visits, then drifted down. A couple hundred the next day, a couple dozen the day after. It's a clean curve you can see coming and plan around.
Hacker News did nothing for three days. A handful of visits on Wednesday, almost none on Thursday and Friday. Then on Saturday morning Ten Years of Franz hit the front page and everything happened at once. 5,802 visits in a single day, dropping off fast over the weekend. Almost all of the Hacker News traffic came in that one window.
One channel arrived on a phone, the other at a desk
Then I looked at the devices, and the size comparison wasn't the interesting part anymore.
Hacker News came in on mobile. 58% of those visits were phones, against just 11% on desktop. Product Hunt was the opposite: 6% mobile, 72% on a laptop. Same product, same week, and the two audiences showed up on completely different devices.
Volume on the wrong device is not a win
Here's the part that should make any product person wince. Franz is a desktop app. It runs on Mac, Windows and Linux. It doesn't run on a phone.
So Hacker News handed me about 3,675 mobile readers, the biggest traffic day in months, all pointed at a product they couldn't install on the device in their hand. The biggest channel of the week landed on the one screen where the obvious next step does nothing.
A spike is only as good as the next step you give people
I knew a day like this would come, so the next step was already built. On mobile, Franz doesn't just show a dead download button. It lets you send the download link to your inbox, so it's waiting there when you get back to your desk.
On a day where 58% of your biggest channel is on a phone, that one feature is the difference between a spike and a wasted spike. It's a small thing. It only gets built if you think about the audience ahead of time and ask what device they'll show up on and what they can actually do once they're there. It's the same kind of product judgment I wrote about in Why product people make the best AI operators. Knowing what to build is the work, and this is one design decision where you can see it.
Product Hunt had the opposite problem, which is to say it didn't really have one. Fewer people, but they were already on a laptop, one click from installing. Less traffic and more intent, and there's almost nothing between reading about Franz and running it. Those visits were worth more to me.
One channel is a spike, the other plants seeds
The device split explains that one day. The weeks after are a different story.
Hacker News is a one-day flash with almost no tail. It's huge on Saturday and basically gone by Monday. Product Hunt has a second life. Ranking as a top product gets you into newsletters, weekly roundups and "best new tools" posts for weeks afterward. That traffic shows up later under a dozen newsletter and creator links, not under producthunt.com. So the referral number undersells a good Product Hunt day quite a lot, as much of what it earned you never shows up with its name on it.
The map splits too
Even the map told two different stories. Hacker News leaned toward the usual Western developer crowd. 1,418 visits from the US, 1,074 from Germany, 499 from the UK, then the Netherlands, France, Austria and Switzerland behind them. 116 countries in total, but most of the weight in the usual places.
Product Hunt reached further into the newer maker scene. India was high on the list, the second-biggest source after the US, with Brazil, Turkey, Pakistan and Bangladesh all showing up where Hacker News barely did. A smaller map at 76 countries, but quite a different crowd, and probably a hungrier one.
A feature buys reach, not loyalty
Both of these were awareness spikes and nothing more. Both were shallow, right around one page per visit. People came, they looked, and most of them left again. A front-page feature buys you reach and a moment of attention. It was never going to buy you loyalty.
What you do with that one day is the part you control. Which device they show up on, and what happens next. Whether the thing they came to see even runs on the screen in their hand.
Franz has been around since 2016 and a good story still travels. Both front pages proved that. The harder part is making sure the story can travel the last few inches, from the phone it gets read on to the machine it actually runs on.
Working out what your traffic is actually telling you, and building the next step before the spike arrives, is the kind of thing I do with founders. Let's talk.
