Hacker News new | past | comments | ask | show | jobs | submit login
Getting our first thousand users in one day (francescodilorenzo.com)
154 points by frankdilo 5 months ago | hide | past | favorite | 55 comments



8% conversion rate is pretty good for a PH/HN audience.

Three things I'm sure helped the signup rates, which others should copy:

- Low-friction registration using SSO.

- Glimpse of the product in the background.

- Let guests see and click on gated features. (Even if visitor clicks "Or try it instead" they still see "+ New Draft" button, which leads to the signup prompt again.)

I've tested these in other apps and found they significantly increase signups. The first because of low friction, the second because--I suspect--it shows the visitor what they're about to experience in just a few more clicks, and the third because it shows the guest what's possible and offers multiple chances to sign up.

Well done!


I'm curious as to what else you've found helps increase sign ups. Do you have the results published anywhere?


Almost... I've been sitting on a draft for months. Will publish it on my site when ready.

Some quick ones:

- Make value proposition clear. Just because they reached the signup screen doesn't mean they're already convinced to sign up.

- Show social proof on signup screen, such as customer logos or approximate number of users. Same reason as above.

- Offer SSO options. Google is especially popular

- Have an Obvious Next Step(tm)(R)(patent pending). Make it clear what the visitor is expected to do next by making that thing visually stand out, and muting (or removing) everything else. Notice in OP's signup screen that the background app preview is tinted, and the Twitter login button is the brightest thing, dead-center of the page.

- Have a product people want. This is a hard one. :)


> - Make value proposition clear. Just because they reached the signup screen doesn't mean they're already convinced to sign up.

I've sat at the signup/register/purchase screen so many times and muttered to my computer, "but I don't know what this thing does, why would I sign up?"

Playing coy was maybe cute in high school. Grownups communicate with each other, and nobody wants to do your job for you and pay you for the privilege (in cash, credit, or PII).


I often find myself in the same situation. A few years ago I started just putting in "whywouldisignupidontevenknowwhatimlookingat@notcomingback.com" or somesuch tailored comment. My hope is that it some human in the chain gets the message and can bring it to the exec that demanded agressive sign up tactics.


I've been on the other end of signups like that. Sorry to tell you they elicit little more than an eyeroll and a swift deletion. Even from executives.


I just close the tab.


Uh, what's PH?



AH! Thank you :)


Product Hunt?


Yes, Product Hunt.


Not PornHub.


Product Hunt


> sparking a controversial discussion in the comments.

I suppose that generates views but to me it seems a cynical abuse of the platform.


He’s done it again!


I don't believe "sparking" was the correct gerund for the OP to use, because indeed it sort of implies that it was intentional.

I'd word it simply:

- a spirited discussion


Yeah definitely meant spirited discussion. Not a native speaker.


It's also not clear how much it helps. I believe HN penalizes overly controversial posts, and it uses quantity/frequency of comments as a signal of controversy.


not his fault people seem to like a fight. It says more about human nature than it does about him.


Well, it certainly says something to profit off it.


I've posted a lot of threads here and you have no control over what people will find controversial or not. If you post a thread and it hits people's nerves it generally happens accidentally. And "profiting off it" is always going to happen if you're posting your own work. So in the end I don't really think it says anything.


this indeed.


Not related to your achievement, but...

If you want “users” - post a well-paid but fake job on Upwork and ask applicants to first check out your site/app.

I don’t suggest this but I’ve noticed it from a couple real jobs I’ve posted. Of course, these “users” will be beyond meaningless for any useful metric. I guess if your app/site was geared toward Upwork freelancers, it could work as a strategy.


lol, nice


Congrats to the Typefully team, I really like the idea, and I'll give a it try soon.

Also thanks for sharing the numbers, and the your insider thoughts.

For everyone else building a product, don't let this success stories discourage you from launching your product sooner, you are not expected to have 1000 user in a one day, launching is not what you want to celebrate, you need to launch as soon as possible, and celebrate product market fit if it ever happens, learning from your users is what can get you there, but you need to launch first!


Yeah definitely. Did many other launches where we had 1/100 of this reach. Just keep shipping.


It's interesting to see the difference these days between PH numbers on launch day and after being featured in the newsletter.

I launched last week and got to the number 3 product of the day. Traffic stats similar to yours. Next day after the newsletter went out and it was a bit of a letdown. Not what I expected.

Would be interesting to see these numbers for you. Maybe PH isn't the launch pad it used to be


Thanks so much for being forthcoming about the donations you received, as well as your commitment to maintaining free features. I've been weighing a paid subscription against voluntary donation for my own bootstrapped product launch, and I seem to keep finding more evidence that the donation model probably isn't the way to sustainability.


I have not yet seen a business that can comfortably survive long term with the “make a single donation” model. Recurring donations are something that can work, however.


We almost had a similar experience with our front-page HN post here https://news.ycombinator.com/item?id=25314170

Except we were not prepared for the traffic, and it took the site down for over 24 hours! Lesson learned...


On my bucket list is using the "Referer" header to aggressively redirect traffic from hug-of-death websites toward cached pages.

It's been a long time since I've worked somewhere that would benefit from that trick, so I haven't had a chance to use it.


That post got just over 100 points - very curious to know roughly what kind of traffic you say from it?


Perhaps it was a slow day if 100 points isn't considered that many - I think the high traffic was from being on the front page!

I wish I could quantify the traffic, but it really did wipe out the server, preventing everyone (including myself) from accessing it! I couldn't even remote in to shutdown/reboot.

The last number I had was just as things were grinding to a halt. It was just over 350 "realtime" users according to google analytics!


Hehe, I didn't mean to say 100 points wasn't a lot, only that it was fairly "standard" for being on the front page, and would make for an interesting metric.


> Yesterday we launched Typefully, a write-only interface for Twitter, after weeks of work and refinement. I told the whole story on Twitter @frankdilo.

The idea of a write-only interface is interesting in itself. I was expecting to see something about it on the Typefully page but didn't.


this will blow your mind... write-once, modify three times.. and read only once on Mondays between 5:30 and 6:30.


chmod 222


One very slightly annoying thing is that the orange/red "over character limit" only appears after the pause/delay.

I don't know if you could signal that more immediately (inline perhaps, without updating the thing on the right)?


Just because people signed up, it doesn't mean they will stick around. I would be curious how much will stick around.


Isn't that a different problem though? It's marketing to get them to sign up in the first place, it's product to get them to come back.


I've bought stuff from two companies I like but seldom patronize this year, and in both cases they started sending me ads every day (usually I say no thanks, so either they didn't ask or they used dark patterns to get me to miss it).

Normally that's okay (it's not okay, but I roll my eyes and unsubscribe), but I've discovered that when my order is delayed, due to out of stock in one case, or Black Friday queuing in the other, I'm really not happy having them prompt me to buy more stuff when they haven't given me the stuff I've already paid for. Each email title gives me a jolt of excitement thinking oh, they've finally shipped it! Nope, they want me to buy more stuff.

And with large departments running multiple campaigns at once, it's not uncommon to get 5 emails in a week, all for different things, which to them makes sense but to me is the titular company acting like a hyperkinetic, needy friend. It is very much the same dynamic that results in every department of a company insisting that they are listed on the company home page, preferably at the top, a common trope for beleaguered UI designers to relate over beers. Not just because of Conway's Law, but because of an internal perverse incentive structure to constantly draw attention to your part of the organization. Lookatmelookatmelookatme.

I find this whole pattern of behavior exceptionally unattractive in luxury brands, which I expect to be more cool and collected than this. Marketing in these cases is jeopardizing the 'get them to come back' part of the equation.


It's definitely a product problem to get them back, but I think in the context of the post, we should be careful how we use the term "user".

It sounds like a user in this case is simply one who viewed the page and clicked "sign in via twitter", which is a nice metric for engagement, but just one in many. It makes for a great secondary blog post for additional engagement though. :)


Yes and no. You need the right marketing to attract the right users to your product.


Right, nothing will kill a bad product faster than good marketing.


You mention a tip-jar “patron” strategy, and I took a second look having not noticed anything like that when you launched the actual product.

I still don‘t see anything.

Are you guys hiding a tip button somewhere in the interface after you sign in?

Are you asking for one-offs? Or recurring?

#alwayscurious


Congrats on launching the app! You have mentioned the earnings from a tip-jar system. I also noticed that Typefully links to Mailbrew. It will be interesting to see the traffic numbers to Mailbrew and how many of them ended up signing up/paying for Mailbrew.


Yeah this project was originally born as a “marketing” side project for Mailbrew.

If Mailbrew is focused reading, this is focused writing.

That being said, we haven’t seen that many signups coming from Typefully. Roughly 100.

So, probably didn’t work as a stunt, but worked as its own product. So we want to invest more in it.


Gotta love the whole "frontpage of HN 2.0: meta article about getting to the front page" approach many have found success with recently :D

Seriously though, love seeing the actual metrics on how this process went, thanks!


Thanks for respecting dark mode in browsers that have it set. Sent a tip.


Is this a sort of Twitter Reddit crossover ? I like the two the way they are. One for short form and the other for long form expression.


To even go to Typefully.app, it is asking me to signin with Twitter. Is that how you are getting new users?


You can click "Or try it instead." under the Twitter sign in.


yeh, but trying is just read & understand. For posting I need to signup anyways - which is fine, at least it tempts me to do that.

Feedback:

0. This is similar to https://getchirrapp.com/ but UI is much better here.

1. Though it lets me to do the writing, I want to also

a. Read my posts

b. Read posts by other writers (may be by the root tweet link) Similar to

c. Delete my published article (if not edit)


get this spam off the front page




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: