Landing page Views blew up but signup low
Hi,
I am marketing my product on various platforms like twitter, indie hackers, hacker news, medium, and stack overflow. I got about 600 page views of my landing page https://stormyapp.com .
I got the stats from the Amazon CloudFront Referrer. 70% of my request has the referrer Not Specified. Does that mean People were directly entering the URL? which is very unlikely as most people don't know about it yet.
Out of those 600, only 20 clicked on the Request Access button and only 4 gave their email addresses.
Since the number of people clicking the request access button is low. Does that mean I am solving a problem that does not appeal to developers?
Thanks and Regards
Two things:
Thanks @sowenjub I am going to talk to the Android Devs about it. How is IOS compile time?
Not happening in the terminal for me. But people using fastlane could be interested I guess.
referrer Not Specified is a common anonymising feature in browsers
you can add your own context to the link with paramters
'Request Access' isn't the best CTA
surveys don't commonly require email as a must, creating users for example do, your possibly using the wrong format for your desired outcomes... don't invent the wheel, look at examples of more popular services and be closer to them..
Your premise isn't very good, why would a developer think his terminal is slow?? is this a bad translation or something? cause later you ask about testing/CI systems... which aren't bound by the terminal, like I feel it's a language/terminology issue...
reading your description after... the wording sounds so much for less technical users, yet your addressing programmer issues.. it's like a miss-match between audience and language...
Thanks, @hatkyinc for reviewing my landing page. I am thinking to add a demo video to the landing page https://www.loom.com/share/483862112a4a438bbafeeba03ed5acbe
What do you think about it?
I will change the CTA to Join Waitlist or something and I will make the email optional.
it at least gives something tangible
your pitch could be more like "waiting forever for npm install? we cut the time by 10 folds"
it's not the best, but it's a direction as it's more relatable than slow terminal IMHO
Are you thinking of this as a payed service, like I'd pay 5$/month to save 1 minute once in a while?
Yes, I am thinking of making it a subscription service based on the duration of the commands you executed. My target audience is people running heavy projects and it takes time for them to run commands locally.
For example, compiling, building android is time-consuming and developer spends a considerable amount of time daily compiling the code.
Do many people do long builds locally? Like not CI
Why do they? Cause if it's money or trust, your service probably won't be much different for them..
I was talking to some of the users. I found that people working in big companies do long builds. By the way how much time it takes for you to compile/build/run code locally? My service is very easy to use and does not require any setup (That's what I am trying to do)
I don't currently have any compile/build stage...
and I normally offload such things to a CI system
I think this belongs more to some worlds more than others..
Maybe cause I was working as DevOps it's a non issue for me to setup these things...
Most companies I worked for put it in a CI straight away, rarely do I see anyone running full testing locally...
I know some silly one offs in some cases like you mentioned npm
people usually pulling off a new repo first time
upgrading a ruby version with an rbenv setup
but I don't think anyone would invest time to solve these, the best case they go vent for a couple of minutes and forget about it for probably 6 months
Also often these silly things have some config that actually makes them faster like using parallel, not getting docs... silly options which you wonder why they aren't default and some people just tweak them.
At other times is knowing how to shorten the tree considerably, for example I've seen someone part rails into it's components and remove one of the larger ones...
on that base I did the reverse and only took a select gem or two.
I get it sometimes your forced into a silly badly designed thingies,
I've always been around people that won't tolerate frequent runs of a minute, and would work around it
so a test that takes 20 minutes with 50 processes, locally you run only the tests that are related to things you changed.
The longest ones people don't fix are the one off installs
but as they are one off, people don't find it valuable to fix.
if I need to do it twice a year, I'll deal with getting a drink or having a chat while it runs...
@raxrb (can't reply any deeper)
You don't have your messaging right yet, so you won't be getting good conversions. Communication is key. This is part of marketing and selling. You need to get that done right first.
"slow terminal" is not an issue anyone has. period.
"slow build times" is something people have for example.
No one that has a "slow build time" issue, would answer a 7 question with forced email, titled "slow terminal" with a button called "please allow me access" to indicate he does
look at the YC pitch video under the curriculum for start ;)
But I guess this is not such a big problem that needs to be solved based on the low click on the Request to access button
Let me share with you my experience. I used to work in Microsoft and when I want to make some change and run the app locally, it used to take around 20 minutes for minor changes, because the codebase became huge. I was talking to other developers in other companies and I found that the monolith codebase has build time of around 20 minutes.
Being a developer, I don't feel well when I'm shown a typeform with a bunch of options. Unless your product is so good (for which I only have some text describing it), I don't see people completing the form.
Why not try collecting emails and then send them an email with some incentive to fill details? This way, you would also get to build your email list for the product.
The problem is people are not clicking to the Request to Access. I think not most of the people are facing the pain point that I am trying to solve. How long does it take for you to do a local build?
My builds are usually less than a minute. But that's also because of the frameworks of my choice. Usually Nextjs.
Will you be interested in a product that reduces the build time to around 20 seconds?
It will be of great help if you can take a survey https://appstormy.typeform.com/to/tnXxgabZ It will less time than your build.
I've taken the survey 👍🏻 Also, just curious, can this be something that can be used in CI/CD pipelines as well?
Thanks for taking the survey. Yes, it can be used with the CI/CD pipeline.
Maybe that's something that would attract more eyes. I'd sure use it if that's the case.
sure, let's discuss over a call? I will schedule a call with you.
I can do that. DM me on twitter :)
I think the first reason is that default terminal is not slow and the second is that annoying survey.
thanks, @RaygunTec, Terminal is fast for a large number of developers. On a scale of 1 to 10 (fastest), How fast is your terminal when you execute commands?
@RaygunTec gentle ping on this
Any possible issue with usability? Bounce rates? If you need a free resource on optimizing your landing page, check this out - https://mprify.com
thanks @parrylee
Enjoy! Downloads are hot now...
I'm a developer who's reasonably happy with the speed of various tasks in my terminal. That being said, I am always be interested in making things faster, but you've got to show me how much faster up-front. I need a video showing a side-by-side comparison of the same command run locally and with Stormy. If that demo is impressive I might be interested for sure.
The link to the demo video which shows comparison of normal command execution and command executed using stormy
https://www.loom.com/share/483862112a4a438bbafeeba03ed5acbe
Explain how it works with less text and a visual, add product shots and add pricing details.
"Make terminal faster" sounds a bit weird grammatically
What os does it work on? Who is it for?
If it's free then give me the download straight away
I'm also kind of confused technically how this could work without a trade off on network or something
Show a gif speed comparison
Feedback is taken. I will add a visual that will tell the user the difference in executing commands using the stormy vs directly executing them.
Also, I will think of changing the text "Make Terminal Faster"
@harlanz The amount of time it takes to send the changes to the server is very low compared to processing the command on the local machine.
Most of the time the bottleneck is the local processing. There are so many processes that keep on running locally which slowdowns the execution of the command.
Take a look at the speed that I achieved by using the Stormy https://www.loom.com/share/483862112a4a438bbafeeba03ed5acbe
@harlanzw will you be interested to try it out?
Maybe I'm getting it wrong, but in your video it looks like you start the local npm install a few seconds after the remote version. Seems like you're putting your finger on the scale a bit. If anything, you should start the local version first - that will look a lot more impressive. As-is I feel immediately suspicious.
@GriffinJohnston yes you are right. I pressed the key but it did not start immediately. The final video that you are seeing is my 4-5 video so I made the mistake of starting the remote first. Anyway, If you take a look you can see that install using the stormy happened in 7 sec while the one using the local or traditional way took about 41 seconds.
If you are not convinced about the time. I can create a different video for you. You can also try for yourself by giving me your details or requesting access on the landing page https://stormyapp.com I will send you the steps to try it. Thanks