The importance of user feedback and community support before launching a new product is huge. Users are full of surprises, it's long been established that developers make many assumptions about how a user will approach, use, or interpret their application. Many of which will be wrong!
What follows is the story of our first app built completely in the open from start to finish - an attempt on our part to help with this problem.
Let us begin by setting the scene…
Chapter 1 - Idea
'Twas the month before Christmas, and all through the house, not a creature was is stirring, not even a… Okay, so not quite, but what was actually happening was Black Friday.
Yes, Black Friday, that slightly repulsive holiday focussed on sucking as much money out of consumers as possible by offering countless deals and pretties.
One can't browse the web at all without being inundated with last chances. No I don't want 40% off everything at Cacti-R-Us, free shipping on a couch, or a half-price jalapeño burger THANK YOU! Actually maybe the last one I could go for. But not because of your dumb deals!
But this isn't just another Black Friday bashing article (even though some places went to "Cyber Tuesday" this year! Come on, that's pushing it).
No, what caught our attention on this most illustrious of days was the emails! So many marketing emails! Most of which I have no memory of signing up for, and certainly no intention of buying from again.
I understand that this is a great medium to capture consumers attention, but after 24 hours I felt like my attention had been blasted out of my ears.
Even opting out or unsubscribing from these emails is a drain on my energy. It's annoying to hunt for the opt out link, especially because that usually means I have to read your stupid marketing garbage. And if you were naïve enough to simply move them to the trash, then you'll quickly realise that this doesn't solve anything, the following week your inbox is swamped again.
As it turns out there is a solution to this problem. However, in return for using services that unsubscribe to these things for you, you have to agree to let them sell all your data for marketing purposes. Call me suspicious, but aren't I trying to get away from marketing purposes?
So what can we do about this?
Chapter 2 - Validation
Well we've discovered a problem, let's see if it's something that bothers other people. So we picked a name and threw together a 10 minute landing page and started sharing it around on social media.
The response was overwhelmingly positive! Leave Me Alone was to become an ethical and privacy first email unsubscription service. Within a few hours we had 50 potential beta users, and a load of ideas and feature requests. All this before we'd written a single line of code.
We decided that going forward we would get community feedback on every decision we made. This included features, design, colours, domain names, and even bits of copy.
Okay, time to retire the .xyz domain. What is the better replacement for @LeaveMeAloneApp ?17:12 PM - 04 Dec 2018
3
4
So I guess the next step was to actually build something?
Chapter 3 - Coding
Coding itself is the piece of the product puzzle that we are most familiar with, and as we're beginning to understand, it's also arguably the least important part.
Nonetheless it had to be done. We whipped together our ideas and in a few days had a working prototype. Leave Me Alone scanned a users Gmail inbox for marketing emails, and allowed them to unsubscribe from them with a single click.
We're pleased to say that we managed this without exposing a single piece of email content on our servers. Thus achieving the privacy aspect that was essential to us. We still store a little meta data from any emails that are unsubscribed (so that we know they have already been actioned if we see them again), but this information is encrypted and so we think this is a small concession.
(Though if anyone can think of a way for us to have the same functionality without storing even this little bit of data then let us know!)
Anyway, with something now working it was time to let our beta testers run free!
Chapter 4 - Closed beta
Beta testing is a scary prospect, letting users into your app before it's quite ready? What if things break?
Well it turns out that things do break, and sometimes they break hard. But beta users aren't expecting a finished product, and they are surprisingly forgiving! In our case we swapped free use of our beta product in return to listening intently to absolutely everything they had to say about the app.
Some stats from the @LeaveMeAloneApp closed beta after only one day!
👩💻 20 active beta users
👀 50 scans
💌 10,000 emails scanned 😱😱😱
🙅♀️ 150 emails unsubscribed
Huge thanks to our testers! ❤️ You've really given the system a battering over the last 24 hours!20:01 PM - 30 Nov 2018
As a result we found and fixed a LOT of bugs, tweaked the UI, and came up with some new features that we hadn't thought of that are now essential to the app.
We're of the opinion now that without letting users loose as soon as possible, Leave Me Alone wouldn't be half as effective as it is now.
Chapter 5- Open Beta
Since the beta was so successful, we are now opening Leave Me Alone up to anyone. A soft launch if you like. This will help us to spread the word, as happy users make for the best salespeople!
Epilogue
We approached making Leave Me Alone a lot differently from our previous apps. We wanted to build in the open and share our progress with other makers and potential customers. We shared our progress on Twitter, logged our tasks on Makerlog, and asked for feedback in Slack and Telegram communities that we're active in.
This turned out to be an excellent decision for two reasons;
We were hostage to accountability. We were openly responsible for building and delivering an idea that we had committed to make. Immediately, enquires started about promised features, progress updates, and the launch date. This helped by validating that the product was needed, and gave us motivation to keep developing.
Having a community of users that want to see our product succeed was invaluable. We discovered isolated bugs that would have been insanely difficult to track down without being able to work one-on-one with dedicated users. For instance one user has a single malformed email that caused the entire app to crash if it was seen. One other had the biggest inbox of spam I have ever seen, which quickly smashed our email streamer 😂. Overall, we had a busy 72 hours handling bug reports, pouring through logs, and pushing fixes. Without this process we would never have known about any of these problems and we would have launched a very broken product.
Final thoughts
We want to say a tremendously huge 🎉 THANK YOU 🎉 to all of our beta testers for their help. Leave Me Alone is now ready to be released into the wild. We'll continue to develop and improve the service. You can see our public roadmap here - let us know which features you'd like us to work on next.
If you've made it this far and you'd like to try Leave Me Alone you can head on over to https://leavemealone.xyz and use the coupon WE_HATE_SPAM for 10% off any scans!
We hope you find our email service useful and that it can help you. If you have any questions or feedback don't hesitate to contact us on Twitter or via email at leavemealone@squarecat.io.
Enjoy your clean inbox!
Cover photo by Randy Fath on Unsplash
Top comments (22)
Great to see the process, and all the best to y'all!
One potential problem I see down the road: design by committee. Even if the committee is your user base, that can still be a problem. Similarly, how do you handle feedback overload? How will you handle making decisions as your user-base grows, where you cannot factor in everyone's feedback? How do you maintain a centralized vision, instead of something designed by five hundred different people?
(Historical context: Minecraft was developed in the open by Notch, and he incorporated a lot of people's ideas before starting Mojang. Now, they get flooded with feedback, and have to ignore most of it.)
Hey Jason! Thanks for a great comment, it made me think!
Right now we haven't yet reached feedback overload. We'll keep asking for feedback from the maker communities which will remain moderately small and keep questions on social media to something easier to manage like polls. I also anticipate more A/B testing with some marketing to drive people to the website.
I'm not sure if we want a completely centralised vision. We knew the features the product needed to get off the ground, but now we're not the ones driving the future of it - our customers are. So we've created a public roadmap where people can suggest and vote for the next feature.
We've built several apps in the past which have not been a success because we built what we thought users wanted instead of asking them. For Leave Me Alone we want the users to be the driving force behind the app. And I may regret it but I hope we grow enough to be flooded with feedback!
Thank you again for the comment :)
This comment made me think there must be a better way to allow people to give feedback...
So I've just released a new version with better way for people to give feedback leavemealone.xyz/feedback and report a bug leavemealone.xyz/bugs - they're both Airtable forms!
What do you think?
Hi Danielle, this post is great as is your approach. A public roadmap and maybe a direct line to the beta testers seems like a good way to steer the product, even when you'll grow so much you won't possibly be able to answer to everyone.
Good luck and thanks for sharing!
Thank you! Haha I can only hope we'll grow too much :P
Yesterday I added a way to provide feedback and report bugs directly from our website at /feedback and /bugs
These go into an Airtable spreadsheet for sorting out (feature requests to the roadmap, and issues onto our bug tracker). Hopefully this method will keep it easy to manage as we grow!
Absolutely fabulous look into your process. Now it's time to put in the long hard work of refining and refining and refining.
Great to have you in the community Danielle.
Thank you Ben! Absolutely, the grind starts now haha
Why on gmail? Why don’t you develop a mail server extension service instead , dependency on Gmail is bad. The way Gmail is now controlling third party access to mailbox is really concerning.
Gmail is the mail provider that myself and James use and we were testing and solving the issue for ourselves first. Google have recently made it more difficult to get API usage approved; we had to record videos of the sign in process and justify why we wanted the read email scope etc which I think is a good thing :)
Is it good thing? Gmail will completely control your users !! and your business in a way, how is it good thing? I would not want to show google how software works with user's emails, it is not Google's property, emails belong to users.
I'm not sure I understand what you're saying. Our users already use Gmail, we just built a service to help them clean their inboxes. We'll support other providers eventually :)
This is a really great article Danielle! Quick question, do you always approach your idea validations with landing page websites or was it something used just for this application?
We didn't used to but we would spend time building something that no one wanted! Now we try to validate the idea first by building a landing page, getting people to join an early access mailing list, and tweeting about it :)
Am asking myself !! could this tool remove my dev.to subscription ?
Haha yes it could! However, one of the features we'll be working on next year is a "keep list" of subscriptions you don't want to unsubscribe from. We have a public roadmap where you can vote for what we work on next leavemealone.xyz/roadmap - the keep list is currently winning!
Nice. Is it possible to join your team ?
It's just my partner and myself at our company right now, we don't have the resources for any more right now, sorry!
Nice approach !! it make me think of slack's launching approach : OPENED , community oriented. Thank for sharing this amazing story !!!
Thank you so much! I'm so glad you enjoyed reading 😊
Fabulous story
Thank you for reading Liana!
Thank you Jag! That's what we were hoping for :) - we've built too many things in the past that were code first, market later and they did not work out so well!