Well, after a really disappointing launch of the premium plan for Markfolder.com, I finally got my first paying customer!
It's on the annual plan, so we're on $30 ARR.
While this is great, clarity on product-market fit is still elusive. So there's still a lot of work to do.
But at least now we know at least one type of user that IS paying for it. I need to find more of those users.
Markfolder has been a free app thus far.
But after 500 signups and seeing real validation through customer feedback and traction, it's time to start releasing premium features.
Free users still get unlimited bookmarks and folders, but they only get basic features. Pro users get more advanced features, especially on the bookmark organisation and consumption side.
Excited to see how this next phase develops!
It's small potatoes compared to most SaaS apps, but today Markfolder surpassed 50,000 saved tweets by its users.
This is with zero marketing and promotion campaigns. Most traffic is from organic search and word of mouth.
The first PRO feature for my Twitter bookmarking SaaS, Markfolder.com, is done!
Markfolder has been a free SaaS, but soon I'll be launching Markfolder PRO. The first PRO feature is Search. Been working on this last few days and I think it's now ready.
In building Search, I looked at various 3rd party options instead of rolling my own, so that I can launch it quicker. Algolia is supported by Laravel (the dev framework I use), and I'm familiar with Algolia. So this looked promising, but, it would have cost a lot of money. Forget it.
So I looked at what MySQL (the database running Markfolder) can do in terms of search. Turns out, quite a bit! MySQL's FULLTEXT search is good enough for this first version and it won't take too much work.
Getting the search keywords formatted correctly took the most time. And MYSQL's default minimum 4 letters for search keywords is a little annoying.
But it's good enough. If this becomes a popular feature and people are paying for it, then I can always upgrade it to a more sophisticated search service (like Algolia).
Next: Create a teaser for free users to entice them to upgrade.
Bookmarking is an inherently private activity, so I wanted to see if a bookmarking tool can, in fact, have some sort of virality built in. So I'm experimenting with public folders as a way to bring in traffic. The assumption is that people who love to curate will use it to share their curated tweets, which in turn will bring traffic to markfolder.com. We'll see!
This week's stats for markfolder.com, Twitter bookmarking tool.
The % compared to prev week are in brackets.
New users: 8 (+100%)
Tweets saved: 148 (+159%)
Folders created: 23 (+228%)
Active users: 12 (+45%)
The product is still on pre-release version. I haven't really worked on acquisition. Most of the traffic is from me engaging with people on Twitter.
Markfolder was released as an early access version just over a month ago. Its goal is to help Twitter users organise tweets they want to save/bookmark, by giving them the ability to group tweets with folders.
Still in its "early access" version, users have already saved over 500 tweets.
Even though there are now 30 signed up users, the bulk of the bookmarks have been saved by just 3 "power users". These guys are prolific in the bookmarking!
The next stage is to release a mobile version of Markfolder. Since 80% of Twitter users are on mobile, this is sorely needed.
One of the biggest issues (and complaints) with Twitter's bookmarking feature is that it's like a black-hole - you save tweets into it and never actually get back to looking at it again (with all the intention in the world to do so!).
So an early goal for Markfolder is to help you surface your saved tweets regularly so you don't forget about them.
This is as easy as just showing you what you've saved, once a week, by email.
And this is exactly what I shipped today!
Lots of technical lessons learned on how to encourage repeat use of your product. For example, if a user did not save anything in the last week, then what can we show them instead? For this, I chose to just surface one random bookmark of theirs, rather than saying "you didn't bookmark anything last week, sorry". Hopefully this will help pull them back into engaging with Markfolder.
But for those that have yet to bookmark anything, then it becomes a different story - how do we get them to save their first bookmark.
Need to think about this.
Thanks to a happy early user of Markfolder who recommended it to someone else on Twitter, Markfolder got its first user from word of mouth.
This is great because I believe on order for a product to spread by word of mouth, it not only needs to be useful, but also pleasant to use. I wouldn't want to recommend a clunky product to someone even if it's useful!
One important item I wanted to tick off as early as possible is backups.
Markfolder's users have entrusted me with their data and I want that to be fully respected. Making sure they never lose their data is the least I can do (followed by security of their data and their privacy, which I'll continually work on).
So now, at 00:00 every night London time, a backup of all code, files and database is made, zipped, and saved to an Amazon S3 server in the UK. Having the backup on a server separate to the web server ensures that if anything happens to the website, the backup is still safe.
Power Twitter users are underserved by Twitter's bookmarking feature, which lacks folders and search capabilities. Markfolder exists to solve this problem.