Lessons Learned Running A Side Project For One Month
Exactly four weeks ago my friend Andreas and I built and launched BugRex.com, a live chat help line for developers.
Our goal is to make coding less frustrating by connecting developers who struggle with experts who’re willing to help them for free.
In this post we’ll share the lessons we’ve learned so far. We’ll also explain why we’re switching the service from paid to free.
The Most Important Lesson
The most important one is actually about how to learn.
You see, sometimes you can feel like having learned something after a thinking about your product or having a discussion with your team. Stuff like weighing pros and cons, drawing logical conclusions and sketching out diagrams on whiteboards.
However, that’s not learning.
Lessons Learned #1: The only way to learn is through execution.
Once you get the product in front of people, reality surprises you in ways you simply could not have foreseen. Many of your fears turn out not to be issues after all, and many of your assumptions turn out to be false.
Note: I’m counting talking to users as a sub category of execution, as that’s clearly also a way of learning.
The Users
At the time of writing this article, BugRex.com has had a total of 4820 unique visitors, with 680 completed chats.
In other words: 11% conversion rate from visitor to chatter — a number we’re quite happy about.
The visitors are from all over the world, though mostly from the US, which isn’t really a surprise.
However, out of all of these people, only 12 converted into actual sales. This means we’re converting around 0.25% of our traffic, which is an order of magnitude lower than what we had hoped for. In other words: quite bad.
Lesson learned #2: Very few developers are willing to pay for coding help. At least in our current format.
Below is a breakdown of the completed sales, divided into technical categories.
All the sales are somewhat related to web development, with a bit more focus on front-end than back-end.
The Marketing
As you can see from the peaks on the graph above, we’ve done two marketing stunts: our launch day push (reddit, Hacker News and Facebook Groups) and when we published our first Medium article (below).
After the peaks though, the traffic quickly fall back to around 30–70 users per day.
Lesson learned #3: Peaks in traffic has nothing to say in the long run. Recurring users have.
Though it’s not visible on the graph above, the Medium article continues to provide traffic longer than HN/reddit/FB. It’s also of a bit higher quality, based on conversion rates and time spent on the page.
In addition to this, the article has resulted in a lot of feedback from people who resonate with our mission, which must have been critical in gaining us over 240 BugRex expert applications so far.
Lesson learned: Share the story behind your product with the world. Especially if you can reach potential users and collaborators that way.
The Experts
As I mentioned in the previous article, it seemed like most BugRex experts actually don’t care much about the money, but rather find a genuine joy in helping other developers.
Out of the 30 experts I interviewed last week, only a handful fell off once I told them about our plans to remove the payment. The rest were still interested in becoming BugRex experts regardless.
Lesson learned #4: Developers actually enjoy to helping others. It’s much more of a of pay-it-forward culture than I expected.
This tendency has been true for our current experts as well, at least up until now. How this will effect their motivation in the long run is yet to be seen, and can only be answered through time.
One thing I’ve noticed though — both before and after we switched to free — is that more experts applicants than I expected tend to fall off in various parts of the funnel.
The funnel goes like this: TypeForm → Email invitation to interview → Interview on Skype → Get access to Olark → Become active expert.
At every step in this process, we loose or disqualify some candidates. So out of 240 expert applications in TypeForm, only 25 have ended up as BugRex experts so far.
Lesson learned #5: Expect a recruitment funnel to look like a sales funnel.
The Pivot
The decision to to pivot from a paid marketplace to a free service is the result of five different factors, with the first two already being mentioned: that people are hesitant to pay (0.25% conversion rate) and that experts seem to want to help anyway.
The three other reasons are the following:
Potential impact
If it actually (against all odds) is possible to create a service where developers help other developers for free, the potential impact will be greater than if it’s limited by payments. We’d rather aim higher and miss, than aim lower and hit.
Motivation
An equally important part is me and Andreas’ motivation. It’s more motivating to see 10 people getting help for free, than one person getting help for 10 dollars. At this stage, capturing the value isn’t as important as creating the value.
Increase number of chats
We’re struggling to grow, and have had days with very few chats. This has multiple negative side effects, with the most important being that experts have left us due to lack of incoming chats. By introducing the free version, the site will hopefully spread easier through word of mouth, resulting in more activity.
Perhaps we’re doing a big mistake here. Though as we’ve learned, the only way to find this out is to execute.
So the next few week will indeed be exciting. And we’ll definitely write about it here on Medium, so be sure to follow the journey!
Thanks for reading! My name is Per, I’m a co-founder of Scrimba — a better way to teach and learn code.
If you’ve read this far, I’d recommend you to check out this demo!
