My Product Failed, Rewritten
Hey folks, Joshua Harms here. I've recently merged all of my blogs into this one, but I've had a few popular posts in the past that I wanted to preserve and move over here.
This is one of those posts; it was by far my most popular and had over 200k views from Hacker News alone. While I do want to preserve the post, I've also decided to rewrite it for a couple of reasons:
- The post rambled a bit too much and came off as "I wish I could build a SaaS business, poor me", which I did not intend at all.
- There was confusion, anger, hysteria, and a general hubbub over my decision to call the product "failed". I didn't explain that I considered it failed because I was never interested in what it did and who it was for, and because I had no intention to work on it again.
That said, the majority of the post remains intact. If you're interested in a reasonably objective (or so I've been told) post-mortem of a "failed" SaaS product, read on.
My product failed
Sometime in the fall of 2012, I started getting into the Software-as-a-Service scene. I was listening to and reading advice from several established SaaS entrepreneurs such as Patrick McKenzie, Brennan Dunn and Rob Walling.
A few months later I felt like it was my turn to take a stab at building a successful software business. My product, called "Rakasheets" (which was poorly named at best), was a realtime inventory management solution for small businesses.
I came up with that solution by trying to follow the gospel as preached by most SaaS entrepreneurs: find a group of people with a real problem, that I could solve, and that they would pay money for.
So I scoured the depths of the internet looking for problems, and eventually I started gravitating toward the Small Business subreddit. Many of the users there were complaining about inventory management, or looking for a new inventory system.
By that time I had built a few custom inventory solutions for local businesses, so I quickly latched onto the problem and I started to ignore all of the other warning signs.
Although pricing was never my problem, I should have considered the most common complaints and requests about inventory management: "Why does it cost so much?" and "Where can I get it for free?".
Four months later, at the end of May and after hundreds of hours put into building the product and trying to market it, I finally decided to call it quits. I had one total customer paying $29/month, and less than 20 trial signups.
Even worse: that single paying customer only joined because I built two big features specifically for him. (More on that below).
Looking back, there are three big reasons why I didn't sell my inventory solution and eventually gave up:
1: I built the entire thing in secret.
Common startup convention says that you should at least put out a "coming soon" page and start directing as much traffic toward it as you can. It helps to not only build up an audience for your launch, but to also validate that you're solving a real problem that people will pay for.
I did the exact opposite of this: I built every single feature before I 'launched' it. It took around four weeks to get the entire thing ready, and the whole time I built it without knowing that somebody out there was going to buy it.
Why did I do that? Fear. The entire time I was building it, I kept telling myself that if I announced a product that wasn't ready, people would dismiss it as amateur and tell their friends not to buy it (or something crazy like that).
If I had been building an audience and talking to potential customers, I would have figured out that what I was building just wasn't what the customers wanted.
2: I built two custom features for one customer.
After the product was finally ready, I started posting on the Small Business subreddit. Each time I saw somebody talking about inventory management, I'd pop in to give them some advice and recommend my product.
A couple of weeks went by with no interest in my app (warning! warning!) until finally one person got in touch with me. After a couple of emails, he let me know that he'd probably buy a subscription if it had two features that he needed:
- His inventory had expiration dates, so it needed to have support for detecting expired inventory and alerting him via email or text message.
- He needed to scan barcodes and generate a report on whichever item he scanned.
Now, don't get me wrong: those are useful features for any inventory application. The problem is I spent several weeks trying to get them perfect for a single potential customer who would probably buy a subscription.
I had even toyed with building an entire Android app just to make the barcode scanning easier. Luckily he was able to purchase a USB barcode scanner, with the promise that an Android app would come eventually.
Thankfully, after the two features were pushed live, this person actually did end up buying a subscription. He got in on the grandfather plan at $29/month, and ended up being a paying customer until I decided to call it quits.
(A paying customer sounds good, but he rarely used the application; I think he was probably paying out of pity.)
3: I was too attached to my vision for the product.
I think this was my biggest problem with my product. Sure, inventory management was useful to them, but only if it did order management, customer management and vendor management too.
That didn't fit my vision though. I wanted the app to be strictly inventory management, and I wouldn't budge when potential customers told me that it needed more.
More importantly, I didn't want to pour more time into a product that I was, by all accounts, burnt out on.
That's where potential customers turned to my competition. No matter how terrible my competitors' interfaces were, potential customers would flock to them because they did what the customers wanted.
That's how my first product, an inventory management solution that didn't really solve any problems, ultimately failed. After nearly six months of time poured into development and marketing, and with only one customer to show for it, I had officially burned out in a big way.
It didn't fail because I didn't get enough customers; it failed because I refused to make it better. It failed because I wasn't interested in what it did and who it was for. It failed because I didn't know how much work it was going to take to build a real SaaS business.
It failed because I was never going to work on it again.