Broken Lightbulbs and Dead Ideas

I've read lots of stories about entrepreneurs validating their ideas and going on to launch successful companies. But what about the ideas that don't pan out?

I want to share the story behind the recent invalidation of an idea I'd been working on.

I was excited to start something new, so I moved forward with the idea, tested it, and finally decided that it wasn't going to work out. Now I'm free to move on to other ideas without any lingering thoughts of, "Could this have worked?"

It feels better than you might expect. You should try it. ;-)

Finding an Idea

I'm an entrepreneur who's been building and running a digital hiking guide called enziano for almost five years now.

Before founding my first company I worked as a developer and usability engineer in Germany, so drafting, designing, and building a new product from scratch is what I do best.

In the interest of focusing on one thing and one thing only, I've always tried to ignore new product and business ideas. But, as it turns out, I can't permanently shut down the idea monkey in my mind.

This year I set aside some time to try out some new business ideas that require the minimum amount of effort possible. My goal is to validate and build a self-sufficient business in the digital space — one that doesn't depend on my personal presence and schedule, but can be run alongside my daily life.

A few weeks ago, I stumbled upon Zack Burt's monthly reports on his business, Code For Cash, on the Indie Hackers forum. In exchange for a monthly fee, Code For Cash helps developers get freelance work by sending them relevant project/job offers via Slack.

It was great to read about a working concept, and my first thought was, "Hey, if this works well for developers, it might also work for freelance designers who are looking for projects."

Feeling Things Out

The first thing I did was reach out to Zack directly to get a feeling of whether he was open to some kind of cooperation.

Given my goals, I wasn't eager to build and maintain yet another complex software system for a side project.

I was hoping for a white-label or joint-venture cooperation where he would maintain the software system and I could use it to market and serve the UX and designer vertical.

Of course, I learned that Zack already thought about expanding to other verticals where he could apply his project-scraping technology. He suggested that it would be no problem to adapt the software for design jobs, too, and that we could drive traffic to a niche site especially targeted at UX designers.

That response was more than encouraging enough for me to decide to move forward.

Reaching Out to Potential Customers

My next goal was to talk to as many designers as possible in order to find out what their #1 challenge is when looking for new design work.

Why? Because only targeting the #1 "hair on fire" problem in a given market is one of the best practices advocated by many people I respect, including Pat Flynn of Will It Fly?, Eric Ries of The Lean Startup, Patrick McKenzie of Kalzumeus, Dan Norris of The 7 Day Startup, The Foundation, etc.

My hypothesis at this point went something like this:

  • A freelance designer is looking for new remote working projects frequently.
  • She doesn't enjoy wasting time searching through dozens of listings only to discover that remote is not an option, the budget is too low, the required skills don't match up, or other deal breakers that are hidden within the job descriptions.
  • Since good project leads mean more revenue, she's willing to pay a monthly fee to get those leads.

I started by posting in two Facebook groups with about 20,000 designers each. I also posted in a few designer forums and newsgroups. Finally, I directly reached out to 90 designers who I found via their public profile pages on Dribbble, Behance, and XING — the German version of LinkedIn.

I didn't want to spend a lot of time copy-and-pasting, so I used a tool called Mailshake to send my emails. It also helped me track who'd already responded, and it sent an automatic follow-up email after 5 days.

Cold email sent to a designer to validate their #1 challenge

I only spent an hour or so writing my email template and setting up Mailshake.

From my background as a usability engineer, I know to avoid asking customers leading questions that artifically get them to agree with your desired solution.

Instead of pitching my idea directly, I tried to carefully assess whether each designer's #1 problem naturally aligned with my hypothesis.

I also ran a split test by sending out two different versions of my email. The version you see above contained a section explanining why I was asking for this information, and the other did not. Both had the subject, "Your advice?"

Setting Up a Landing Page

While I was waiting for the designers to reply, I decided to set up a simple landing page.

Although I didn't have a working product to sell at this point, I figured a website could still serve as a smoking-and-mirrors test of whether or not people were willing to purchase the offer based on the description alone.

I wanted to get the page up with as little investment as possible, so I just piggybacked off of another WordPress website that I already had online.

In total, it only took me about 6 hours to outline the content and set up the site.

I used copy similar to what's been working for Code For Cash, but included some extra information below the fold based on my guess that the designers I was targeting would mostly be looking for remote projects.

Landing Page for Smoke-and-Mirrors Test

In addition to this landing page, I also added a payment page and a "more info" page.

To get a better sense of what people thought, I set up Inspectlet, a recording tool that gives you insight into what visitors are doing, reading, and clicking on your site. I was specifcially interested in seeing how far they scrolled down the landing page, and the text and images at which they either exited the site or clicked through to purchase.

Now that the site was up and running, all that was left was to drive traffic to it.

I decided to run a small Facebook ad campaign targeted at designers. In total, the ad cost me $23 (USD) and brought in 229 visitors, meaning I was only paying 10 cents per click!

Later during the campaign I noticed that I'd set up ad to target a worldwide audience. As a result, I'd been getting lots of clicks from Asia, which might be why the cost-per-click rate was so cheap. Oops.

It was my first time running a Facebook ad campaign, so I chalked it up to being an amateur and moved on.

The Results of My Testing and Outreach

The two forum posts I made completely failed. In one case I got zero responses, and in the other a discussion broke out about how low-rate $5 designers from Fiverr ruin rates for everyone else. I wasn't able to salvage the discussion and find any relevant pain points there.

The Facebook groups fared a little better. One of my posts picked up some traction, and a few designers left comments describing their biggest challenges.

As for the website I set up, the end result was that around 20% of visitors to the primary landing page clicked through to the payment page. Not bad! Unfortunately, nobody who got that far ended up making a purchase.

My guess is that lots of people clicked through to the payment page without reading the info on the landing page. The screen recordings from Inspectlet seem to confirm this.

Since the payment page contained almost no information at all, visitors quickly left.

The next time around I'll add some reinforcing information and testimonials on the payments page, too.

Finally, the 90 cold outreach emails I sent got a whopping 84% open rate. Unfortunately, I only got 8 replies in total, even after following up with everyone.

The email containing explanation of why I was asking resulted in a 17% response rate, while the version without the explanation only got a 5% response rate. The lessons I learned here were to always split test email copy, and to follow up at least once or twice if you don't get an answer.

Analyzing My Potential Customers' Problems

After having all of these conversations, I compiled a list of the designers' pains points and my takeaways.

For most of the freelance designers I talked to, it turned out that finding new leads and clients was absolutely no problem.

They tended to get project offers from former clients who've been happy with their work in the past, and from new people who have been referred by a former client.

Lesson learned: Word-of-mouth is king, and designers should proactively facilitate that by asking for referrals, staying in contact with "passive" clients, and just generally having a system for this.

My other takeaway was that most freelance designers avoid bidding for projects on online platforms like Fiverr, 99design, and Upwork, since they consider it bottom-feeding — they don't like to compete on $5 projects with hourly rates from overseas designers.

Some, however, have found success using those platforms as lead-gen tools after taking the time to filter through everything to find the "good projects".

So what were their challenges? Here's what I recorded in my notes:

  • filtering through projects to find the ones that match their skills and expectations
  • finding the right kinds of clients with whom they can have healthy working relationships, so they can avoid stress and facilitate more referrals in the future
  • juggling multiple projects and their associated timelines and deadlines
  • getting their collaborators to do the necessary work on time, provide input, and work within deadlines so as to avoid delays and last-minute requests
  • balancing sales and business tasks with design work and self-improvement


Based on all of the above, I decided that my idea is probably not worth pursuing any further, at least not in its current form.

I wanted to deliver releveant projects to designers via chat in return for a monthly fee. However, word-of-mouth referrals were already working well for most designers, and none of the visitors to my smoke-and-mirrors test converted to paying customers.

One could argue that I didn't talk to enough designers, or that I talked to the wrong ones, or that I didn't drive enough traffic to my website.

That might be true. Who knows?

However, I hope that you've gleaned some valuable or interesting insights from my "invalidation process," because I certainly did!

I had fun working on the idea, connected with my fellow designers, and learned a lot about Facebook ads, cold email outreach, and the daily struggles of freelance designers.

If I let things sink in for a while, it's possible that some new angle or insight will occur, and I will give it another shot. But for now I'll move on to other ideas and attempt to find an actual problem that's worth solving as a side project.

Feel free to share your thoughts on the process and the results! Did I quit too early? What else could I have tried? I'm curious to hear what you think! ;-)