Some weeks ago I had a idea: an app that shows the best stocks for short or long term investments and for dividends, plus a tracking portfolio with buy/sell alerts.
Then I thought maybe it could interest to the rest of the world or at least to a little group of people and the business model would be a monthly subscription.
So here I’m gonna share how I tested a failed startup idea.
In my previous experiences, I spent some months to build the product first and then discovered no one wanted it or didn’t pay for it. Yes, very frustrating.
I’m sure this can be done better, so this time I wanted to really test a startup idea in a right way. And I felt very inspired by how Buffer started.
So the plan was build a tiny landing page with plans, a signup form in order to measure interest and pay intention, get traffic and see what happens.
What I did
- I got a Bootstrap theme and made a fake screenshot in less than 1h. Yeah, I can make it faster with HTML than Photoshop! 😀
- I built the landing page with pricing, a signup (splitting the
PREMIUMplans with internal variables for texts and different URLs) and a thank you page. Every page was tracked with Google Analytics. I spent about 4h in the nights of a couple of days.
- I love NodeJS but I coded it with PHP (CodeIgniter) in order to deploy in my LAMP multi-hosting account already payed 😉
- Spent 2 hours searching for a cool name and its available domain. Got crazy and finally purchased a domain.
- Create and customize the social media accounts in Twitter and Facebook.
When I was ready to post, share and follow:
- I share on my Facebook and Twitter accounts
- Post on Hacker News with HN Show tag
- Run Google AdSense campaigns
- Run Facebook Ads campaigns
- Create & watch a Google Analytics Custom Dashboard (See below)
- Follow in realtime the Google Analytics
- Made minor changes in the pages, like texts.
For each signup I personally sent an email asking what they loved or liked more and how could I improve the tool for them.
What I got
- Twitter: no replies, no RT.
- Facebook: 3 friends commented, no one engaged.
- Hacker News: a lot of visits, one comment (see above).
- Facebook Ads: See the screenshot
- Google Adsense: just very similar to Facebook Ads.
- And here is the Google Analytics Custom Dashboard:
Metrics: - Visits: 1420
Unique users: 984
Step 1 Conversion (See pricing): 10.99%
Step 2A Conversion (Select Free): 2.21%
Step 2B Conversion (Select Premium): 0.44%
Step 3A Conversion (Done Free Signup): 0.71%
Step 3B Conversion (Done Premium Signup): 0.28% – Really 0%
Traffic by source: - Facebook: 37.4% (inc. ads)
Google: 30.3% (inc. ads)
Hacker News: 17.60%
About the emails I sent when signup: no one replied and about the premium signup was done with a non-existing email. 🙁
- Domain: 18.12 EUR (2 years offer)
- Google AdSense: 32.50 EUR
- Facebook Ads: 33.22 EUR
- Total: 83.84 EUR
- Work Time: About 20h
That time I spent when I could: weekends, some evenings and nights. The campaigns were running about a week.
So that’s what I spent to get about 1,000 users in the landing page and evaluate that the signups are very low.
If I would have at least 1% of premium signups conversion rate and some good feedback, I would continue. But it’s not the case.
I prefer to pay 84 EUR and work about 20h to test an idea than develop during a few months for nothing.
Now I think the idea was not completely right but I truly believe in this testing method.
Bonus: What I would do different now
- Talk with your friends directly before share on social media. Even better: talk to everyone who could be your client or user. I already talked with someone but it was not enough. Never is enough.
- WTF! Buy the domain for 1 year and in a cheap provider.
- And don’t spend too much time for the name and the domain.
- If with about 500 unique users I don’t get good feedback, any sight of traction and already get a 0%, 1% or even 2% conversion rate (set your own), maybe it’s enough to stop everything and don’t continue.
- Don’t create and customize the social media accounts unless you have a cool name, it’s would be available or get traction. Else it’s losing time.
Any suggestion or comment? Leave a comment! Feel free to ask