Guarantee VS Free Trial

It’s over 6 months since I started offering 60-day money-back guarantee for FirstOfficer. Unfortunately it’s not working for me and today I’m offering 30-day free trials.

Why I chose to start with a money-back guarantee

I’ve been tempted to try out money-back guarantee since Jason Cohen’s 2013 MicroConf talk Designing the Ideal Bootstrapped Business.

Money-back guarantee felt like a perfect cure for my pre-launch fears:

  • I wouldn’t have to worry about filling my API usage limit by serving people who aren’t yet paying
  • I wouldn’t have to spend a lot of effort to support people who are just ‘kicking the tires’
  • My app would have smaller load and I’d have more time to improve the performance

My app is a good candidate for this model - it starts providing value from day 1 with very little setup.

I also knew that most people don’t actually ask for their money back unless they are really unhappy - so I’d be getting more money even if people wouldn’t convert.

As an extra boost, one of my competitors seems to be happy with that model, so I knew I wasn’t doing a business suicide by starting the same way.

What worked?

In a way, the guarantee did just what I wanted and I’m happy I started with it.

The biggest benefit for me was to gently restrict growth and filter out people who weren’t serious about buying.

Stripe API is feature-rich. For several months, I was worried that new customers would be using some API feature that FirstOfficer wouldn’t be able to handle.

I have a double-entry system similar than in accounting and cross-checkers so I know when the numbers aren’t right - but in worst cases I had to keep the customers waiting for weeks while I was debugging the problems and implementing support.

Some weak moments I felt that I’d never cover the API fully and I was happy to know I did that work for serious prospects.

But there’s no reason to hold back anymore. And still, one size does not fit all when it comes to financial analytics.

I find comfort in numbers - I have enough customers to say no to people whose dashboards would require extensive customization to work properly. The basic customization work is automated.

What didn’t work?

What usually goes wrong with money-back guarantee is that people lose money with it.

However, my bigger problem was that I was losing time - my most valuable resource. Why? Because I ended up giving free trials anyway and then had to manage the whole onboarding manually.

Most of my ‘free trials’ were caused by people signing up too early for financial analytics. If you have few customers or too short history, some of the metrics can’t be calculated reliably. I don’t want to show metrics that can’t be trusted.

So I had two options - let people evaluate the app with some parts not available - or to give them enough time so that they can actually see all the numbers. I decided to give them the time.

In addition to that, some people - especially from larger companies - just asked for a free trial anyway.

It creeped up on me. One day I woke up and my calendar was filled up with notifications to write onboarding emails. I was spending time for something that should be fully automated.

When I sat down to think how to solve this - I decided that I don’t want to support two different onboarding flows.

Why I think there’s also more money in free trials

When I look at my metrics, there are several things that stand out - lousy landing page conversion rate (1.5%) and unexpectedly low refund and churn rates.

I can interpret the numbers in 2 ways:

  1. Customers love my product! Sweet. I’m doing the right thing!

  2. I’m filtering away a large number of potential customers. It can’t be that everyone just loves my product - I’m not converting people optimally.

When something looks too good to be true, it usually is.

If it were only the conversion rates, one could argue if the conversion rate could be brought up with a demo or better landing page. And they might be right - but it’s time  to try and ditch the guarantee. I never intended to keep it forever.

Guarantee is a choice of whether to have money up front or to have more people in the onboarding funnel. Today, I’ll rather have more people.

It’s thrilling to see what the change will do to my numbers.

What types of SaaSes might benefit from the guarantees?

Even though I’m not continuing with money-back guarantee, there are businesses that have tried both guarantee and trials and found out that the guarantee makes them more money.

I suspect that the main ingredient for succeeding is the customer commitment and the trouble from switching - like what you see with WPEngine, Jason Cohen’s product.

They do WordPress hosting and moving your WP content over to a new host is a major headache. People don’t do that just for the fun to compare different options and when they pick a host, they don’t change on a whim. I’m a happy WPEngine customer myself.

With SaaS analytics, especially with one-click setup, there’s little commitment.

New apps popping up every day, it’s easy to switch and try different ones. When it’s easy to switch, the situation is different. People may sign up to 3-4 different tools and try them all side-by-side. In a situation like that, free trial is a nice thing to have and I’m glad that I’m now able to offer that.