Fake door tests: know before you build
(Without infuriating your customers)
What if I told you that you can predict how much traction your product is going to get, without developing anything at all. Sounds scammy? It doesn’t have to be. But let me explain.
Fake door testing is a validation technique you can use to assess market interest before investing significant resources into a product’s development. Or, to put it in simple terms, it’s a test to see if you can sell something that doesn't yet exist.
In this article, I’ll look at this tool as a way to find problem-solution fit, and even some signs of product-market fit.
As the name suggests, a fake door test involves creating the illusion of a product or feature, measuring how many potential customers try and “get through the door”, and then deciding if it’s worth your time to build the real thing.
Controversy alert, controversy alert!
As you can imagine, a lot can go wrong when fake door tests are involved. However, these little experiments can be so valuable and save you such an immense amount of time and money, that it’s worth seeing if you can get them right before ruling them out prematurely.
Like most tools, it’s the way you use it that will make all the difference.
Before you get started, let’s look at ways to reduce the risk of PR catastrophes, and see how you can embed fake door tests as part of your validation efforts, in an ethical way.
Think brand
How does the idea of doing something cheeky, but ultimately well-intended, sit with your brand? For some brands, it’s going to be right up their street and so aligned with brand personality that customers may find the whole thing entertaining and bold. If that’s you, make sure you grab this low-hanging fruit with both hands.
Another angle to consider is if you have a brand at all. The lesser known you are, the more freedom to experiment you’ll have - your reputation isn’t at stake, after all. Just don’t take this as a license to do, ehm… shady stuff.
Don’t test too early
Is your product timeline measured in years rather than months? Maybe this isn't the best approach for you. How long is a reasonable time for your customers to wait? If an MVP can be ready in weeks, go for it. But if you're building a complex technology that’s years away from fruition - or worse, if the build duration is still unknown, hit the brakes. Your customers might lose interest (or reach retirement age) before you can deliver.
Use the right call to action
If your buttons and links flash 'Sign up now!' or ‘Get it today’, users will naturally expect immediate access. Mitigate disappointment by tempering expectations of immediacy. CTAs like 'I’m interested' or 'Tell me more’' still help gauge intent without overpromising.
Once you see that your idea has gained some interest, you can even drop every pretense and simply say ‘Join the waiting list’ or ‘Get notified’. Albeit slightly less effective, these can still generate valid results. Besides, if your customers are willing to wait a little for the product you can be more confident that your solution is needed.
Offer an interim solution
The product isn’t ready yet… but can you offer some form of value in the meantime? This could involve manual processes or simplified versions of your offering. For example, let’s say you’re a nutritionist building a meal planning app, you could manually create personalised meal plans while the product is being built.
It's not scalable, but it keeps people engaged and provides valuable insights for your eventual product. Not to mention it could even bring in revenue.
Be honest to rebuild trust
In your experiment environment, people will naturally feel frustrated when they discover that there isn't anything behind that “door”. An eye roll isn't the end of the world, but make sure you manage it effectively and defuse it, rather than escalating it into rage.
Firstly, people’s frustration will be a function of how much time they’ve invested, so refrain from asking them to complete forms or take lengthy quizzes.
Secondly, honesty pays off. For good or for worse, you have the person's attention. So you might as well use it to immediately rebuild trust:
Say that the product isn't ready yet.
Say when will it be.
Say thanks for their patience and highlight why it's worth it (in customer benefits terms).
Offer to onboard them first, or at a discounted price, when the product is ready.
Your copy matters, so choose your words carefully!
Don’t trick people into sharing their contact details
Allow users to opt-in to updates or further communication only after they fully understand the current status of the product. Quality leads are always better than quantity, especially when the quantity is made up of disgruntled individuals you just tricked into giving you their phone number. Not cool.
Keep reading
〰️
Keep reading 〰️
So…how do you build a fake door test?
There are about a hundred different ways you can build a fake door test. The one I’m about to share works best for validating a product idea when you have already identified one or more potential audiences.
Create a landing page
It doesn’t have to look extra fancy but drop the MVP look. Sloppy won’t do, in this type of experiment. I like to use Unbounce.Set up tracking
Choose the behavior you want to track on the page. Heat maps and session recordings (check out HotJar) can be a source of great insight, too. With the latter, expect to generate lots of data and spend significant time analysing it.Choose your traffic source:
If you can, stay clear of your own social media channels unless your test has very low potential for publicity accidents.
Ultimately, you’ll want to get your potential customers where they are, so ensure that both your test audience and your chosen channels align with your target market. Misaligned traffic will skew your results.
Search ads or targeted social media ads are a great way to reach people actively looking for solutions in your space, and (with the former) to do so while maintaining a low profile. Something to be careful about is the quality of your ads. Bad ads = bad results, even if your product may otherwise win some traction.Allocate a modest budget:
£100-200 is often sufficient for initial testing. More, if your product is costly to build. This small investment could save you thousands in misguided product development.Prepare your follow-up:
Create a short email sequence for people joining your waiting list or point them to a place where they can see product updates if you’re building publicly.Analyse your campaigns:
Is there any traction? Where is it coming from? Can you spot any audience and engagement patterns to help you segment your potential market? Are people motivated to pay? These are all great questions to ask when interrogating the data you’re collecting.
And finally, decide. Small experiments alone won’t give you absolute certainty, but they sure help you prove (or disprove) some important hypotheses. Depending on where they take you, map out your findings, choose your next steps, and iterate your way forward.
That’s all folks.
Personally, I love fake door tests.
Yes, they can be tricky to pull off, but all it takes to avoid disasters is a bit of care, consideration, and gratitude for the people taking part in your experiment.
(As well as suppressing your inner Dr. Evil).
(I’m only joking).
Happy validation!
Alice