What Is a Boomerang Experiment? Definition
A boomerang experiment has target customers use a competitor's or existing product while you observe and interview them. Their frustrations and unmet needs with that incumbent reveal the opportunity for your solution, so you validate real demand before building anything of your own.
Quick Answer: A boomerang is a discovery experiment from Testing Business Ideas by David Bland and Alex Osterwalder. You get target customers to complete a task using a rival's existing product, then watch and question them. The pain points they hit with the incumbent become the evidence for what your product should do differently.
How a Boomerang Experiment Works: Study Customers on a Rival's Product
A boomerang tests your understanding of the problem by observing real behavior with a real product. Instead of pitching your idea or building a prototype, you hand customers something that already exists, an incumbent tool or a competitor's product, and let them use it for a genuine task.
In Testing Business Ideas, Bland and Osterwalder group experiments by what they help you learn. The boomerang sits in the discovery stage: its job is to surface unmet needs and confirm that a problem is real and painful, not to prove your specific solution works. You are borrowing a competitor's product as a research instrument.
The mechanics are simple and cheap. A typical run looks like this:
- Recruit target customers who genuinely have the job or problem you want to serve.
- Ask them to complete a real task using the existing or competitor product, not a mockup.
- Observe closely where they hesitate, work around limits, or express frustration.
- Interview immediately after to understand why those moments felt painful and what they wished the product did instead.
The name captures the logic: you send customers out to a rival's product, and their reactions come back to you as insight. Because the product is already built, you spend almost nothing and risk no engineering time. This makes the boomerang a natural early move in any sequence of experiments, a topic covered more broadly in what is a validation experiment.
Watch behavior, not just opinions. The strength of the boomerang is that customers act rather than speculate. What they say they want in the interview matters, but where they actually stumble while using the incumbent is the harder, more trustworthy signal.
A Concrete Boomerang Experiment Example
Consider an illustrative founder exploring a better tool for freelance bookkeepers. The following example is hypothetical and used only to show the mechanics.
Imagine you suspect that existing invoicing software is too clunky for bookkeepers who juggle many small clients. Rather than build your own app on a hunch, you run a boomerang. You recruit five working bookkeepers and ask each to set up a recurring invoice for a mock client inside a popular incumbent tool while you watch over a screen share.
The friction shows up fast. One bookkeeper hunts for three minutes to find where recurring schedules live. Another gives up on bulk-editing tax rates and does it manually. A third complains that switching between client accounts logs her out. In the follow-up interview, each one describes the same underlying job the incumbent handles poorly: managing many tiny clients at once.
You leave with evidence, not a wish. You have not built anything, yet you now know which specific frustrations recur across real users of the current alternative. Those pain points define the opportunity and sharpen your product's focus. Pairing this with a structured competitor analysis playbook turns scattered observations into a clear map of where the incumbent leaves customers underserved.
Boomerang vs. Concierge Test: What Each One Learns
Both are cheap discovery tactics, but they answer different questions. A boomerang studies customers on someone else's finished product; a concierge test has you deliver the value by hand. The table below contrasts them qualitatively.
| Dimension | Boomerang Experiment | Concierge Test |
|---|---|---|
| What you use | A competitor's or existing product | Your own manual, human-delivered service |
| What you observe | Pain points with the incumbent | Whether customers value your delivered outcome |
| What it validates | The problem is real and underserved | Willingness to accept and pay for your solution |
| Who does the work | The customer, on a rival's tool | You, by hand, for the customer |
| Best question it answers | "Where does the current alternative fail?" | "Will customers use what I would actually offer?" |
Takeaway: Reach for a boomerang when you still need to prove the problem and locate the gap; reach for a concierge MVP once you want to test whether your own delivered value lands. Many founders run the boomerang first, then use its findings to shape the concierge offer.
When a Boomerang Is the Cheapest Way to Learn
A boomerang shines when a credible alternative already exists and you are unsure whether its weaknesses are real openings. Because the product is already built, you get grounded behavioral evidence for almost no cost or engineering effort.
It fits best in these situations:
- A crowded or established market. Incumbents exist, so customers can actually use something today. Their frustrations point to the wedge for a new entrant.
- You are early and problem-focused. You need to confirm the pain is real before committing to a build, exactly the discovery job Bland and Osterwalder describe.
- You want behavior over opinion. Watching someone struggle with the current tool beats asking whether they would like a better one.
Know its limits, too. A boomerang tells you where the incumbent falls short; it does not prove that your specific product will win. Customers reacting to an existing tool cannot confirm demand for a solution they have not seen. Treat it as a strong first read on the problem, then follow with experiments that test your own offer directly. Tools like Edmired can help you log these observations as evidence so the decision rests on what customers did.
Key Takeaways
- A boomerang experiment has target customers use a competitor's or existing product while you observe and interview them.
- It is a discovery experiment from Testing Business Ideas by David Bland and Alex Osterwalder, built to validate the problem, not your solution.
- The value comes from real behavior: the frustrations customers hit with the incumbent reveal the unmet needs your product can address.
- It is fast and cheap because the product already exists, so you risk no engineering time to gather grounded evidence.
- It differs from a concierge test, which delivers your own value by hand to test willingness to use and pay.
- Use it early in crowded markets where credible alternatives already exist and you need to locate the gap.
- Respect its limit: it proves where the incumbent fails, not that your specific product will win.
Frequently Asked Questions
Is a boomerang experiment the same as a competitor analysis?
Not quite. Competitor analysis usually studies rivals from the outside, their features, pricing, and positioning. A boomerang goes further by putting real target customers in front of a competitor's product and observing them use it. The evidence comes from watching behavior and frustration firsthand, not from reviewing the competitor's marketing or feature list.
What does a boomerang experiment actually validate?
It validates that a problem is real, painful, and underserved by the current alternative. By observing where customers struggle with an existing product, you confirm the opportunity exists. It does not validate that your specific solution will succeed, since customers are reacting to the incumbent, not to anything you have built.
How many customers do I need for a boomerang?
There is no fixed number, but a small group of genuine target customers is usually enough to see patterns. When several people hit the same friction on the same task, that repetition is your signal. Bland and Osterwalder frame discovery experiments as fast and lightweight, so favor speed and real users over large samples.
When should I run a boomerang instead of building a prototype?
Run a boomerang first when a credible competitor or existing product already exists and you are still unsure the problem is worth solving. It costs almost nothing and produces behavioral evidence before you write code. Once the boomerang confirms the gap, a prototype or concierge test is the natural next step to validate your own solution.