Home
← The Inventor's Notebook
Essay

Top-Down vs Bottom-Up Market Sizing: Which Should You Trust?

A plain-language comparison of two ways to size a market, and why most early ideas need both before anyone believes the number.

Top-Down vs Bottom-Up Market Sizing: Which Should You Trust?

The short answer

Top-down market sizing starts with a broad published figure and narrows it with assumptions, while bottom-up market sizing builds the number up from real counts of customers, purchase frequency, and price. Top-down is faster but easier to overstate; bottom-up is slower but far more defensible, and the two should roughly agree before you trust either one enough to act on it.

What is the difference between top-down and bottom-up market sizing?

Top-down market sizing starts with a big published number, usually from an industry report, and narrows it down with assumptions and percentages until it looks like your slice. Bottom-up market sizing starts with real, countable units, like how many potential customers exist and what they would actually pay, and builds the number up from there.

Top-down is faster and sounds more credible on a slide. Bottom-up is slower but far harder to fake, because every step traces back to something you can defend. Most investors and advisors trust bottom-up more, and most beginners default to top-down because it is easier to produce.

Why does the market sizing method matter this much?

Because the number you land on shapes decisions long before you have revenue. It affects how much you plan to spend on a prototype, what kind of investor conversation makes sense, and whether the idea is even worth the next six months of your time.

A market size that is wrong by a factor of ten is not a rounding error. It is the difference between building a feature for a niche and building a company for a category. Getting this wrong early is one of the quieter reasons promising ideas stall during progressive idea testing, because the founder spent months chasing a market that never existed at the size they assumed.

How does top-down market sizing actually work?

You start with a total addressable market figure, often from a research firm, government data, or an industry association. Say a report states the global market for a product category is $40 billion. You then apply a series of percentages: maybe 15% of that is relevant to your geography, then 8% of that fits your specific niche, then you assume you can capture 2% of that niche in five years.

Each multiplication feels reasonable in isolation. But stack three or four assumptions together and the final number is really a guess dressed up in decimal points. The original $40 billion figure also often includes categories, business models, or customer types that have nothing to do with your actual product.

Top-down is useful for a first gut check and for framing a pitch. It is not useful as proof that demand exists for what you are specifically building.

How does bottom-up market sizing actually work?

Bottom-up starts with units you can count or reasonably estimate: the number of people or businesses who have the problem, how often they would buy, and what they would pay per purchase or subscription.

For example: if there are roughly 120,000 independent bakeries in a country, and your tool would be relevant to about 30% of them, and they would pay $40 a month, that is a specific, checkable path to a number, not a borrowed one. You can go test pieces of that chain directly: survey actual bakery owners, check existing tool pricing, look at how many bakeries show up in business directories.

Bottom-up takes longer because you have to find or estimate real counts instead of lifting one from a report. But it forces you to confront whether your customer actually exists in the numbers you need, which is exactly the kind of question a Product Strategist is trained to push on before you spend money building anything.

Which one should you actually use first?

Use top-down first, briefly, to sanity-check that you are not chasing a market too small to matter. If the entire category is worth $2 million a year globally, that tells you something important fast, even with rough numbers.

Then move to bottom-up before you trust any number enough to act on it. Bottom-up is the method that holds up when someone asks, "where did that number come from?" and you need an answer better than "a report said so."

The two methods should also roughly agree. If your bottom-up estimate of $6 million and your top-down estimate of $900 million are wildly far apart, one of your assumptions is broken, and it is worth finding out which one before you build anything around either figure.

What mistakes do people make with market sizing?

The most common mistake is treating the top-down number as the whole answer and skipping bottom-up entirely, because bottom-up requires more digging and feels less impressive on paper.

A close second is double-counting. Founders sometimes size the market for the problem (everyone who experiences the pain) and the market for the specific solution (everyone who would pay for this particular product) as if they are the same number. They rarely are. Most people with a problem never buy any solution at all.

A third mistake is ignoring willingness to pay entirely and sizing based only on how many people have the problem. Counting people is not the same as counting dollars. This is one of the reasons a market sizing exercise benefits from being checked against actual financial modeling rather than left as a standalone slide, which is where a Financial Analyst perspective earns its place in the process.

How do you build a bottom-up estimate without existing data?

Start narrow. Pick one well-defined customer segment instead of trying to size the whole category at once. Count what you can verify: directory listings, membership numbers, app store category rankings, trade association figures, census or labor statistics for a profession.

Then validate the assumption about who would actually buy, not just who has the problem. This usually means talking to real potential customers or running a small test, which is covered in more depth in guides on how AI tools can help validate product demand before you spend on manufacturing or development.

Finally, be honest about the capture rate. Assuming you will win 20% of a market in year one, with no existing brand, distribution, or product, is the kind of assumption that collapses the whole exercise. Single-digit capture rates in early years are far more defensible.

Does market sizing matter for a side project or a hobby idea too?

Yes, even at small scale. If you are weighing whether to turn a side project into a startup, a rough bottom-up number tells you whether you are looking at a sustainable small business, a modest side income, or something too thin to support the time it would take.

It also helps you decide how much to invest in the next steps. An idea sized at a few thousand potential customers paying a small amount might justify a lean, low-cost build. An idea sized at hundreds of thousands of potential customers might justify a more serious investment in concept drawings or a 3D model before you commit further.

How does this fit into validating an idea overall?

Market sizing is one piece of a bigger validation picture, alongside checking for prior art (existing patents, products, or publications that resemble your idea, a term explained more fully in the IP Education Center), stress-testing the business model, and running basic financial projections.

Treating market size as a single number you calculate once and move on from is a common misstep. It should be revisited as you learn more, the same way a SaaS business model canvas gets stress-tested repeatedly rather than signed off once. If you want a structured way to run an idea through this kind of check, alongside a Product Strategist, an IP Consultant, and a Financial Analyst, the free idea assessment walks through market framing as one part of a fuller picture before anyone talks about a prototype or a patent filing.

Frequently asked questions

Which market sizing method do investors trust more?

Bottom-up, generally, because every step traces back to a countable, checkable assumption rather than a borrowed industry percentage.

Can I use top-down and bottom-up together?

Yes, and it is recommended: use top-down for a quick sanity check, then bottom-up before trusting the number, and compare the two for a rough match.

What is the biggest mistake people make in market sizing?

Treating everyone who has the problem as the market, rather than only those who would realistically pay for a solution.

Do I need market sizing data for a small side project?

A rough bottom-up estimate still helps, since it shows whether the idea supports a sustainable small business or is too thin to justify further investment.

How does market sizing relate to checking for prior art?

They are separate checks within the same validation process: market sizing tests demand, while a prior art search looks at existing patents, products, or publications that resemble the idea.

Sign inSign up free