How to gauge product demand before you launch

Launching a product without measuring demand is like opening a restaurant without knowing if anyone in the neighbourhood is hungry. Yet most founders build first and hope later. You can flip the order: demand is measured before you build, for free, from the traces people leave when they look for a solution. Here's how to read those traces and turn them into a clear decision.

Demand ≠ interest ≠ market size

Three notions we constantly confuse, leading to three different decisions:

• Market size — every theoretically possible buyer. A flattering figure, but unreachable short-term.
• Demand — the fraction actively looking for a solution now. This is what you can capture fast.
• Interest — “oh that's cool”. Only turns into revenue if there's purchase intent behind it.

You validate a product on demand, not on the pitch's market size. A huge market where nobody actively looks for a solution is a trap.

The 4 free sources to measure demand

1. Google Autocomplete — the real queries

Type the core of your product into Google and read the suggestions: these are real searches, ranked by frequency. Many variants = a rich, segmentable demand field. No suggestion beyond the generic term = a warning. Above all, note suggestions containing a commercial modifier (“price”, “software”, “reviews”, “best”).

2. Google Trends — the trend

Demand can be strong today and dead in a year. Run the keyword over 24 months: old, steady growth = structural demand; a recent vertical spike with no history = a likely fad. The method is detailed in our guide using Google Trends to validate an idea.

3. YouTube — the appetite for learning

Recent videos with lots of views on your topic = an audience that wants to solve this problem and learn. Especially telling for educational, software or “onboarding-heavy” products.

4. Communities — the intensity of the complaint

Forums, groups, Reddit, competitors' 1–3 star reviews: this is where the pain speaks plainly. Phrases like “I waste so much time on…” are worth more than any declarative survey.

Telling real demand from curiosity

The classic trap: mistaking informational traffic for solvent demand. A query “how to do X for free” isn't worth the same as “best software for X”. The first wants to avoid paying, the second already holds its card. Sort the queries captured in step 1 into three buckets:

• Cold — “what is”, “how does it work”. Curiosity.
• Warm — “how to”, “example”, “tutorial”. Active interest.
• Hot — “price”, “best”, “buy”, “alternative to”. Intent.

A product is validated when the “hot” bucket isn't empty.

What demand signal means go

Cross the four sources and file the verdict into one of three boxes:

• GO — volume present, healthy trend, active community, existing hot queries.
• REFINE — the vein exists but a signal is off (fad, or no hot query). Tighten the angle.
• PIVOT — demand missing or purely cold. Don't build.

Example — a meal-planning tool. Autocomplete: very rich (“weekly meal planning”, “menu app”, “batch cooking app”). Trends: steady growth over 3 years, not a fad spike. YouTube: high-volume channels on batch cooking. Communities: recurring complaints about the mental load of groceries. Hot queries: “best meal-planning app”, “price”. Verdict: GO. Still to test willingness to pay with a landing page.

From demand to proof of payment

Measuring demand hugely reduces the risk, but doesn't remove it: the only judge is a customer who pays. Once demand is confirmed, build a pre-sell landing page to turn the signal into real commitment.

The shortcut: measuring product demand in 30 seconds

Crossing these four sources by hand takes a good half hour per product. Sondari does it automatically: it aggregates Google Autocomplete, Google Trends and YouTube, separates demand from intent, and returns a usable demand score in 30 seconds — enough to screen a list of products before testing a single one.

To go further on reading the number, see how to interpret a search volume.