Method 11 of 100 · Stage 2: Check demand before you build

Google Trends Analysis

Compare search interest over time and by region to see whether a problem is growing, steady, seasonal or fading.

Google Trends analysis means using Google’s free Trends tool to see how interest in a search term changes over time, across regions and against other terms. For a founder it answers one narrow question very well: do more people, fewer people or the same share of people search for this problem than a few years ago?

The scale runs from 0 to 100, and here is why that matters. Google divides each data point by all searches in that place and period, then scales the result, so 100 is the peak share for that line in that chart and 50 is half of it. Terms with very low volume show as 0. Trends shows direction and shape and says NOTHING about how many people searched.

When should you use it?

Use it when:

  • You have a problem or product category in mind and want to know whether interest is rising, flat, falling or tied to a season.
  • You sell to consumers or small business owners who google their problems: health, parenting, home, hobbies, local services, ecommerce products.
  • You are torn between two phrases for the same thing (“meal prep” or “batch cooking”) and want the words customers actually use.

Skip it, or cap it at 20 minutes, when:

  • You sell to a narrow B2B niche. “Freight audit software for cold storage” will read as 0 and teach you nothing.
  • Your category does not exist yet. People can’t search for a thing they can’t name, so look up the problem they have today.

How do you do it, step by step?

  1. Write 5 to 10 search terms. Mix problem phrases (“back pain after pregnancy”), solution phrases (“postpartum physio”) and the words your competitors use in their headlines.

  2. Run each term on its own first. Go to trends.google.com, enter the term, set the location to the country you will sell in and the period to “Past 5 years”. Five years shows whether a rise is real or a one-off spike. A line that sits on 0 with random jumps has too little volume to read, so broaden it or drop it.

  3. Choose between term and topic on purpose. As you type, Trends offers the search term and, for many words, a topic. A term matches your exact words in one language. A topic groups searches that share a concept in any language, so the “London” topic also catches “Londres”. Use the topic for overall interest and the term for wording.

  4. Compare terms in one chart. Add more terms with “Compare”. All lines now share one scale, so you see which phrase wins. Always add one benchmark you understand, say “pilates” if you sell fitness, so the other lines have a reference point.

  5. Read the shape, then the season. Look for a repeating pattern across the 5 years: a peak every January, a dip every August. Write down the peak and low months. A seasonal market is fine, but a test you run in a low month will look worse than reality, so plan around it.

  6. Check the regions. Scroll to interest by subregion. If you plan to start local, see where your city or region ranks. These are shares of local searches, so a small region can score 100 with few actual searches.

  7. Mine the related queries. “Top” lists terms most searched with yours in the same session. “Rising” lists the terms with the biggest growth in your period, as a percentage, and Breakout means growth above 5,000%. Copy every rising query that describes a problem into your sheet as a fresh angle for interviews.

  8. Put a real number on the winners. Trends gives no volume. Take your best 3 terms to Google keyword validation for monthly search ranges before you decide anything.

What does a worked example look like?

Say Ines is a physiotherapist in Lisbon who wants to sell an online pelvic floor course for new mothers in the UK and Portugal. The numbers below are invented to show the reading process.

She sets the UK and “Past 5 years” and compares “pelvic floor exercises”, “kegel exercises”, “postpartum workout”, “diastasis recti” and her benchmark “pilates”. Say the chart shows “pelvic floor exercises” steady around 45, “kegel exercises” sliding from 30 to 20, and “diastasis recti” climbing from 10 to 25. Two lessons: British women say “pelvic floor” more than “kegel”, and a related problem is gaining attention.

She opens Rising for “diastasis recti” and, in this example, finds “diastasis recti exercises after c section” marked Breakout. That becomes a possible module and an interview question.

Then she switches to Portugal and Portuguese phrases. Most lines sit at 0. Her call: start in the UK with English content, park Portugal, and check the 3 strongest UK phrases in Keyword Planner for real volume.

Which free tools help?

Tool What you use it for Free plan
Google Trends Relative interest over time, regions, rising and top queries Free, works without signing in
Google Keyword Planner Monthly search volume for the terms Trends picked Free inside a Google Ads account
Google Sheets Log terms, peak months, rising queries and your decision Free with a Google account
AnswerThePublic Turn a winning term into the questions people ask Free searches with a daily cap

What mistakes should you avoid?

  • Reading 100 as “a lot”. 100 is the top of that line in that chart. A niche term can hit 100 with a few hundred searches.
  • Comparing screenshots from different charts. Every chart rescales. Compare terms only inside one chart with the same region and period.
  • Judging on a 30-day window. A news story or a viral video creates a spike. Use 5 years for direction and 12 months for the season.
  • Searching for your solution’s name. A new category reads as 0 and scares you off a real problem. Search the problem.
  • Treating a rising line as demand. Rising interest shows curiosity. Payment intent comes from a landing page test or a pre-sale.

How do you know it worked?

You are done when you can write three lines in your sheet: the phrase customers use most, whether interest over 5 years is rising, flat or falling, and which months peak.

Our rule of thumb for a go: your main problem term averages above 20 across 5 years in your launch country, shows no steady decline, and Keyword Planner confirms a few thousand monthly searches across your best terms. No-go for search as a channel: every term hugs 0, or the 5-year line falls year after year. The idea can still work in that case, though you will need proof from conversations and direct outreach instead of search.

What should you do next?

Take your strongest terms to Google keyword validation for volume, then collect the exact questions people type with Google autocomplete research. If the problem looks real, hear it in people’s own words through problem interviews. The F/MS startup game walks you through these validation steps with an AI co-founder if you want a second opinion on what your charts mean.

FAQ

No. It shows each term’s share of all searches in a place and period, scaled from 0 to 100. For monthly volume use Google Keyword Planner, and use Trends for direction, season and wording.

Terms with very low search volume appear as 0. Broaden the phrase, switch to the topic, widen the region or pick a longer period. If it stays at 0, search is a weak signal for this idea.

For web search, the period menu reaches back to 2004. For a startup decision the last 5 years matter most, because older patterns reflect how people searched before phones and apps changed habits.

Updated 26 October 2026 · Written by the Fe/male Switch team

Methods that go well with this one

Method 12

Google Keyword Validation

Use Google Keyword Planner to see how many people search for your problem each month and whether advertisers pay to reach them.

Validate demandBeginner2 to 3 hours
Method 14

Google Autocomplete and People Also Ask

Type your problem into Google, letter by letter, and log what it suggests plus the People Also Ask questions: free, fast demand clues.

Validate demandBeginner45 minutes to 1 hour
Method 13

AnswerThePublic Research

Turn one seed word into hundreds of real search questions, then group them into problem clusters and pick the strongest one.

Validate demandBeginner2 hours
Method 1

Reddit Problem Research

Search subreddits for repeated complaints and wishes to find problems people already talk about, in their own words.

Find the problemBeginner3 to 4 hours
Method 31

Landing Page Test

Build one page that sells the product as if it exists, send real visitors to it and count how many leave an email or pre-order.

No-code prototypeBeginner1 to 2 weeks (about 6 hours of work)

Go deeper