Google Trends is a free Google tool that shows how search interest in a term changes over time, place, and category. It reports relative popularity on a 0-100 scale rather than absolute search volume, which makes it the fastest way to spot seasonality, compare topics, and catch rising queries early.

Category
SEO
Difficulty
Beginner
Updated
Mar 2026
Read time
5 min

Google Trends is a free Google tool that reports how search interest in a term has changed over time, by location, and by category. It has covered Google search data since 2004.

The critical thing to understand is what the numbers mean. Google Trends does not give you search volume. It gives you relative interest, indexed to a 0-100 scale where 100 is the highest point in the range you selected and every other point is scaled against it. A score of 50 means half the interest of the peak, not 50 searches. Change the date range or the region and every number on the chart changes with it, because the peak it is measured against changes.

That design is a feature, not a limitation. It strips out the growth of the internet itself, so a query in 2011 and the same query today are comparable on the same axis.

Google Trends answers questions that volume tools answer badly or not at all. Volume tools give you a monthly average, which flattens exactly the patterns you need to plan around.

  • Seasonality is visible. A monthly average of 4,000 tells you nothing about whether the demand arrives evenly or lands in a six-week spike. The trend line shows you which.
  • Direction, not just size. A term with modest volume that has doubled in six months is often a better bet than a larger term that peaked three years ago.
  • Relative comparison. You can put up to five terms on one chart and see which one your market actually uses.
  • Geographic breakdown. Interest by region and by city tells a local business whether a topic matters in its own market.
  • It is free and needs no account. For a small business with no tooling budget, it is the one demand-research source with no gate on it.

The trade-off is that you never learn absolute size from Trends alone. Pair it with a volume source and use each for what it is good at.

The core mechanics

Trends takes a sample of Google searches, filters out duplicate searches from the same user and queries with very low volume, then normalises what remains. Each data point is divided by the total searches for its time and place, so the result reflects share of interest rather than raw count. That number is then indexed against the peak in your selected range to produce the 0-100 scale.

Two consequences follow. Low-volume terms return a flat or empty chart because they were filtered out, not because nobody searches them. And any two Trends screenshots are only comparable if they used the same range, region, and category.

The filters that change your answer

Four controls sit above every chart, and each one rewrites the result:

  • Region. Worldwide down to metro or city level. A national trend can be irrelevant in your service area.
  • Time range. Past hour through 2004-present. Short ranges expose noise and news spikes; multi-year ranges expose seasonality and structural decline.
  • Category. Disambiguates a term. "Jaguar" filtered to Autos is a different query from "Jaguar" filtered to Pets and Animals.
  • Search type. Web, Image, News, Shopping, and YouTube each have separate indexes. Shopping interest and web interest peak at different times for the same product.

You can also compare a search term against a topic. A term matches that exact string; a topic groups the concept across spellings and languages. Topics give a truer picture of a subject, terms give a truer picture of the phrasing people type.

Below the chart, Trends lists related queries in two tabs. Top ranks the most searched related queries in your range. Rising ranks by growth rate, and this is where the value is. Rising surfaces queries gaining share right now, which is the closest thing to early warning that a free tool provides. A query labelled Breakout grew by more than 5,000% and usually has too little history to compute a normal percentage.

Rising and Breakout queries are the ones worth checking against Google Autocomplete and turning into pages before the term becomes competitive.

Timing a seasonal page. A roofing company checks "roof repair" over five years at state level and sees interest climb from a February floor to a peak in late summer, every year. Published in July, a storm-damage page is competing with everyone else and has no history. Published in March, it has four months to be crawled, indexed, and to accumulate signals before demand arrives. The chart does not tell you what to write. It tells you when to have written it.

Choosing between two phrasings. A firm cannot decide between "HVAC repair" and "air conditioning repair" for a service page. Put both on one chart, set the region to their state, set the range to five years. Whichever line sits higher is what their market types. If the lines cross seasonally, that is a signal to target both, one as the page and one as a section within it, rather than guessing. Confirm the winner against keyword difficulty before committing.

Sanity-checking a trend before investing. Before building a cluster around a topic, run it at 2004-present. A term in structural decline can still show healthy monthly volume in a keyword tool, because that tool is reporting a twelve-month average of a falling line. Trends shows you the slope. A five-year downward slope is a reason to reduce scope, not expand it.

Reading a spike correctly. A sudden vertical spike is usually news, not demand. Switch the search type to News and check whether the spike is mirrored there. If it is, the interest is transient and will not sustain a page. If web interest rose while news stayed flat, something real changed in the market.

  • No absolute numbers. Trends cannot tell you whether a term gets 200 or 200,000 searches. Confirm size elsewhere before you commit budget.
  • Low-volume terms return nothing. An empty chart is not evidence of zero demand. Most useful long-tail keywords are below the reporting threshold.
  • It is a sample. Numbers shift slightly between refreshes, so treat small differences as noise.
  • Local granularity thins out. City-level data for a niche term in a small metro is frequently unavailable.
  • It measures interest, not intent. A rising query can be curiosity rather than buying. Check search intent in the live results before building for it.
  • Always set the region before reading the chart. The worldwide default is the wrong answer for almost every local business.
  • Use five-year ranges for planning, twelve-month for scheduling. Long ranges reveal the pattern. Short ranges tell you where in the cycle you currently sit.
  • Compare, do not read in isolation. A single line at 0-100 is nearly meaningless. A second term on the same chart gives it scale.
  • Mine the Rising tab first. Top queries are already contested. Rising queries are the ones you can still win.
  • Feed findings into a dated plan. Trends output is only useful once it becomes a content calendar with publish dates set ahead of each seasonal peak.
  • Validate against your own data. Google Search Console shows what your site already gets impressions for. Trends shows the market. The gap between them is your opportunity list.

Frequently Asked Questions

It is a free chart of how much people search for something over time. The line runs 0 to 100, where 100 is the busiest point in whatever date range you picked. It shows you whether interest is rising, falling, or seasonal. It does not show how many searches happen.

No. It shows relative interest on a normalised 0-100 index, not counts. A score of 80 means 80% of the peak interest inside your selected range and region. Use a dedicated keyword research source when you need absolute numbers.

It is accurate for direction and comparison, and unreliable for anything small. The data is a filtered sample, so low-volume queries are excluded entirely and figures move slightly between refreshes. Trust the shape of the line. Do not trust a three-point difference between two terms.

Breakout replaces a percentage when a related query grew by more than 5,000%, usually because it had almost no search history before. It flags genuinely new demand. Treat it as a prompt to investigate the cause, since a breakout driven by a news cycle will fade and one driven by a product or regulation change may not.

Yes, with one adjustment: set the region to your state or metro before reading anything. National interest curves often peak at different times than local ones, and category-level differences by region are large. Expect city-level data to be unavailable for narrow terms.

Want the research turned into published pages without doing it manually? theStacc plans, writes, and publishes SEO articles straight to your site. Sign up for free →

Sources

Akshay VR

Akshay VR

Marketing Head · theStacc · ex-Sr Marketing Specialist, ARKA 360 · Malappuram, Kerala

Akshay leads editorial and content operations at theStacc. He writes about SEO craft, content operations, and the practical tactics that compound into search visibility.

LinkedIn Author page More terms by Akshay