Most arguments about whether a trend is finished get settled by feel. One person says a shoe shape is over, another says it is only now arriving, and neither has anything to go on beyond their own feed. There is a free public record that settles a decent part of the question, and it has been running since before most of the trends now on sale existed.
Search interest is not sales, and it will never tell you what to like. What it does well is date things. It can show you roughly when a look started climbing, how long it has been flat, and whether the spike in front of you happens every single year at the same point on the calendar.
A free record going back to 2004
Google Trends launched in May 2006 and covers searches from 2004 onward. Google's own guide to the tool, published in August 2021, describes it as showing what people were searching for at any date from 2004 up to a few minutes ago, split between a historic set and a real-time set covering the past week. That guide sits on the Google blog and is the fastest way to learn the controls.
Two decades of continuous data is what makes the tool useful for clothes specifically. A fashion cycle is long, often longer than the memory of anyone arguing about it. Being able to put a five-year window next to a twelve-month one turns a line that looks like a rocket into the third small bump in a slow decline, or the other way around.
What the number on the chart is not
The scale is where people go wrong, and getting it wrong produces confident nonsense. The values run from 0 to 100 and they are relative, never counts. Google's help documentation explains that each point is divided by the total searches for that geography and period, then placed on a 0 to 100 range against the term's own high point, so a reading of 100 marks the busiest moment inside the window you selected rather than any particular volume. The Trends data FAQ lays out those steps.
Two consequences follow. Changing the date range changes every number on the chart, because the peak it is all measured against moves with the window. And a term with tiny real volume can still draw a full-height curve, since it is only ever compared with itself. Height alone carries no information. Shape and timing are the parts you can use.
Curve shapes and what they mean
After looking at a few dozen garments the shapes start to repeat, and each one points at a different buying decision.
| Shape | Rough meaning | Sensible response |
|---|---|---|
| Steady climb across two years or more | A direction | Buy a good version |
| Vertical spike, fast fall | A microtrend | Borrow or skip |
| High and flat for years | Settled staple | Buy on price |
| Repeating annual sawtooth | Seasonal, not new | Buy off-season |
| Long slide with small bumps | Fading | Wait or pass |
The shapes worth the most attention are the boring ones. A long flat plateau is the profile of something that has stopped being a trend and become a normal garment, which is exactly what you want if you plan to keep it. The dramatic shapes are the ones that reward caution, because a curve that went vertical in eight weeks has no track record at all.
Choosing the search term carefully
Most bad readings come from the phrase typed in, not from the tool. Clothing vocabulary drifts constantly. A garment that everyone called one thing in 2019 gets a different name by 2023, and the old label shows a decline that never happened to the actual item. A few habits keep that from misleading you.
- Test the obvious synonyms together before concluding anything, since a fall in one name is often a rise in another for the same shape.
- Pick the topic option over the plain search term where it is offered, because a topic groups spellings, plurals, and related phrasings.
- Set the region to where you actually shop, as a shape that has finished in one country can still be early in another.
- Open a five-year window before a twelve-month one, so an ordinary seasonal dip does not read as a collapse.
- Avoid brand names as stand-ins for a silhouette, because those curves track publicity and price cuts more than the shape itself.
Seasonality that looks like a trend
Coats climb every autumn and sandals climb every spring, and both look like momentum if you only opened the chart last week. A rise that repeats at the same calendar point across three consecutive years is weather, not novelty. The fix is to compare like with like: set the chart to five years and check whether this year's peak is higher than the same month in previous years, rather than higher than last month.
That single comparison catches a great many false alarms. A sandal term at its yearly high in June proves nothing. The same term reaching a June high that beats the four previous Junes is the beginning of an actual signal, and one that is lower than all four is a category quietly losing interest under a curve that still looks healthy month to month.
Where search data stops being useful
The limits are real and worth holding onto. Searches measure curiosity, not spending, so a term can climb while nobody buys anything. Narrow terms are noisy, and a chart that jumps between 0 and 100 with nothing in between usually means the volume is too small to read. One televised moment can lift a term for a fortnight with no follow-through at all. And none of it knows your body, your climate, or your closet, which is where every real decision actually gets made.
Treat the chart as a date stamp rather than a verdict. It tells you where an idea sits in its own life, which is genuinely hard to judge from a feed, and then hands the decision back to you unchanged.
A ten-minute check before you buy
Next time something has your attention and it costs real money, run this before the purchase. Open Trends, type the garment name, set the region to your country, and set the window to five years. Note which of the five shapes you are looking at. Switch to twelve months and see whether the recent movement is a genuine change or the seasonal bump it makes every year.
Then write one sentence: the shape, the age, and what you plan to do about it. Something like a steady four-year climb, so a well-made version is safe, or a spike that started in March, so borrow it or wait. Ten minutes of chart reading will not tell you whether a garment suits you, but it reliably tells you whether you are buying near the start of something or paying full price at the end.