Research · 5 September 2026
Every kind of browser-tool search grew after 2023. Developer utilities did not.
Yearly search volumes for 6,823 tool queries with a full five-year history, grouped by the job they name. Converters, calculators, timers and generators all rose between 2023 and 2025. The JSON formatters, encoders and validators stayed where they were.
One line per category, the developer utilities in the accent. Six categories left 2023 behind. One stayed.
Table view
| Category | Queries | 2021 | 2022 | 2023 | 2024 | 2025 | 2023 to 2025 |
|---|---|---|---|---|---|---|---|
| Developer utilities | 192 | 82 | 86 | 100 | 107 | 105 | +5% |
| Converters | 613 | 69 | 70 | 100 | 157 | 192 | +92% |
| Calculators | 1,739 | 83 | 79 | 100 | 145 | 162 | +62% |
| Generators | 313 | 59 | 67 | 100 | 128 | 148 | +48% |
| Tests and checkers | 183 | 81 | 85 | 100 | 122 | 144 | +44% |
| Timers and clocks | 175 | 82 | 88 | 100 | 126 | 177 | +77% |
| Randomizers and games | 274 | 77 | 85 | 100 | 107 | 116 | +16% |
| Whole pool | 6,823 | 74 | 75 | 100 | 123 | 133 | +33% |
What I tested
I build small tools that work in the browser, and I buy search volume data to decide which ones are worth building. The vendor's history goes back to 2021, a year per query, so I had a question I could answer with data I already owned: which kinds of tool are people searching for more, and which less?
Of the 14,557 tool queries with any history, 6,823 have a volume for every one of the five full years, 2021 to 2025. Only those count here, so no line bends because a query appeared or disappeared from the vendor's records. Each query is grouped by the job it names: a calculator, a converter, a generator, a test or checker, a timer or clock, a randomizer or game, or a developer utility (formatters, encoders, validators, hashes, placeholder text and the like). A query that names none of those is left out. Volumes are summed per category per year and indexed to 2023, the year the current wave of AI assistants arrived in editors and browsers.
Six categories grew. One did not.
From 2023 to 2025, the whole pool grew +33%. Part of that is real demand and part is how the vendor measures, which is why the comparison that matters is between categories measured the same way.
The slowest of the other six categories grew +16%. Developer utilities grew +5%.
Table view
| Category | Queries | Total change | Median query | Fell over 20% | Rose over 20% |
|---|---|---|---|---|---|
| Developer utilities | 192 | +5% | +5% | 24% | 44% |
| Randomizers and games | 274 | +16% | +59% | 14% | 69% |
| Tests and checkers | 183 | +44% | +78% | 9% | 79% |
| Generators | 313 | +48% | +95% | 6% | 71% |
| Calculators | 1,739 | +62% | +73% | 9% | 71% |
| Timers and clocks | 175 | +77% | +115% | 9% | 78% |
| Converters | 613 | +92% | +94% | 15% | 68% |
The largest developer searches
These are the developer searches with at least 100,000 searches a month in 2023, one per group of phrasings the vendor reports at the same volume. None of them grew.
Table view
| Search | 2023 | 2025 | Change |
|---|---|---|---|
| lorem ipsum | 165,000 | 140,000 | -15% |
| json formatter | 152,500 | 135,000 | -12% |
| format json | 121,225 | 114,150 | -6% |
| ascii table | 110,000 | 82,250 | -25% |
What it might mean
The two years in question are the two years in which every code editor and every browser gained an assistant that will format JSON, decode a token or produce placeholder text on request. A developer who used to open a new tab for that now asks the thing already open. The searches did not collapse; they stopped growing while every other kind of tool search kept climbing. That is a coincidence in time, not a proven cause, and I have no data on what those people did instead. What the data does say is narrower and firm: among the jobs people search a browser tool for, the developer ones are the only group that did not grow.
Limits
- Volumes are one vendor's yearly figures, derived from Google Ads data, rounded and grouped: phrasings of one search often carry an identical number. The named list keeps one phrasing per group for that reason.
- Requiring all five years excludes queries that appeared after 2021, so the pool's own growth is understated. The comparison between categories is unaffected: every category is held to the same rule.
- Categories are assigned by the words in the query, with my lists. A query that fits two is placed in the first, developer utilities before converters.
- The pool is queries with tool intent, mostly US, gathered while choosing what to build. It is not a sample of all search.
The data
The category tables and the named searches above are one file, the same file this page is rendered from, so a figure quoted from here can be checked against it. It does not hold the per-query list.
Download the data as JSON