The Sixteen Second Wait · Research · September 2026
School Website Speed: We Tested 636 Indian International School Websites on a Phone
Slowly. In August 2026 we ran Google Lighthouse's mobile test on 636 Indian international school websites. The test simulates a mid-range phone on a slow 4G connection. The median website took 16.23 seconds to show its main content, usually the hero image or headline.
Google counts 2.5 seconds as good. Only 6 of the 636 websites (0.9%) met that mark, and 467 (73.4%) took longer than 10 seconds. Layout stability was mostly fine; loading was the problem. Below we show the full spread, which Core Web Vitals fail, what slow websites have in common and what good looks like.
16.23s Median time to main content (LCP) on a phone Google's "good" line: 2.5s
The benchmark
How fast should a school website load?
Google's line for Largest Contentful Paint (LCP), the moment a page's main content appears, is 2.5 seconds. Up to 4 seconds counts as "needs improvement". Anything slower is rated poor (Google, web.dev).
On that scale, 608 of the 636 school websites we tested (95.6%) fell in the poor band. Another 22 (3.5%) needed improvement.
The chart below places every website in a 5-second band by its median load time.
How long 636 school websites took to show their main content
The two busiest bands are 5 to 10 seconds (119 websites) and 10 to 15 seconds (117). So the 16-second median describes a typical school website. The 41 websites that took more than a minute sit in their own bar at the end.
Core Web Vitals
Which Core Web Vitals do Indian school websites fail?
Core Web Vitals are Google's three page-experience measures: loading (LCP), responsiveness (INP) and visual stability (CLS). INP needs real visitors tapping a page, so a lab test cannot record it. Lighthouse reports Total Blocking Time (TBT) instead, which tracks the same kind of delay (Google, web.dev).
The bars show the share of the 636 websites that met each "good" line in our test.
Share of school websites meeting Google's "good" line, by metric
Most websites held their layout steady and responded without long script delays. Loading is where they failed, which leaves just 4 websites (0.6%) meeting all three lines at once.
Lab results differ from the field data in Google Search Console, which comes from real visitors on their own phones and connections. A lab test is harsher, but it measures every website under the same conditions.
Lighthouse score
What Lighthouse performance score do school websites get?
Lighthouse rates performance from 0 to 100. Scores of 0 to 49 are poor, 50 to 89 need improvement and 90 to 100 are good (Chrome for Developers). The median school website in our test scored 54.
The chart counts websites in each 10-point range of the score.
Lighthouse mobile performance scores of 636 school websites
227 websites (35.7%) scored below 50, and 406 (63.8%) landed in the middle band. The 50s alone hold 247 websites.
A score in the 50s can sit alongside a very slow load. The score blends five lab metrics, and stable layouts and low blocking time lift it even when the main content arrives late. Only 3 websites reached 90; we show them in the last section.
Technology
Why are school websites so slow?
A lab test shows what a homepage loads. It cannot show why a school's website was built that way. So we set a scan of each homepage's technology beside its load time. What follows are correlations from one test, and none of them proves a cause.
jQuery is on most school websites
jQuery is an older JavaScript library that many themes and plugins still depend on. Our scan found it on 473 of the 645 websites we could scan.
WordPress runs half of them
Our scan identified WordPress on 336 websites (52.1%). It found no content management system on 278 (43.1%), which may mean a custom build or a system the scan could not recognise. Other systems make up the remaining 31 (4.8%).
What 645 school websites are built on
How load time differs with and without each technology
Each row compares the median load time of websites with a technology (filled dot) against websites without it (open dot).
Median time to main content, with and without each technology
WordPress barely moves the median. Websites on it took 16.48 seconds, against 16.19 seconds for the rest.
Websites loading jQuery were 3.57 seconds slower at the median. The library may simply mark an older, heavier build, so removing it alone may not close the gap.
Websites where we detected a content delivery network (CDN) were 5.27 seconds slower. A CDN should speed up delivery, so the likelier reading is that heavier websites are the ones that add one. Either way, a CDN alone did not make these websites fast.
What parents see
Does a slow website cost a school admissions enquiries?
Our test cannot count enquiries, and we have not found a study that measures this for schools. What the test can show is the wait a parent sits through before the page is any use.
Here is that wait for the median school website, in seconds from the first tap.
The median school website loading on a phone
For the first 4.65 seconds, the parent sees nothing. Google's good line for that first paint is 1.8 seconds, and 18 of the 636 websites (2.8%) met it.
Whether this wait costs a particular school enquiries depends on its parents, its city and the other schools on their list. We would rather show you the wait than put an invented number on it. Your free scorecard puts your own website's timings next to these.
What good looks like
How can a school make its website faster?
Start with the websites that already load well. Three school websites scored 90 or above on Lighthouse performance in our test, and each showed its main content within 2.5 seconds.
| School | City | Performance | Main content (LCP) | First content (FCP) | Blocking time (TBT) | Layout shift (CLS) |
|---|---|---|---|---|---|---|
| K.R. Mangalam Global School | New Delhi | 99 | 1.81s | 1.46s | 0ms | 0.006 |
| Hiranandani Upscale School | Chennai | 96 | 1.31s | 1.18s | 48ms | 0.016 |
| JS Global School | Chennai | 96 | 2.10s | 1.80s | 14ms | 0.054 |
Two of the three, K.R. Mangalam Global School and JS Global School, run WordPress and load jQuery. That is the same combination found on many of the slow websites above. The platform did not stop them loading quickly.
What to ask your web team
- Run the homepage through PageSpeed Insights on mobile, and note LCP, TBT and CLS before anything changes.
- Find the element the report names as the LCP element, often a hero image or slider. Ask for it to be resized, compressed and loaded first.
- List every script the homepage loads, including sliders, chat widgets and tracking tags. Remove or delay the ones a parent does not need on arrival.
- Retest on mobile after each change, so you know which fix moved the number.
Sometimes the theme itself is the weight, and patching it costs more than starting again. Our sister agency Pixelmattic builds school websites when a rebuild is the better route.
Method
How did we measure school website speed?
We ran Google Lighthouse on the homepage of each school in our study on 19 August 2026, using its mobile setting. That setting slows the processor and the network to mimic a mid-range handset on slow 4G. The results are lab data from one controlled run per homepage, not field data from real visitors.
We attempted 683 homepages and 665 runs succeeded. We excluded 3 runs that recorded a load time of zero, which means the measurement failed. That left 662 homepages.
Some school groups run several campuses from one website, and each campus returns the same result. So every speed figure on this page counts each website once: 636 websites, each taking the median of its campus rows.
Technology figures come from a scan of each homepage's HTML on the same day, covering 645 websites. The comparisons of load time by technology use the 630 websites that have both results.
We report medians throughout. A long tail of very slow websites pulls the average up, so it would overstate the typical wait. We name schools only where they did well, after checking each name against the school's live homepage.
We published this because we could not find a speed benchmark for Indian international schools. When we asked ChatGPT how Indian international school websites compare on speed and whether any benchmark exists, it answered:
"I would not treat any single published 'Indian international school' benchmark as authoritative."
ChatGPT with web search, asked to compare Indian international school websites on speed and Core Web Vitals and to name a benchmark, 15 September 2026.
This page is one part of our wider audit. The full method for the study is in the International Schools in India: Digital Report 2026.
Where does your school's website sit in this spread?
We run the same mobile test on your website and place it against these 636. The free scorecard also covers AI readiness, the admissions path, ads and reviews.