We fetched 30 Tamil Nadu engineering college homepages as raw HTML with JavaScript switched off. Seven offered a crawler neither an enquiry form nor a phone number, and five of those seven returned no readable content at all. Four of the 25 readable homepages showed no enquiry route. Fix the path before you raise the ad budget.
A parent finishes dinner, opens your college homepage on a phone, and wants to ask one question about seats. Thirty seconds later they are back on Google. The budget was spent, the click happened, the parent arrived. What was missing was the route to a conversation.
19 of 30
gave a crawler both an enquiry path and a phone number
7 of 30
gave a crawler neither one
9 of 30
had no phone number in the raw HTML
What did we actually measure?
On 29 August 2026 we fetched the homepage of 30 Tamil Nadu engineering colleges as raw HTML, with JavaScript execution switched off. The sampling frame was a public enumeration of engineering colleges in the state, read in document order across Chennai, Coimbatore, Cuddalore, Madurai and Erode. We kept the first 30 that resolved an own-domain homepage.
- A visible enquiry path means an admission or enquiry form in the raw HTML, or a link whose own text offers one
- A generic Contact us in the footer did not count as an enquiry path
- A phone number on its own did not count either, because we recorded that separately
- Both definitions are ours, written down before collection started, and neither is a platform standard
What this instrument can and cannot see
One caveat runs through everything below. A page that returns nothing to a non-rendering fetch may still look completely normal to a human in a browser. Every claim here is about what a crawler receives, on one date.
What does a crawler see on a college homepage?
Twenty five of the 30 homepages returned readable content. Five did not: four returned nothing at all, and one returned fragments with no body copy. Those five are not proof of a bad website. They are proof that one class of visitor, the non-rendering kind, gets handed nothing.
| What the raw HTML offered | Colleges | Denominator |
|---|---|---|
| Readable content | 25 | 30 |
| No readable content at all | 4 | 30 |
| No phone number anywhere | 9 | 30 |
| No visible enquiry path | 4 | 25 readable |
| Neither an enquiry path nor a phone number | 2 | 25 readable |
Read the denominators before the numbers. The nine with no phone number include the four that returned nothing at all, where a missing number is part of a larger absence rather than a separate fault.
Does the average college site fail this test?
No, and that is the honest headline. Counting across all 30 homepages, 19 offered a crawler both a visible enquiry path and a phone number. Most colleges in this sample handle it adequately. The interesting number is the tail.
- 7 of the 30 offered neither an enquiry path nor a phone number
- 5 of those 7 are the rendering failures already described
- The other 2 returned readable pages carrying no form, no admission link and no phone number
- One of those two problems is a development ticket, the other is a decision somebody made
If you want the raw-HTML view of your own homepage before term starts, that is the first thing our admission funnel audit checks.
Free admission funnel audit. Your site, your ads, and your follow-up.
Why does a missing phone number matter more than it looks?
A phone number in the rendered page is enough for a parent using a browser. It is not enough for everything else reading your site now. Google's guidance on generative AI features in Search is direct about the precondition: a page “must be indexed and eligible to be shown in Google Search with a snippet”.
"A page a crawler cannot read cannot be quoted back to a parent who asks an AI assistant which Coimbatore colleges to shortlist"
- The same guide tells owners to ensure your content is crawlable, because the generative systems work from publicly accessible, crawlable content
- It also notes that browser agents may work by inspecting the DOM structure, and interpreting the accessibility tree
- A number that exists only after a script has run is a number several of your visitors never receive
None of this is an argument for stripping JavaScript out of your website. It is an argument for making sure the contact details survive without it.
Does Google see your enquiry form if JavaScript builds it?
Usually, eventually, and not always the way you assume. Google's JavaScript SEO basics documentation says Search runs JavaScript with an evergreen version of Chromium, and queues every page returning a 200 status code for rendering. The gap is timing and coverage, in three parts:
- Rendering happens later, in a queue, and the queue belongs to Google rather than to you
- Other crawlers are under no obligation to run your scripts, and Google's own wording is that not all bots can run JavaScript
- The same Google documentation still recommends rendering on the server, for that reason and for speed
"server-side or pre-rendering is still a great idea because it makes your website faster for users and crawlers, and not all bots can run JavaScript"
Which link patterns can Google actually crawl?
The link rules are stricter than the content rules. Google's link best practices say it can “only crawl your link if it’s an <a> HTML element” carrying an href attribute. Three colleges in our set had this fault: one Apply Now link resolved to a bare hash, and two form links returned empty pages.
| How the Apply Now button is written | Google's classification |
|---|---|
| <a href="/admission-enquiry"> | Recommended |
| <a href="/admission" onclick="track()"> | Recommended |
| <a routerLink="admission"> | Not recommended, may still be parsed |
| <a onclick="goto('/admission')"> | Not recommended, may still be parsed |
| <a href="javascript:goTo('admission')"> | Not recommended, may still be resolved |
Is a footer contact page an enquiry path?
We decided in advance that it is not, and that decision is ours rather than a platform rule. The reasoning is about what the parent is being asked to do, and the four routes below are not interchangeable.
- An enquiry form asks for a name, a phone number and a course, and starts a conversation
- A contact page publishes an address and a landline, then asks the parent to start it
- A counselling code link sends the parent to a government portal, which answers a different question
- A PDF prospectus answers questions the parent has not asked yet
- Four of the 25 readable homepages had no route meeting the first definition, and two of those four did publish a footer contact page
Where do the better sites put the form?
Read this section as colour, not as a finding. We tried to count clicks from the homepage to a working form, and that measure resolved for only 16 of the 30 sites against a completeness threshold of 20 set before we started. It is incomplete, and it carries no headline anywhere in this post.
- Two colleges embedded the enquiry form on the homepage itself, which is the shortest path available
- Eleven put it one click away, on a page reached from an admissions link
- Some routed to an external form service, which works for the parent but moves the record of the enquiry outside the college's own systems
- One published an admissions page carrying PDFs only, with no web form anywhere on it
When does this cost you the most?
The Tamil Nadu engineering admission year has a hard shape. Across ten admission keywords measured over the three complete years 2023 to 2025, 4 of the 10 drew under five percent of their annual search volume in January to March. Five percent is our cut, not an external standard.
Three caveats attach to that number, and they matter more than it does:
- It measures search demand, not enquiries. Nobody has measured enquiry volume for this population, including us
- The five percent line is ours. The raw distribution, not the pass count, is the real instrument
- The figures are Ahrefs index estimates for India, not Google data, on ten keywords rather than a census
The practical reading is narrow and still useful. Demand for these queries concentrates between May and August, so a broken enquiry path costs almost nothing in February and costs the year in June. The month-by-month shape is in our study of when Tamil Nadu engineering admission search peaks.
What should you fix first?
In this order, because each step is cheaper than the one that follows it.
- 1
Fetch your own homepage as raw HTML and read what actually comes back
- 2
Put a phone number in that HTML, as text, not inside an image or a script
- 3
Make the enquiry CTA a real anchor element with a real href attribute
- 4
Click that href yourself and confirm a form loads at the other end
- 5
Move the form to one click from the homepage, or onto the homepage
- 6
Repeat the fetch after the fix, before the campaign money starts moving
Steps 1 and 6 need no tool you do not already have. The related question of which crawlers execute JavaScript at all is covered in our read of four crawler documentation sets.
The metadata half of this same 30-college frame, covering titles and descriptions rather than contact routes, is in what most Tamil Nadu engineering college sites get wrong.
What this does not tell you
Thirty homepages, one state, one day. That is a snapshot of a population, not a rate and not a trend, and a site that failed the readability check may still render perfectly for a human.
- It says nothing about schools, which run a different admission cycle with no centralised counselling
- It says nothing about engineering colleges in other states
- It says nothing about how fast any of these colleges reply once an enquiry does arrive, because we never submitted one
- It measures the offer of a route, not whether anybody used it, and not whether the enquiry turned into an admission
- It is one instrument. A rendering browser, a screen reader and an AI crawler each see a different version of the same page
Key takeaways
- 19 of 30 Tamil Nadu engineering college homepages gave a non-rendering crawler both an enquiry path and a phone number. The problem is a tail, not an average
- 7 of 30 offered neither route. Five of those seven were rendering failures, two were readable pages with nothing on them
- 9 of 30 had no phone number in the raw HTML, and four of those nine returned no readable content at all
- Google can only crawl a link that is an anchor element with an href, so an Apply Now button driven by a click handler is invisible to it
- Demand for these admission keywords concentrates between May and August, so a broken path costs the year in June and nothing in February
Frequently asked
How do I check what a crawler sees on my college website?
Fetch the homepage without JavaScript and read the response. Any raw-HTML view, including your browser's view-source, shows roughly what a non-rendering crawler receives. Look for a phone number as text and an anchor element pointing at your enquiry form.
Does a Google Form count as an admission enquiry form?
It works for the parent, and two colleges in our set of 30 route Apply Now to one, one directly and one through an extra page. The tradeoff is that the enquiry record then lives outside your own systems, which makes response-time tracking and follow-up harder to run.
Will structured data get my college into AI Overviews?
Google states that structured data isn't required for generative AI search and that there is no special markup for it. Publish contact details in readable HTML first. Structured data is still worth adding for rich results in ordinary Search.
How many clicks should it take to reach the enquiry form?
We cannot answer this reliably. Our click measure resolved for only 16 of 30 sites and was declared incomplete before analysis, so any number here would be invented. Zero or one click is what the better sites in the sample did.
Does this study apply to schools in Tamil Nadu?
No. The population is Tamil Nadu engineering colleges. School admission runs on a different cycle with no centralised counselling, and nothing measured here transfers to it.
What should I check first if my ad campaign is already running?
Fetch the landing page the ads point at, not just the homepage, and confirm the form and the phone number are in the raw HTML. A campaign pointed at a page whose enquiry route only appears after a script runs is paying for arrivals it cannot convert.
Sources
- 1.Link best practices for Google, Google Search Central
- 2.Understand the JavaScript SEO basics, Google Search Central
- 3.Optimizing your website for generative AI features on Google Search, Google Search Central
- 4.Kula original analysis, Tamil Nadu college enquiry path audit, Kula Digital
- 5.Kula original analysis, Tamil Nadu engineering admission search seasonality, Kula Digital
Adith Krishnan
Co-Founder & COO, Kula Digital
8+ years building marketing that is measured in revenue. Runs strategy, AI search visibility, and education marketing at Kula. Written from the studio in Coimbatore.