Short answer

We read the websites of 30 mid-sized Indian manufacturers the way a buyer and a search engine read them, and marked each on 18 points. Nearly all gave a phone number (26 of 30). About half stated capacity as a number (14) or gave product specifications as figures (16). Very few gave what a buyer needs most: 2 described a past supply in any detail, 2 let a buyer attach a drawing to an enquiry, and on 1 a certificate could be opened.

Before a buyer calls a manufacturer, he looks the company up. We wanted to know what he finds. So for Arun Agrawal's guide Before the Buyer Calls, the websites of 30 Indian manufacturers who supply industry and projects were read and marked on 18 points a buyer looks for. This article sets out the results and what they mean for an owner. The study page, with its method in full, is here.

What did the 30 websites show?

"Yes" means the fact was found in the text of the pages read. "Unclear" means the evidence was ambiguous, or the page that would carry it was not among those read. It is never counted as yes or no.

Point checkedYesNoUnclear
First home page heading says what the company makes19101
Production capacity stated as a number14160
Machine or equipment list given15141
Plant location with town and state2505
Plant area or size stated13170
At least three clients named in text11145
Past projects or supplies described in any detail2271
Approvals or vendor registrations named6240
Certificates named1857
A certificate can be opened or downloaded11217
Product specifications given as numbers16140
Downloadable brochure, catalogue or datasheet (PDF)11172
Quote request form separate from the general contact form7212
Form lets the buyer attach a drawing or file2271
Phone number in text on the home page2640
WhatsApp link present12162
Named person or designation for sales8193
Statement about export or countries supplied15114

What do most sites get right?

The basics of being reachable. 26 of 30 put a phone number in text on the home page, and 25 gave the plant location with town and state. 19 opened with a heading that says what the company makes, and 18 named their certificates.

Where do they fall short?

On the facts a buyer uses to decide.

  • Past work. 11 named at least three clients in text, but only 2 described a past project or supply in any detail. A buyer sees who you worked for and learns nothing of what you did.
  • A route to a quote. 7 had a quote request form separate from the general contact form. 2 let the buyer attach a drawing, which is the one document that turns an enquiry into a quotation.
  • Proof he can open. 18 named certificates, yet on 1 could a certificate be opened or downloaded. 6 named approvals or vendor registrations.
  • Size of job. 14 stated production capacity as a number and 13 gave the plant area. The rest left the buyer to guess whether his order would fit.
  • A person to speak to. 8 named a person or designation for sales.

Why do these gaps matter more now?

Because the buyer is no longer the only reader. Google and AI assistants such as ChatGPT read the same pages, as text, to decide which suppliers to show or name. A fact that is missing, or shown only in a picture, cannot be used by either. The study read the sites as text only, which is roughly how a search engine or an assistant reads them.

There is an opening in these figures as well. The facts fewest sites give are the ones buyers need most. A manufacturer who writes down two past supplies, states capacity and adds a quote form that takes a drawing is already ahead of most of the field.

How were the sites chosen and read?

The sites were found through ordinary web searches for seven kinds of manufacturer that supply other companies: pre-engineered buildings and roofing, structural fabrication, castings and forgings, machined and sheet-metal components, industrial equipment, electrical products for projects, and industrial packaging. Each was chosen to look mid-sized, roughly Rs 20 to 200 crore a year. The home page was read with up to three inner pages, as text only, through an automated fetch tool, on 6 October 2026. The companies are not named.

What are the limits of the study?

  • Four pages a site at most, so the "no" figures are upper limits and the "yes" figures lower limits.
  • Text only. Logos, certificates and figures shown as pictures were not counted, though a person looking at the screen may see them.
  • Turnover was not verified. "Mid-sized" is a judgement from what each site says.
  • It is not a random sample. These are sites that came up in search and could be read, with Gujarat and Maharashtra the largest of twelve states. The weakest sites are under-represented.
  • One day. Sites change.

Read each figure as a count of what was found on the pages read, and not as a measure of any one company.

How does your own website score?

If you run a manufacturing business, you can check your site against 14 points in about ten minutes. It is free, and you get a scorecard showing what a buyer can and cannot find.

Score your website

For an outside view, the AI Search Visibility Blueprint is a written report on where your business stands with Google and the AI assistants, and what to fix first. It costs Rs 10,000 + GST, is delivered within 24 hours, and is refunded in full if it is not useful.

Questions about the study

Which companies were checked?

The companies are not named. They were found through ordinary web searches for seven kinds of manufacturer that supply other businesses, and each was chosen to look mid-sized: one to a few plants, not a listed giant and not a single-shed workshop.

Does a no mean the company does not have it?

No. It means the fact was not found, as text, on the pages read on that day. A site may say more on pages that were not read, or show it in a picture.

What is the single most common gap?

Past work described in detail and a quote form that accepts a drawing. Each was found on 2 of the 30 websites.

Can I check my own website against the same points?

Yes. The self-check on ebizindia.com covers similar ground in 14 statements, plus 2 for dealer-led makers and exporters, and takes about ten minutes.