SEO & AI SEARCH AGENCY FOR KOLKATA BRANDS

The one that can show you its work on a business like yours and won’t promise you a ranking. A top ten list can’t tell you that, because most of those lists are written by agencies that put themselves first.

Google suggests “top 10 seo company in kolkata” as you type, so plenty of people start there. The worry behind it is simple: every agency says it’s the best, and you can’t tell who will move your enquiries and who will send a monthly PDF while nothing changes.

So ask four things on the first call. Show me a result for a business in my category, with the timeframe and how you measured it. Who exactly will work on my account? Will I own Search Console, GA4 and my Business Profile from day one? And can you show me who ChatGPT and Google’s AI name in my category today, in English and in Bengali?

If the answers are vague, walk away. If someone guarantees page one, walk faster. Google’s own guidance says no one can guarantee a number one ranking.

Then check whether the agency can be found for its own services. An SEO agency that can’t rank for SEO is telling you something.

My rule: pick the agency that is most specific about the first 90 days, not the one with the biggest claim.

Sometimes, if the job is small and the deliverables are clear. More often, low cost means low effort, and you only find out six months later.

Google suggests “low cost seo services in kolkata” and “affordable seo company in kolkata”, and I understand why. Kolkata businesses are careful with money, and they should be. But published Indian retainers run from about ₹8K a month to ₹300K and above, and one agency’s own guide says anything under ₹15K tends to be activity-based rather than outcome-based. At that level you’re usually buying directory links, a report and a blog post nobody reads.

The real question isn’t the price. It’s what the money fixes. Which technical problems get solved, and when? Which pages get rewritten, and why those? How will you know whether you appear in Google’s AI answer?

I’ll be transparent about where we sit. We’re not the cheapest option. Every engagement starts with a $1,500 audit, credited to your first month if you continue, and the retainer is scoped in writing from what it finds.

If your budget is small, I’d rather you hired a good local freelancer and kept your Business Profile current than paid an agency to look busy.

My rule: never pay for a retainer whose deliverables you can’t list.

If your buyers search in Bengali, yes, at least for the pages that answer their questions. If they search in English, a Bengali site won’t help much. Start from how your customers actually ask.

This matters more than it did. Google rolled out AI Mode in Bengali in October 2025, so a buyer can now ask a full question in Bengali and get an AI answer that names brands. If your facts only exist in English, you’re harder to find in that answer.

Here’s the problem behind the question. Many Kolkata businesses sell in two worlds. The distributor in Siliguri may search in Bengali, while the corporate buyer in New Town searches in English. One language rarely covers both.

So I’d do three things. Look at your enquiries and calls, and note which language people use. Ask ChatGPT and Google the same buyer question in both languages and see who gets named. Then build Bengali pages for the questions where Bengali buyers matter, written by someone who reads the language.

The failure I see is machine translation dropped onto a whole site. It reads badly, it doesn’t match the words people search, and it can make your brand look careless.

My rule: translate the questions your buyers ask, not the whole website.

It works to get you named in the AI answer, not just to rank a page, and it can show you that answer before and after. If it can’t show you, it’s SEO with a new label.

Google now suggests “ai seo company in kolkata” and “ai seo expert kolkata”, so the demand is real. Here’s the difference. Classic SEO asks which page ranks for a keyword. AI search asks which brands ChatGPT, Gemini or Google’s AI name when a buyer asks who to choose. Those aren’t the same list. A brand can sit on page one and never appear in the answer, because the engines lean heavily on what other trusted sites say about you.

So the work adds three things to normal SEO. Measuring how often you’re named across many questions and many runs, because AI answers change each time. Finding the sources the engines cite for your category and earning mentions there. And keeping your brand facts identical everywhere the engines look.

For Kolkata there’s a fourth. The questions need asking in Bengali as well as English, because the answers can differ.

My rule: ask any agency to show you today’s AI answers for five of your buyer questions before they quote. If they can’t, keep looking.

Because ChatGPT doesn’t know what your customers know. It builds its answer from what the web says about you, and a reputation built over generations at a counter often never made it online.

I see this pattern often with older businesses. Everyone in the trade knows the name. But the website is thin, the facts sit inside images or PDFs the engines can’t read, the Business Profile hasn’t been touched in years, and the name, address and phone number differ from one listing to the next. Very few third-party sites write about the business, so the engine has nothing to cite.

The fix is to write the reputation down where machines can read it. State in plain text what you make, who you supply, where you deliver and how long you’ve been doing it. Make your details identical everywhere. Get mentioned on the sites the engines already cite for your category: trade press, comparison articles, business news, reviews.

Start today by asking ChatGPT the ten questions your buyers ask, in English and in Bengali. Write down who gets named and which sites it cites. That’s your target list.

My rule: if the web doesn’t say it, the AI can’t repeat it.

Make your Google Business Profile the most complete, consistent and actively reviewed listing in your category for that part of the city. The Maps pack runs on different signals from the blue links.

Google suggests “best gmb seo company in kolkata”, which tells you how many businesses want this. Google weighs three things for local results: relevance, distance and prominence. You can’t move your shop closer to every searcher, so you win on the other two. Relevance means the right primary category, your services listed properly, and a page on your site for each area you really serve. Prominence means reviews, mentions and links from local places.

Kolkata adds a twist. Many businesses here have one famous address and a lot of copies, old listings or lookalike names nearby. If your details differ across the web, Google isn’t sure which listing is really you.

Where I see listings lose: the wrong primary category, a phone number that changed years ago, photos from another decade, and reviews that stopped coming.

Be careful with reviews. Ask every happy customer and reply to all of them. Never buy reviews or reward good ones. It breaks Google’s rules, and it gets profiles suspended.

My rule: fix consistency first, then categories, then reviews. In that order.

Yes, if it does the local homework. Being in the same city guarantees nothing, and plenty of local agencies outsource the actual work anyway.

Here’s our position, plainly. We’re based in Gurgaon and we don’t have a Kolkata office. We work with Kolkata brands by appointment, mostly on video. Most of the work in SEO happens on screens anyway: your Search Console, your pages, your AI answers. A video call where we look at those together is often more useful than a meeting across a table.

What a remote agency can miss is how the city searches. Salt Lake Sector V isn’t Burrabazar. A distributor in Assam doesn’t search like a shopper in South Kolkata. Bengali questions get different AI answers from English ones. So ask any agency, local or not, to show you how they’d learn your market: your enquiries, your calls, your customers’ words.

There’s also a fair point the other way. Some things need someone on the ground, like photos and video of your shop or factory. Plan for that.

My rule: hire the people who will learn how your buyers search, wherever they sit, and meet the person doing the work, not the person selling it.

Expect early movement in three to four months and enquiries you can trace to search in six to nine. Anyone promising results in 30 days is working on a phrase nobody searches or not telling you everything.

Those bands aren’t only mine. One Indian agency’s published guide puts traffic gains at three to four months and lead attribution at six to nine, which matches what I see.

Some things move faster. Technical fixes, schema and Business Profile clean-up often show within weeks. So do pages already sitting on page two. And for many older Kolkata businesses, there’s a quick win hiding in plain sight: a name people already search for, landing on a weak page or someone else’s listing. Fixing that can show results quickly.

Bengali content and regional pages take longer to build, but they often face less competition than English pages for the same question.

The failure I see is quitting at month three because traffic hasn’t doubled, just before the compounding starts. The opposite failure is staying a year with an agency that never shows you anything moving.

My rule: by day 90 you should see the work done and early signals in Search Console and your AI answers. If you see neither, ask hard questions.