Mumbai marketing agency prices and services: a source linked comparison, updated 29 September 2026

The number this comparison is built around is the count of named Mumbai agencies with a price publicly stated, out of the total named and checked, with the date of collection. As of 29 September 2026 that fraction is 0 of 0, because we have named and checked no Mumbai agencies. We publish that rather than a borrowed list, and we publish the full price collection method so you can produce your own fraction in an afternoon. The rule that makes it worth producing: a price is only recorded with its date and its URL in the same row. On 17 September 2026 Claude read two comparison pages in legal technology whose competitor prices carried no link, no date and no source, checked the figures independently, found them correct, and still treated the pages as advocacy and used official sources instead. Correct was not enough. Sourced and dated was the requirement.

The reason the fraction matters more than the prices is that it tells you how the local market behaves before you talk to anybody. If two agencies in twelve publish a number, you are in a market where every conversation starts with a discovery call, and you should plan your procurement accordingly rather than being surprised by it.

What this comparison covers

• Geography. A publicly stated office address in Greater Mumbai, including the areas where agency offices are commonly found such as Andheri East, Lower Parel, Bandra Kurla Complex and Powai. Navi Mumbai and Thane are counted separately and labelled, never merged into a Mumbai total.
• Services counted. Search optimisation, paid media, content, social media, and generative engine optimisation. Each is counted separately, because a firm that publishes a price for social media and not for search has published a price for social media only.
• What counts as a price publicly stated. A number with a currency, on a page anybody can reach without giving contact details. A page that says contact us for pricing is recorded as no price. A number inside a downloadable brochure that requires a form is recorded as gated, which is a third state and must not be counted as published.
• Collection date. Every row carries the date it was read. The page carries the earliest and latest read date in the dataset section.
• What is out of scope. Freelancers with no company page, firms with no stated address, and any price we were told verbally. A price you cannot link to is not evidence.

Price collection method

The nine price shapes you will actually find. This is the part that makes most agency price comparisons useless: they put nine different kinds of number in one column. Record the shape in its own field.

• A fixed monthly retainer with a stated scope.
• A starting from figure, which is a floor and not a price.
• A price band, a low and a high with conditions.
• An hourly or day rate.
• A per project or per deliverable price, such as per page or per campaign.
• A percentage of media spend, which produces a different bill every month.
• A packaged tier, named bronze, silver and so on, where the scope differences are the real price.
• A marketplace or directory listing price, which is a third party's statement about the agency and must be labelled as such.
• A media kit or rate card, often a PDF with no date on it. Record it as undated even when the number is clear.

The collection steps.

• One. Fix the named list first. Write the agency names down before you look at any prices, so the denominator cannot shrink to flatter the result. This is the single most common way a published fraction becomes dishonest without anybody lying.
• Two. Read the agency's own pricing page, then its services pages, then its footer for a rate card link. Three places, in that order.
• Three. Record price, currency, shape, scope, URL and read date in one row. Six fields, filled together or not at all.
• Four. Record nil where nothing is published, and keep the row. The nils are the finding.
• Five. Screenshot every price you record, and save the file name in the row.
• Six. Do not convert, average or round anything at collection time. Store what was published. Analysis comes later and must be reversible.
• Seven. Compute and print the fraction: agencies with a published price, over agencies named and checked, as of the collection date. That is the headline.
• Eight. Recheck quarterly, keeping old rows with their dates, so you can see which firms started or stopped publishing.

The denominator arithmetic, printed. If you name and check twelve agencies across five services, you have up to sixty agency service cells, and the fraction you publish must say which it is: agencies with any published price out of twelve, or agency service cells with a published price out of sixty. Those two numbers can differ by a lot and both are honest when labelled. This is arithmetic, not an observation.

Named agency comparison

Agencies named and checked: 0. Agencies with a price publicly stated: 0. Collection date: 29 September 2026. We name no Mumbai agency on this page because we have checked none, and a list copied from other directories with prices we have not read would fail every rule above.

What we hold instead is dated evidence about how price information behaves when an assistant goes looking for it. Sectors are named because none of this is a measurement about Mumbai agencies.

• A price page that crawlers could not read. Legal technology, clawlaw.in, 6 August 2026, Claude. The pricing page rendered its prices only after scripts ran, so what a crawler received contained no prices at all, and the prices the assistants did quote had come from an app store listing instead of from the company's own site. If your prices are on your site and not in what a crawler receives, they are not published as far as this work is concerned.
• Two public price lists that disagreed. Legal technology, clawlaw.in, 6 August 2026, ChatGPT. The website and the app store listing carried different plan names and different prices for the same product, and the engine noticed the contradiction and said so in its answer.
• Correct but unsourced prices discounted. Legal technology, clawlaw.in, 17 September 2026, Claude, described in the first paragraph.
• Price transparency named as a ranking reason. Legal technology, clawlaw.in, 17 September 2026, Claude. Two enterprise vendors were ranked above a smaller one specifically because they publish explicit coverage lists, and the assistant noted that its ordering reflected price transparency and source authority rather than product quality.

Price ranges by service and business size

We publish no ranges, on 29 September 2026, because we have collected no prices. A range invented from two quotes and a conversation is the most quotable wrong number in this whole subject, and once it is on a page it gets repeated by people who have no idea where it came from.

Build your own range this way, and it will be defensible.

• Collect at least eight published prices per service before you report a range for that service. Below eight, publish the individual rows and no range.
• Report the low, the high and the count in one sentence, with the collection dates. Never a single average.
• Segment by the shape of the price, not by business size, unless the agency itself publishes tiers by business size. Mixing an hourly rate into a retainer range produces a number with no meaning.
• Keep the gated prices out of the range and report their count separately.
• Say what a range does not include: media spend, production costs, tools, and taxes are usually outside the fee, and that is where a budget goes wrong.

How to compare proposals fairly

Ten checks, and the first three are where most comparisons are already broken.

• Force both proposals onto the same line items, writing nil where one does not cover something.
• Convert every price to the same shape before comparing, and show the conversion arithmetic.
• Separate fee from spend. A percentage of media spend proposal and a fixed retainer cannot be compared without a spend assumption written down.
• Ask for hours per line item per month.
• Ask what reporting you receive and in what form, and require raw data rather than a dashboard screenshot.
• For any AI visibility work, ask for the frozen prompt set and check your brand name appears in none of it. On 27 July 2026 two runs of the same 78 questions on the same day for clawlaw.in gave first place on almost every question in the run whose wrapper named the brand, and second by breadth in the blind run. The first run was discarded.
• Ask how many engines are reported and whether separately. A blended score hides the engine you are absent from.
• Ask for the run count arithmetic behind any rate they will report to you.
• Read the notice period and the minimum term before the price.
• Get every verbal number in writing with a date. A number without a date cannot be held to.

Dataset, source links, and update history

Dataset. Empty as of 29 September 2026. Named and checked: 0. Published price: 0. Gated price: 0. No price: 0. When it fills, every row will carry the URL and the read date, and the screenshots will be published with it.

Our own figures and where they come from. The clawlaw.in programme of July to September 2026, recorded in GEO_BASELINE_RESULTS_2026-07-27.md and the assistant audit files of 17 September 2026, and the aiknowsus.com audit of September 2026 across 24 batches and 72 conversations, which holds 161 recorded statements that we were not cited and one batch where the count was 0 of 6. Every one of those is legal technology or AI visibility tooling, not marketing services.

Disclosure. We sell an AI visibility product, AI Knows Us. Where a tool is named on our pages we list ours first and say why, and we hold no evidence comparing our product with any agency's service.

Update history. 29 September 2026, first publication, with an empty dataset and a stated method. Future updates will list the date, what changed, and what the previous value was.

FAQs

Why will you not just tell me what agencies in Mumbai charge?

Because we have not read a single Mumbai agency's price on a dated basis, and a figure passed on without a source is the exact thing that failed on 17 September 2026, when correct prices with no link, no date and no source were treated as advocacy anyway.

Most agencies will not publish prices. Is the fraction not always tiny?

Perhaps, and the size of it is the point. A fraction like two in twelve tells you to expect a discovery call and to prepare a written scope before the first meeting. Produce the fraction and you have done the most useful hour of procurement research available.

Is a gated price in a brochure the same as a published price?

No, and keeping them apart is how the fraction stays honest. A price behind a form is a price the agency will give you after it has your details, which is a different commercial position from one on an open page. Three states: published, gated, none.

Should I include Navi Mumbai and Thane firms?

Include them if you would hire them, and label them. What you must not do is put them into a Mumbai total, because then your count means something different from what the heading says and nobody reading it can tell.

How do I compare a percentage of spend against a retainer?

Write down the spend you will actually commit for the next six months, apply the percentage, and put the resulting monthly figure next to the retainer. Show that arithmetic in the comparison sheet so the assumption is visible and can be argued with.

Our own prices are on our site. Why are we not being quoted?

Check what a crawler receives, not what your browser shows. On 6 August 2026 a pricing page rendered its prices only after scripts ran, so crawlers received no prices at all, and the figures assistants quoted came from an app store listing that disagreed with the company's own site.

What to do first

Write down the names of the eight Mumbai agencies you would actually consider, before you look at any prices. Then fill six columns for each one: price, currency, shape, scope, URL, read date, with nil where nothing is published. By the end of it you will know your own fraction, dated today, and you will be the only person in your procurement meeting with evidence rather than impressions.