What Does Brandlight Cost and What Does It Cover?

Brandlight is the most widely quoted company in this category that almost nobody can price. Its research is in our own library, in three separate posts. Its Series A was $30 million. Its customer list runs to Volkswagen Group, LG and Estee Lauder.

Its pricing page returns a 404.

We sell no platform. We run the measurement and the content work behind it for clients, so the only question we care about is whether the thing you are about to buy can carry the decision you are buying it for.

For this one, answering that means separating what the company publishes from what the category has inferred on its behalf.

What does Brandlight cost and what does it cover?

Brandlight publishes no price. Its pricing, plans and apex URLs all returned 404 on September 6 2026, and every figure in circulation, from $199 a month to $15,000, comes from aggregators that describe their own numbers as reported or estimated. On coverage, the vendor's own site names five engines. Third parties say six or eleven.

That is the whole post. The rest is what those two gaps cost you.

Brandlight publishes findings. It does not publish figures. Every price you have read about it was written by somebody who was also guessing.

Brandlight is better known for its research than for its product page

This is unusual enough to say plainly, because it changes how you should read everything else about the company.

The 70% to 20% finding travels further than anything Brandlight sells

In May 2026, 5W Public Relations released a report stating that the overlap between top Google rankings and AI-cited sources had collapsed from 70% to under 20%. Brandlight conducted the underlying analysis.

That number is now everywhere in this category. It is in our AI visibility platform buyer's guide, in our six metrics piece, and in our post on why Google rankings no longer predict AI citations. We put it beside Profound's independently measured 19% overlap and called it two-source corroboration.

The release discloses no sample size, no period and no method

We should have said that at the time, so we are saying it now. The 5W release gives the finding and stops. There is no prompt count, no engine list, no date range, no description of how overlap was defined.

The corroboration with Profound's 19% still holds, because Profound published its denominator: more than 1.5 billion real-user prompts. One number in that pair is checkable and one is not.

A statistic without a denominator is a claim wearing a lab coat.

A company that measures citation quality is judged by the same test

None of this means the finding is wrong. Directionally it matches every other read in the category, including our own.

It does mean the company selling you an authority audit shipped its most influential number without the disclosure it would flag on your pages. Fair to raise on a sales call, unfair to treat as disqualifying.

The one thing Brandlight publishes in abundance is answer-shaped content about itself

Brandlight runs a separate subdomain of question-shaped articles, every title phrased as a buyer question, bylined by its own CPO. "Which AI tool separates AI-assisted conversions?" is a title on that site.

That is a live GEO tactic, executed properly, by a company that sells GEO. Whatever you conclude about the product, the subdomain is worth ten minutes as a worked example.

What every Brandlight review asks:

• •How much does it cost per month?
• •Which AI engines does it cover?
• •Which enterprise logos use it?
• •How does it compare to Profound?

What the missing pricing page actually decides:

• •Which tier is being described when a review says "$199"?
• •Is Claude in my engine set or not?
• •What is the prompt quota, and does it multiply by engine?
• •What am I comparing against when I score the shortlist?

Every review answers the first list. Not one of them can answer the second, because the vendor has not published the inputs.

5 things the missing pricing page decides that no Brandlight review mentions

None of these are defects. Each follows from selling enterprise-only with no published tier, which is a legitimate commercial choice with consequences the review sites do not carry.

Consequence #1: Every price you have read about Brandlight was written by someone else

The $199 starter figure, the $400 to $750 band, the $750 setup fee and the $4,000 to $15,000 enterprise range all originate with aggregators.

checkthat.ai says so in its own copy, noting that Brandlight's pricing page 404s and that all public pricing derives from vendor-authored content and third-party aggregators. getaiso.com labels its figures as "directional scores from our structured rubric review" and says plainly they are not independently audited.

Both sites are being honest. The problem is that the honesty sits in a disclaimer and the numbers sit in a table.

Consequence #2: A $199 figure and a $15,000 figure cannot both describe the same product

The circulating range spans a factor of 75. That is not a pricing ladder, it is two different products described by people working from different scraps.

If the $199 tier exists, Brandlight has a self-serve entry point and belongs on a shortlist beside Peec and AthenaHQ. If the real floor is a paid pilot in the thousands, it belongs beside Conductor and is not comparable to any of them.

You cannot score a shortlist where one entry might be either.

Consequence #3: The engine count changes by source, and Claude is the swing item

Brandlight's own homepage names five engines: ChatGPT, Gemini, Perplexity, Copilot and Grok. Claude is not among them.

tooldirectory.ai names six and adds Claude back. getaiso.com says "a reported 11 AI engines" and never enumerates one. checkthat.ai gives a list of eleven that includes ChatGPT, GPT-4o and "OpenAI systems" as three separate entries.

GPT-4o is a model, not a retrieval surface. Counting it alongside ChatGPT is how a five turns into an eleven.

Consequence #4: Claude coverage is exactly where the rest of this bracket separates

This is why the swing item matters more than the price. Claude is Enterprise-only on Peec, Enterprise-only on Goodie, and absent from Ahrefs Brand Radar at any tier. We priced that gate in what Peec AI actually tracks.

If your buyers are developers, security teams or researchers, Claude coverage is the first filter and every other feature is downstream of it. On Brandlight, the vendor's own site and every third-party review disagree about whether it exists.

Consequence #5: With no published quota, cost per answer cannot be computed at all

Across this bracket we normalise every product to one figure: dollars per thousand AI answers. Ahrefs Custom Prompts runs $10.00, Otterly Premium $10.19, Peec Starter $21.11, Profound Growth $44.33, AthenaHQ Starter $81.94.

Brandlight cannot be placed on that ladder. There is no published prompt quota, no stated refresh cadence and no confirmed price, so there is neither numerator nor denominator.

That is not a criticism of the product. It is a statement about what your spreadsheet can hold, and it moves the comparison onto a call rather than into a tab.

What Brandlight sells that the self-serve bracket cannot

Fit is more useful to you than a verdict, and three things here are worth saying plainly.

Multi-brand governance is a real product category and almost nobody else builds it

Brandlight's module list runs to eight: Visibility and Insights, Technical Health, Content, Partnerships, Agentic Commerce, Ads, an Enterprise HQ view across brands and regions, and Attribution marked as coming.

An Enterprise HQ view is the thing a holding company or a multi-market CPG actually needs and cannot get from a tracker built for one domain. Publicis Groupe on the customer list is the tell: that is an agency buying for a portfolio.

The Partnerships module points at the half of the problem software usually ignores

Partnerships analyses which third-party publishers drive citations into AI engines, which is the correct thing to be looking at.

In our own concluded CITE Index study, four of the twelve most-cited domains were brand-owned, which means eight were not. Reddit alone drew 14,698 citations and appeared in 13.6% of all 90,132 AI answers. A module aimed at the eight is worth more than another chart about your own site.

The ads and agentic-commerce positioning is a bet, and it is an early one

The $30M Series A release of February 11 2026 frames the company around AI-native ads and a new marketing channel. Pelion Venture Partners led, with Cardumen Capital and G20 Ventures participating.

If AI ad surfaces mature the way the release assumes, buying measurement and media planning from one vendor will look prescient. If they mature more slowly, you will have paid enterprise money for a tracker with an ads tab. Both outcomes are live and nobody knows which yet.

Read a vendor's funding announcement as a statement of where it is going, not a description of what it does today.

How to run a Brandlight evaluation before the demo

The diagnostic half is done. Here is the sequence we run with clients evaluating this product specifically.

Step 1: Get the tier you are being quoted named in writing

Open the call by asking which tier the quote describes and what its published name is. Then ask whether a self-serve or starter tier exists at all, and at what price.

That single answer decides whether Brandlight is on the same shortlist as a $295 tracker or a different shortlist entirely. Nothing public can tell you.

Step 2: Ask for the engine list, enumerated, with Claude confirmed either way

Do not accept a count. Ask for the names, and ask specifically whether Claude is included at the tier you are being quoted.

Then ask whether GPT-4o, ChatGPT and Google AI Overviews are being counted as one surface or three. The answer tells you how the eleven was assembled.

Step 3: Ask which interface produced the Gemini and Copilot numbers

Google's grounding terms forbid using its results to build an index, and Microsoft's grounding terms carve out an exception that excludes exactly this use. We worked through both documents in what Conductor AI measures.

Grounding API, consumer interface or a licensed third-party route are all defensible answers. No answer is the finding, and it applies to most of the shortlist rather than to this vendor alone.

Step 4: Ask for the prompt quota, the cadence and the overage rate

These three numbers turn a quote into a cost per answer, which is the only figure that compares across the bracket.

Multiply your prompt count by the engines step 2 confirms, then by 30 for daily. Divide the monthly price by that product. We costed out which prompts earn a slot in how to select prompts for LLM tracking, and most sets come out half the size teams first propose.

Step 5: Ask for the method behind the 70% to 20% figure

You are buying measurement from the company that produced it. Ask for the prompt count, the engines, the period and the definition of overlap.

A vendor that can produce those four things on request has just demonstrated the exact capability you are paying for. A vendor that cannot has told you something too.

Step 6: Fund the earned half from the same budget line

Whatever the licence costs, hold back a matching share for the work that puts you inside other people's pages: earned mentions, comparison placements, review-site position, community presence.

Across those 90,132 answers the category leader flipped on only 18.7% of day pairs, and in four of the ten categories it never changed once. The average category leader held 78.2% of its own category's answers. The full corpus sits in the final report. That is incumbency rather than instrumentation, and it is why a managed GEO agency sits beside tooling rather than inside it.

Where Brandlight fits, by what each option constrains

Option	What it constrains	Right when
Brandlight	Price and scope, both of which are quoted. No published tier, quota or cadence, so cost per answer cannot be computed before a call	You run many brands or markets, need a portfolio view, and the Partnerships and agentic-commerce modules map to real budget lines.
AthenaHQ	Prompt count. Nine engines at daily cadence leaves about 13 prompts on $295 a month	You need daily depth on engines the cheaper tools cannot reach. The credit arithmetic is worked out in full.
Peec AI	Three of six engines on every self-serve tier, with Claude held for Enterprise	Your buyers cluster on three engines and you need unlimited seats. What Peec tracks covers the gating.
Rankscale	Credit balance, but scheduling runs from hourly to monthly	You need one converged reading rather than a standing dashboard. What Rankscale measures prices the depth.
Conductor	Sampling depth. The AI allowance is annual rather than monthly, and no price is published	You want AI visibility read beside ten years of your own organic data. The annual credit is priced out.
Goodie AI	Optimization actions. Ten fixes a month against roughly 15,000 observed answers	You sell through retail assistants, or you want the acting half metered rather than assumed. The two meters are divided out.
A managed program	Nothing, if scoped right. It costs a retainer	The measurement was never the bottleneck and nobody owns the weekly loop.

The six jobs to score any of these against sit in the AI visibility platform buyer's guide, and the rest of the field is mapped in our survey of GEO tooling for 2026.

FAQ

What is Brandlight?

Brandlight is an enterprise AI visibility platform that measures how AI engines represent and recommend brands, covering mentions, sentiment, share of voice and cited sources. It was founded in October 2024 by Imri Marcus, Uri Gafni and Didi Dvash, and raised a $30 million Series A led by Pelion Venture Partners in February 2026. Its site names eight modules including Visibility, Technical Health, Content, Partnerships, Agentic Commerce, Ads and an Enterprise HQ view across brands and regions.

How much does Brandlight cost?

Brandlight publishes no price. Its pricing page, plans page and apex-domain variant all returned 404 when we checked on September 6 2026, and there is no free trial or self-serve checkout, only a demo request. Figures in circulation range from roughly $199 a month at a supposed starter tier to $4,000 to $15,000 a month at enterprise, with a $750 setup fee mentioned by one source. Every one of those numbers comes from aggregators that describe their own pricing as reported or estimated rather than confirmed, so treat the whole range as unverified and get the tier named in writing.

What engines does Brandlight track?

It depends which source you read. Brandlight's own homepage names five: ChatGPT, Gemini, Perplexity, Copilot and Grok. tooldirectory.ai names those five plus Claude. getaiso.com refers to "a reported 11 AI engines" without naming any of them, and checkthat.ai publishes an eleven-item list that counts ChatGPT, GPT-4o and "OpenAI systems" as separate engines. Since GPT-4o is a model rather than a retrieval surface, the larger counts are padded. Ask for the enumerated list at your tier, and confirm Claude either way.

Is Brandlight worth it?

It depends entirely on whether you need portfolio governance or a tracker. For a holding company, an agency network or a multi-market brand, the Enterprise HQ view and the Partnerships module do work that no self-serve product in this bracket attempts, and the named customer list is genuinely enterprise. For a single-domain B2B team, the same money buys a published quota, a published cadence and a computable cost per answer somewhere else. The absence of any public pricing means the comparison cannot be made before a call, which is itself a cost.

What are the best Brandlight alternatives?

For enterprise multi-brand coverage, Profound and Conductor are the usual comparisons, with Scrunch AI now inside Sitecore. For single-brand tracking with published pricing, the names are Peec AI, AthenaHQ, Rankscale, Otterly, Goodie AI, Semrush AI Visibility and Ahrefs Brand Radar. Compare on cost per thousand AI answers rather than on headline price, because engine gating moves the real figure by a factor of eight across the bracket. Then ask every vendor which interface produced their Gemini and Copilot numbers, because most of them have never been asked.

The bottom line

Brandlight built a real product, raised real money and sells to brands you have heard of. Then it left the two numbers a buyer needs off the internet entirely, and the category filled the gap with estimates that span a factor of 75.

Do steps 1 and 2 before you take the demo. A named tier and an enumerated engine list turn this from an unpriceable entry on your shortlist into a comparable one, and both take a single email.

Do step 5 as well. You are buying measurement from the company that produced the most repeated statistic in this category, and asking for its denominator is the cheapest audit of a vendor's methodology you will ever run.

Then go do the work no licence covers. Nothing in the software writes the answer block, fixes the passage the model could not extract, or earns the third-party mention that puts you in the source pool. An AI visibility audit will show you which of those your gap sits in before you sign for a year of anything.