If you went looking for Adobe LLM Optimizer recently and ended up somewhere else, you did not take a wrong turn. The product page, its pricing page, and the original announcement all now redirect to something called Adobe Brand Visibility, whose heading states the situation plainly: formerly Adobe LLM Optimizer.
That happened on 17 June 2026, roughly a year after the original launch. Nothing was cancelled. The tool grew into a bigger platform, took a new name, and left a lot of people searching for a product that no longer answers to it.
Here is what it actually was, what it became, and whether any of it is worth your attention if you sell things online.
Strip away the category jargon and Adobe LLM Optimizer answered one question: when someone asks an AI assistant about products like yours, does your brand come up, and what does it say?
Adobe's own documentation breaks that into five jobs. Track how often your brand appears in AI-generated answers, with a score you can watch over time. Compare that against competitors to see who is being mentioned instead of you. Find the gaps worth fixing, which it splits into content work like FAQs and structured data, and technical work like whether the AI can crawl your pages at all. Watch the traffic that comes back, separating visits from AI assistants themselves from visits by people who followed a recommendation. And share all of it across the marketing, SEO, and PR people who each own a piece of the problem.
Read that list again and notice something. Three of the five are measurement. The tool told you where you stood. What you did about it was still your job.
That gap is more or less what Adobe says prompted the rebuild. Brand Visibility is pitched as the step from measurement to action to proven outcome, and the additions are substantial rather than cosmetic.
It now runs on market intelligence from Semrush layered under Adobe's own execution and attribution, drawing on a pool of 261 million real AI search prompts rather than guesses about what people ask. It reads agentic traffic out of CDN logs, which is the only reliable way to see AI crawlers that never render a page. And it connects visits that originated in an AI answer to what those visitors actually bought.
That last one is the meaningful change. Plenty of tools will tell you that ChatGPT mentioned you. Very few will tell you what those mentions were worth.
The numbers Adobe uses to justify the category are worth knowing even if you never buy the product. Citing Semrush data, it says AI-referred traffic converts at 4.4 times the rate of organic search. Citing Bain research, it says 80% of brands have significant gaps in how they show up in AI today. Both come from a vendor selling the solution, so treat them as directionally useful rather than as your forecast. The direction matches what we see: this traffic is small, it is growing, and it buys.
Adobe does not publish a price. The pricing page that used to exist for LLM Optimizer redirects to a product page whose call to action is a conversation with sales.
That is not a criticism, it is information. Enterprise software priced on request is priced per customer, and per-customer pricing exists because the deals are large enough to negotiate individually. If you are running a store doing a few million a year, you are not the buyer this was designed for, and you will spend the meeting finding that out.
The honest shortlist is short.
You already run on Adobe's stack, so the data plumbing and the attribution work without a project to connect them. You have a marketing team with separate people for SEO, content, and communications, because a shared dashboard only helps when several people need to see it. And your brand is searched for by name often enough that being described wrongly in AI answers is a real commercial problem rather than a theoretical one.
If all three describe you, this is a serious product and the Semrush data underneath it is genuinely hard to replicate.
If one or none describe you, the tool will measure a problem you cannot afford to act on, which is the most expensive kind of software there is.
Here is the part most owners miss. Look back at Adobe's own list of what to fix: FAQs, abstracts, structured data, crawlability, indexing. None of that requires the platform. The platform tells you which of them to do first. You can do all of them today.
Start with whether AI systems can read your site at all. If your product pages need a browser to render their content, some crawlers see an empty page. If your prices, stock status, and product attributes are not published as structured data, an assistant has to guess, and it will guess wrong or skip you. This is the same groundwork that decides whether ChatGPT and Gemini recommend your products, and it costs a few days of a developer's time, not an enterprise contract.
Then answer real questions in plain text on the page. The FAQ block that Adobe's tool nudges you toward is not a ranking trick. It is the format an assistant can quote directly, which is why cited pages tend to have one.
Then check the commercial plumbing behind it, because being recommended is only half of it. An assistant that can find you but cannot complete a purchase sends the shopper elsewhere, which is a different problem with its own checklist.
And measure crudely before you measure expensively. Ask the major assistants the questions your customers ask, once a month, and write down what they say about you and who they name instead. That is a spreadsheet and an hour. It will not scale, it will not attribute revenue, and it will tell you eighty percent of what you need to know about whether this is a problem for you yet.
Adobe renaming a product a year after launch is not a scandal. It is what happens in a category that did not exist two years ago and is still deciding what it is called, which is also why you should be careful about building a process around any single tool in this space right now.
The underlying shift is real and it is not going away. What is unsettled is which vendor owns it, what it is named this quarter, and what it should cost. We saw the same sequence play out with Adobe's platform strategy on the commerce side, where the product you buy today and the product you are upgraded to in two years are increasingly different things.
The work underneath, though, stays stable. Clean data, readable pages, honest answers to real questions. That was worth doing before anyone sold a platform for measuring it, and it will still be worth doing after the next rename.
If you are trying to work out whether AI assistants are recommending you or quietly skipping you, that assessment is something we do at Encomage without a licence attached. It usually starts with reading your product data the way a machine reads it, which tends to explain more than any dashboard does.
Let’s build something powerful together - with AI and strategy.
