Home / Company / About us

We help AI tell the truth about you

Plámenos is a small AI visibility agency. We work on the places AI tools learn from, so that when someone asks about you, the answer is accurate and fair.

WHERE THE ANSWER COMES FROM

Six kinds of place a model learns about you from.

Why we started

The answer changed, the work didn't

People now ask ChatGPT, Perplexity, Gemini, Claude and Google's AI what to buy and who to trust. Those tools answer from what others say: Reddit threads, forums, Wikipedia, reviews, press. Most companies have no idea what that says about them.

We saw two kinds of help on offer. Agencies that only looked at your own website, and people selling shortcuts: fake accounts, bought reviews, quiet Wikipedia edits. The first misses where the answer comes from. The second gets found out. We started Plámenos to do the slower, honest version properly.

What we believe

Useful first, mentioned second

A few things we hold to, whoever the client is.

Work in the open

Every account we post from is real and disclosed. We follow each platform's rules, and on Wikipedia we declare our conflict of interest.

→ no sockpuppets

Earn it, don't buy it

No fake, paid or incentivised reviews, and no review gating. A good reputation should come from the honest majority of your customers.

→ no shortcuts

Be straight about limits

Nobody can promise what a model will say. We commit to the work and to honest reporting, and we'll tell you when something isn't worth doing.

→ no guarantees we can't keep

What we work on

Outside your company, and inside it

Most of what shapes an AI answer lives on other people's websites, so that's where most of our work happens. Some of it happens with your own team.

  • Communities and forumsReddit, niche and local forums, and branded communities, where people compare and recommend.
  • Reference and reviewsWikipedia, review platforms, press and directories: the sources models check facts against.
  • WebsitesSites that search engines and AI can read. This one is our own custom-code example.
  • People and habitsHelping teams adopt AI, and making sure someone owns how AI talks about the company.

How we work

Measure, work, measure again

Every engagement follows the same loop, so you can always see what changed and why.

Look first

We ask the major models about you and your category, and trace every source behind their answers.

Pick what matters

A short, ranked list of what's worth fixing. Sometimes the honest answer is "not yet", and we say so.

Do the work

On the platforms that matter for your category, by their rules, from disclosed accounts.

Show what moved

The same questions, asked again every month, with the work and the results side by side.

Who we are

A small team, on purpose

We're senior people who do the work ourselves. The person you first speak to is the person who runs your work, and our change programmes are led by a senior organisational change management lead.

We'd rather do less, well. Staying small means we turn down work we don't think we can do properly, and say so early.

Questions? Ask us

About how we work, what we will and won't do, or anything on this page. We reply within one working day.