Entrepreneur, engineer and philosopher.

• Runs Xfaang (xfaang.com): visibility done professionally
• Co-creates K MAG and the Warsaw Glitch festival
• Writes treatises: Tractatus Volens, Theta, Ludens
• On this stage since WarsawJS #2 (2014, Ember CLI. Yes, really.)

piotrzientara.pl

SEO

Rank in a list of links

Win the click. Measure positions and CTR.

GEO · Generative Engine Optimization

Become the source the answer is built from

Win the citation. Measure how often models mention you.

1 · Rewrite

The question becomes several search queries.

2 · Retrieve

Results come from a search index: Bing, Google or the bot’s own crawl.

3 · Select

Relevant, trusted, fresh, easy to extract wins.

4 · Write + cite

The model paraphrases chunks and links a few.

Numbers with a source

“+42% in 28 days (AlfaUMi)” beats “a lot more”.

Named quotes

A real person, a real title, a sentence worth repeating.

Self-contained chunks

Each paragraph makes sense alone. No “as mentioned above”.

Clear entities

Full names, same spelling everywhere, “X is a…” definitions.

Fresh and dated

Visible “updated on” dates. Refresh pages that still get asked about.

Be talked about

Wikipedia, Reddit, YouTube, other media. Models learn from them.

# robots.txt
# search bots: show me in AI answers
User-agent: OAI-SearchBot
Allow: /

User-agent: PerplexityBot
Allow: /

# training bots: your call
User-agent: GPTBot
Disallow: /private/

1 · The question is the heading

Use the exact words people type. Mine them from Search Console queries and People Also Ask.

2 · Answer in 40 to 60 words

Right under the heading. Full sentence, no “it depends”. Details come after.

3 · Lists and tables

Steps as an ordered list, comparisons as a real table. Engines lift them as they are.

4 · Mark it up

FAQPage, QAPage, HowTo in JSON-LD. Fewer rich results today, but machines still read it.

Invisible 😴

Our thoughts on retail

Shopping has always been part of Polish culture. Since the 1990s, the way we spend weekends has changed a lot, and in this article we would like to take a closer look at…

Quoted ✅

Are shops open this Sunday?

No. Sunday 15 November is not a trading Sunday in Poland. Large shops are closed; small shops staffed by their owners may open. The next trading Sundays come in December, before Christmas.

E

Experience

You actually did the thing.

E

Expertise

You know the subject deeply.

A

Authority

Others cite and link to you.

T

Trust

The one that holds the others up.

Experience: prove you were there

The one thing a model cannot fake for you.

Do this

• Your own photos, screenshots, recordings. No stock.
• First person: “we tested”, “I was at”, “it broke when”.
• Say when and how long: “3 weeks, 40 users”.
• Include what went wrong. Failures read as real.

At K MAG

Reports written by people who were in the room. Like my feuilleton after Iggy Pop at OFF Festival 2022. Yes, I was there. No, ChatGPT wasn’t.

Expertise: a name, a face, a track record

“Admin” is not an author.

Do this

• A named byline on every article, linked to an author page.
• Author page: photo, bio, what they cover, all their texts.
• Stay on your beat. Depth in a few topics beats a bit of everything.
• Money or health topics: add “reviewed by” an expert.

In code

/autor/jan-kowalski
<a rel="author">
"@type": "Person"
"jobTitle", "knowsAbout"

Authority: others vouch for you

You can’t declare it. You can only earn it.

Do this

• Publish original data that others will cite.
• Show up elsewhere: interviews, podcasts, talks (hi 👋).
• One entity everywhere: same name, logo, links. Add Wikidata.
• Pillar pages with strong internal linking on your core topics.

In code

"@type": "Organization"
"sameAs": [
  "wikidata…",
  "linkedin…",
  "instagram…"
]

Trust: the boring pages that matter most

Google calls Trust the most important member of the family.

Do this

• About, contact, who owns and funds the site.
• Editorial and corrections policy. Fix errors in public.
• Label sponsored content clearly. Every time.
• Link primary sources. Honest dates: published and updated.

Trust killers

• Ads on top of the content
• “Updated today” on a 2019 article
• Fake reviews, fake authors
• No HTTPS in 2026 😬

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "Article",
  "headline": "…",
  "datePublished": "2026-11-11",
  "author": {
    "@type": "Person",
    "name": "Piotr Zientara",
    "url": "https://piotrzientara.pl",
    "sameAs": ["…linkedin…", "…github…"]
  }
}
</script>

One identity, linked everywhere.

That is the Experience and Expertise part of E-E-A-T, written in a format a crawler can parse.

Who?

Is it obvious who wrote this, and can a reader learn more about them in one click?

How?

Did we say how it was made: tested, measured, interviewed? Was AI involved?

Why?

Does it exist to help a person, or only to catch traffic?

Google and ChatGPT visibility in one check.

• SEO, GEO, content, technical, images
• Monitor: AI citations, Bing, Search Console, GA, PostHog
• Alerts when tracking silently breaks

alfaumi.pro

1.2M active users · +16.5%

22%

AI visibility: how often ChatGPT and Gemini bring up K MAG

90

Bing score, up 7% week over week. Bing feeds ChatGPT search and Copilot.

90,187

scrolled to the end of an article

+42%

138,964

pageviews (PostHog)

+36%

133

newsletter sign-ups

+50%

1 · Structured data

Article, Person, Organization in JSON-LD. Validate it.

2 · Real authors

Author pages, bios, photos, sameAs links to profiles.

3 · Answer first

Put the answer at the top of the page, then the story.

4 · Let bots in

Check robots.txt for AI crawlers. Server-render the content.

5 · Bing counts

Bing Webmaster Tools and IndexNow. AI search reads it.

6 · Measure it

Track AI citations, not only rankings. And check your tracking still fires.