Executive Summary

TYS in one page: what we do, how we prove it, how we work

TYS is a founder-led digital performance practice operating in two markets, Turkey (tys.net.tr) and Germany (tysd.de). The work: making brands visible and correctly represented in classic search and in generative AI answers (ChatGPT, Gemini, Claude, DeepSeek), and keeping them that way through monthly, human-verified reporting.

This page condenses the practice into what a decision maker needs: the problem, the model, the measurement anchors, the proof principle and the engagement terms.

The problem, in business terms

Buying decisions are increasingly shaped by AI answers, not only by search rankings. A brand can rank well on Google and still be absent, or worse, described incorrectly, when an AI assistant answers a customer question. That is unmanaged revenue risk: invisible in reports, real at the moment of decision.

Most organizations do not measure this channel at all. The cost of finding out is low; the cost of not knowing compounds.

The model: one loop instead of scattered tactics

All TYS work runs on one operating model, TYS Digital Performance, with four phases: Diagnose (measure the current state), Build (make the brand machine-readable), Grow (produce demand through citable content and channels) and Monitor (re-measure monthly and feed findings back into the loop).

The model exists so that budget maps to phases, every deliverable has a defined place, and no work is sold twice. It is documented in full on the operating model page.

Three named anchors

The practice is built around three named, repeatable instruments:

AI Authority Score

A 100-point measurement across four AI engines (ChatGPT, Gemini, Claude, DeepSeek), split into Entity (35), Citability (30) and AI Visibility (35). Live queries, archived with screen evidence, repeatable across time. It is a measurement anchor, not a promise: reports state that AI answers vary and that repeated measurement is the basis.

BrandLock

The brand canon: a verified, machine-readable record of what is true about the brand, used to detect and correct incorrect or incomplete AI representation. Boundary honesty: we cannot edit AI output directly; we align the canonical signals AI feeds on with the verified truth of the brand.

TYS Visibility Pilot (90 Days)

The standard entry: a Day 0 baseline, targeted canon and schema hygiene with at least one concrete fix and one implemented improvement per month, and a Day 90 re-measurement with the same methodology. It ends in a decision gate; continuing into monitoring is the client's call.

The proof principle: baseline, then measure

No growth number is quoted before a baseline exists. Impact is shown as the measured difference between Day 0 and Day 90, and monthly thereafter, in the client's own data: score movement per category, mentions per engine, share of voice against named competitors, and classic search data where Google Search Console and GA4 are connected.

Where numbers appear in our case studies, they were measured for that client and are labeled by evidence type. We do not convert one client's outcome into another client's promise, and we do not use guarantee language anywhere.

How an engagement works

  • Founder-led and deliberately small: concurrent projects are capped, and we do not sell when capacity is unavailable.
  • Entry is low-friction: a free AI visibility check, then the paid diagnosis or the 90-day pilot. Ongoing monitoring has a 6-month minimum and is only decided after the pilot's measured results.
  • Two delivery modes: TYS implements directly, or your team receives a prioritized, technically explicit work list. Every change is logged.
  • Pricing: transparent scope drivers in the Turkish market, including a proposal calculator with real list figures; conditions on request in the German market. Reports are human-verified; billing is currently a manual process, stated plainly.

Risk controls, for the buyer

  • Every report passes human review before it reaches you; no auto-published metrics.
  • Scope is capped in writing; changes are logged with what, why and when.
  • Measurement variance is disclosed, not hidden: the score is a repeatable anchor, not a stochastic lottery ticket presented as certainty.
  • What we cannot do is documented on every relevant page: no direct control over AI output, no Core Web Vitals field commitments without a dedicated measurement pipeline, no features sold that do not exist.

Executive questions

How do we know it works?

You see it in your own data: a Day 0 baseline, a Day 90 re-measurement with the same methodology, and monthly deltas thereafter, with screen evidence. The decision to continue is taken after the proof window, not before it.

What does it cost to find out?

The first step is free: the AI visibility check. The paid diagnosis and the 90-day pilot are scoped offers; in the Turkish market scope drivers and the proposal calculator are transparent, in the German market conditions are provided on request.

Who actually does the work?

A founder-led practice with a production-grade measurement engine, not a large agency bench. That is a deliberate trade: narrow scope, capped capacity, direct senior attention. If your need is scaled media buying or a large embedded team, we say so and point you to the right category.

What are the main risks, honestly?

Two are structural: AI answers are variable by nature, so single measurements can mislead; we mitigate with repeated, evidence-archived measurement. And visibility work depends partly on third parties (search engines, AI providers); we mitigate by building durable machine-readable assets you keep, whatever happens to any single channel.

How fast do we learn something useful?

The baseline lands within the first week, with expert commentary within 24-48 hours of measurement. The first implemented improvements follow in month one; the full evidence file exists by Day 90.

Where to verify

Run the free AI visibility checkContact us