Search changed shape. Buyers still open Google — but a growing share of them now ask ChatGPT, Gemini or Perplexity first, and act on an answer that names two or...
Search changed shape. Buyers still open Google — but a growing share of them now ask ChatGPT, Gemini or Perplexity first, and act on an answer that names two or three companies. If your brand is not one of those names, you were never in the consideration set.
VIDCO is an AI Search and organic growth agency headquartered in Hà Nội, Vietnam, working with brands across global markets. We do two things: rank you in classical search, and get you named in generated answers — and we measure both with a method we publish.
| Classical search | Generative search | |
|---|---|---|
| Surface | Google results, Maps, images | ChatGPT, Gemini, Perplexity, AI Overviews |
| What wins | Relevance, authority, technical health | Being named and cited by sources the model trusts |
| How you measure | Rankings, impressions, clicks | Mention Rate, Citation Rate, Share of Model |
| What it needs | Content and links | Entity clarity and third-party citation |
These are not two strategies. They are one strategy measured on two surfaces — which is why we do not sell them as separate retainers.
An agency tells you it achieves a high AI citation rate. Ask three questions and the claim usually falls apart:
We answer all three in writing, in every report. That is the whole differentiator — not a bigger number, a checkable one.
Most agencies claim AI visibility. Very few publish how they measure it. Here is our method, in full:
| Step | What we do |
|---|---|
| 1. Prompt set | We build 50–100 prompts a real buyer in your category would type. None of them contain your brand name — the point is to see whether the model brings you up unprompted. |
| 2. Engine runs | Each prompt is run against the AI engines in scope, on a fixed date, and the raw answers are stored so you can audit them later. |
| 3. Scoring | Mention Rate (% of answers naming your brand) · Citation Rate (% citing your domain) · Share of Model (your mentions ÷ all competitor mentions). |
| 4. Competitor board | The same run tells you which brands the model currently treats as the category leaders. |
| 5. Re-measure | Same prompt set, same engines, 90 days later. Movement is comparable because nothing about the test changed. |
What we do not claim: engine coverage is stated explicitly in every report. A measurement against one model is not a measurement of Google’s AI Overviews, and we will never present one as the other.
Hà Nội, Vietnam. We work with clients across global markets and operate in English and Vietnamese.
Yes. Engagements run remotely with scheduled reviews in your timezone. Reporting, contracts and deliverables are in English.
We publish the method: prompt set, engines covered, run date, and the raw answers behind every number. You can re-run and audit it.
No, and we would not trust anyone who does. No one controls a model’s output. What we commit to is a measured baseline, a documented programme, and re-measurement on the same prompt set so progress is provable.
Find out how visible your brand is in AI answers.
Send your domain and your three closest competitors. We run the baseline and send back your Mention Rate, Citation Rate and where you sit against them.
Đội ngũ chuyên gia Vidco Group sẵn sàng đồng hành cùng bạn
Bước 1 / 4
Chúng tôi sẽ liên hệ trong vòng 2 giờ làm việc.