Documentation
Brand and monitoring setup
Launch GEO Scout in a few minutes: onboarding, prompts, competitors, and your first monitoring run
Open in dashboardRun onboarding and pick prompts for your first run
Onboarding sets up GEO Scout in three steps: enter your site, review the brand data AI gathered automatically, and pick up to three prompts for a trial run. The whole flow takes a few minutes — then the system checks your visibility in ChatGPT, Gemini, and Yandex with Alice on its own.
- Step 1 — enter your site URL. GEO Scout parses the homepage, detects the brand, and shows what it found so you can confirm it's the right domain
- Step 2 — review the data. AI has already detected the brand name, pulled a description, and prepared business context. Fix anything off or leave as is — the sharper the entities, the better the recommendations and reports
- Step 3 — pick up to three prompts for the trial run. Choose from the generated list or add your own. Narrow, niche prompts surface more accurate competitors than broad queries
- Once onboarding is done, GEO Scout sends the prompts to ChatGPT, Gemini, and Yandex with Alice. The first visibility and competitor results appear in the dashboard within a few minutes
- The trial run is free: 3 prompts every 7 days. That's enough to see your first visibility patterns and learn which AI providers surface you and which skip you — before you upgrade to a paid plan
Start with prompts your customers would actually use to find you. A broad query like "freelance marketplace" creates noise in the competitor list; a narrow one ("where to find a freelancer for a WordPress online store") surfaces direct competitors and clean signals you can act on.
Set up prompts and clusters for monitoring
Monitoring → Prompts is where you manage everything measured each day: the prompt list, clusters, provider slots, cluster rotation, and group geography. Your plan sets the daily run budget — you choose how to spread it across topics and providers.
- Your plan sets the daily run budget. The top panel shows total runs, runs used today, and the response rate per provider. Slot cards above the list set which providers run on each cycle day, while each topic run day is selected from the cluster header — see "Provider and cluster rotation" below
- Split prompts into clusters by meaning: for example, "Finding web-development contractors" and "Finding branding designers" — two different directions. On the dashboard and in reports, metrics — visibility, Share of Voice, citation share — are computed separately per cluster
- Assign each cluster a color. Useful when you have 5–10 directions: the cluster is recognizable in tables and charts without reading the name
- A cluster can have its own geography, separate from the brand: inherit from brand, local (country + city), regional, country-wide, or international. Useful for geo-dependent queries like "find a freelancer in St Petersburg" or "buy in Berlin"
- Cluster geography overrides brand settings for every prompt in the group. This lets you run one brand with multiple geo directions in one workspace — no need to create a separate brand per country or city
If your prompts mix intents — commercial and informational — split them into separate clusters. You'll instantly see where the brand wins on "buy / order" and where it wins on educational topics, and you can optimize each independently.
Rotate providers and clusters by day — stretch your budget
If your daily budget is tight, use slots to rotate providers and topic clusters. One day ChatGPT can run one set of topics, the next day Claude can run another. Clusters left in "Every slot" still run daily; clusters pinned to a slot run only on that slot's day.
- A slot is one day in the rotation cycle. Add 2 slots and the cycle runs for 2 days, alternating the selected providers and slot-specific clusters automatically. Today's slot is tagged with a "Today" badge. Slot transitions happen near UTC midnight
- Use the "Every slot / Slot N" control in a cluster header. "Every slot" means every prompt in the cluster runs daily; choosing a specific slot pins that cluster to the slot's day. Slot cards show how many prompts will run that day
- Daily budget cost is set by the heaviest day: prompts from clusters running in that slot × providers in that slot. Splitting clusters across slots can lower the daily peak without deactivating prompts
- A single slot can hold several providers — they all run on that slot's day. Convenient for grouping by type: a "Russian" slot (Yandex, Alice) and a "Western" slot (ChatGPT, Claude, Gemini), switching every other day
- On the free plan, the provider card shows the fixed ChatGPT, Gemini, and Yandex run; provider and cluster rotation can be customized on paid plans. Partner accounts keep the old behavior: unlimited budget, no rotation needed
Keep critical clusters in "Every slot" when you need daily continuity. Move exploratory or lower-priority topics into separate slots when you need to cover more models and directions within the same cap.
Content language and Telegram reports
Two useful options in Settings help you tune GEO Scout to your team and process: the language of AI-generated content and connecting a Telegram bot for weekly visibility reports.
- In brand settings, pick the content language — Russian or English. GEO Scout uses it for every AI-generated artifact: reports, articles, recommendations, and briefs. This doesn't change the language of the AI providers' answers — only your output documents
- Content language is independent of the interface. You can work in a Russian UI and receive English recommendations — handy if your market is international or your team is distributed
- In the Notifications tab, connect a Telegram bot to receive the weekly visibility report directly in your messenger — no need to open the dashboard. Click "Generate code" and send it to the bot to link the brand
- Each brand is linked separately. The weekly report covers the key changes: Share of Voice and citation shifts, prompts that rose or dropped, competitors gaining ground — all in a compact format for a quick scan
Content language is a decision for the team, not for your site visitors. If your market is English-speaking but the team prefers a Russian UI, set English for content — the AI will write recommendations for your audience while you stay in your preferred locale.
Name variations and additional domains
GEO Scout uses AI to identify your brand and competitors in AI responses automatically — no extra setup needed. Add manual name variations and additional domains only for non-obvious cases: rebrands, alternative transliterations, multiple sites.
- Name variations are only needed in non-obvious cases — when AI detection might not link different spellings to the same brand. Typical triggers: a rebrand (e.g. T-Bank / Tinkoff), non-standard transliteration, or internal abbreviations. Ordinary case variants and obvious spellings are handled by the system
- Without variations, AI can split your SoV between two names, leaving you with artificially low numbers. Once configured, Share of Voice and citation share are computed correctly across every variant at once
- Additional domains. If the brand runs on several sites — main and regional, primary and a blog on a separate domain, landing pages — add them all. A citation of any one domain counts toward your Citation Share
- Especially important for holdings, media groups, and post-M&A brands: one brand in GEO Scout stays unified in reports, regardless of how many legal entities and domains sit behind it
After a rebrand, keep the old name in the variations list for at least a year. LLMs train on historical data and keep mentioning the brand by its previous name long after. Remove the variation — lose those mentions and a real chunk of your Share of Voice.
Invite teammates to work on the brand together
In Settings → Team the brand owner can invite teammates by email — each gets their own access to the dashboard, monitoring, and settings, while billing and brand removal stay with the owner. Available on every paid plan.
- Invite anyone by email. They receive a link; if they don't have a GEO Scout account yet, one is created automatically — no need to share credentials in advance
- Members see the same as the owner: dashboard, reports, monitorings, competitors, sources, AI traffic. They can edit prompts, competitors, and brand settings — full-fledged work with the data
- Billing and brand removal stay owner-only: the team can't accidentally change the plan or delete the brand. Managing the team — invitations and access revocation — is also reserved for the owner
- Revoke access at any time from the same Team tab: one click and the member loses access. A pending invitation link can be cancelled the same way
If your team works across several brands in different workspaces, invite people surgically — only into the brands where they actually contribute. Credits and the subscription are scoped to the brand, not the user, so adding a teammate doesn't increase your bill.
Launch monitoring and start the full cycle
Once your plan is paid, the monitoring page shows a "Ready to launch" banner. Double-check your prompts and clusters, click "Launch monitoring" — and GEO Scout starts the daily cycle of observation, diagnosis, change, and verification.
- The "Ready to launch" banner appears automatically once the subscription is active. Two buttons: "Go to prompts" to review setup once more, and "Launch monitoring" to start right now
- After launch, GEO Scout runs your prompts through the selected AI providers daily within your daily budget. Visibility, Share of Voice, and citation metrics refresh by the next morning
- From day one, history starts building: metric dynamics by day, week, and month; period comparisons; the measured effect of your site updates and external publications
- This is the start of the full visibility cycle. From here on: weekly reports, data-backed recommendations, an action plan for the site and external sources — all in one platform
Don't rush to optimize right after launch — give the system 3–5 days to accumulate data. More observations means sharper signals about real patterns and weak spots, not random noise from one or two AI answers.