Setup Wizard
Walk through the guided onboarding wizard and the seven-step project setup path — define your brand, choose AI models and surfaces, add prompts, and capture your Share-of-Model baseline.
Overview
AEO Optima has two guided setup surfaces that share the same underlying configuration:
- The onboarding wizard (
/onboarding) — an eight-step flow every new user goes through on first sign-in, from creating an organization to capturing a first baseline. It also runs when you create an additional organization. - The Setup hub — a seven-step checklist for any existing project, reachable from the Setup Wizard link in the sidebar. The same seven steps appear on the Dashboard's Setup Progress card and in the per-page setup stepper, so every surface agrees on what's done and what's next.
You can complete either flow in a single sitting or return later — progress is tracked, and every step can be skipped and finished afterwards.
The Onboarding Wizard (8 Steps)
New users are taken through this wizard on first sign-in. It follows the canonical AEO route: definition comes before prompts, because an AI engine cannot recommend you for something you have not first defined yourself into.
| Step | What Happens |
|---|---|
| 1. Organization | Create a new organization, or accept a pending invitation / request to join an existing one. |
| 2. Project | Name your first project and enter your brand name and website URL. |
| 3. Define Your Brand | Establish who you are, what you do, how you qualify, and whom you serve (details below). |
| 4. Prompts | Add tracked prompts manually or import them from CSV. |
| 5. AI Models & Surfaces | Pick the chat models and search-embedded AI surfaces (Google AI Overview, Google AI Mode) to capture. |
| 6. Configure | Brand facts, competitors, and your recurring capture schedule preferences. |
| 7. Capture Your Baseline | Run your first real capture — this becomes your Share-of-Model baseline. |
| 8. All Set | A summary of what you configured, plus a website AI-readiness check. |
Step 3: Define Your Brand
This step has two paths:
- Automatic (plans with Query Universe — Pro-Individual and above — and a website URL): the platform reads a bounded set of your own site's pages (homepage plus the about/products/services pages it discovers, up to 6 pages) and extracts your project definition. You review the extracted items — grouped into what you do, how you qualify (gates such as certifications), whom you serve, and competitors, each carrying a confidence level and, where available, a verbatim quote from your own site as evidence — and apply only the ones you confirm.
- Manual (all plans; the only path on Free and Starter): a compact form capturing your industry, category, certifications (treated as gates — the pass/fail criteria engines filter on), target segments, and competitors. What you enter here pre-fills the brand facts and competitor rows in the Configure step.
Skipping is always allowed — definition improves everything downstream but never blocks setup.
Step 4: Prompts
Add prompts one at a time or import a batch from CSV (a downloadable template is provided). On plans with Query Universe, once the Define step has applied a confirmed definition, a "Suggested from your definition" card offers prompts composed from your building blocks — select the ones you want and add them in one click. Suggested prompts keep their building-block links, so they count toward coverage exactly like prompts added from the Query Universe page.
Step 5: AI Models & Surfaces
Choose which chat models (ChatGPT, Claude, Gemini, Perplexity, and more from the registry) and which search-embedded AI surfaces to capture. Google AI Overview and Google AI Mode are captured natively by AEO Optima's own browser — no connector needed — and the step is shown on every plan; surfaces your plan doesn't include yet appear with an inline upgrade prompt. Bing Copilot additionally requires the DataForSEO connector. Selections save automatically, and you can change them anytime in Settings → AI Surfaces.
New projects come pre-configured with one recommended chat model per provider (capped at your plan's daily snapshot limit, flagship providers first), so there is always something sensible selected. Native surfaces are never pre-selected — you opt into them here.
Step 7: Capture Your Baseline
The wizard runs a real bulk capture across your active prompts and selected models, with live progress. This first capture is framed as your Share-of-Model baseline — the reference point all later trend data compares against.
Your recurring capture schedule (configured in step 6) is created when you finish this step:
- If you captured a baseline, the schedule does not immediately capture again — its first automatic run follows the cadence you chose.
- If you skipped the baseline, one first capture is queued automatically so your project doesn't sit empty until the schedule's first cycle.
Step 8: All Set
Alongside the setup summary, an AI readiness card checks your website's crawler policy — whether robots.txt was found, whether any AI bots are blocked, and whether a sitemap is declared. If it isn't crawled, it can't be cited. The card links to a full GEO Audit for the deeper nine-category check.
The Setup Hub (7 Steps)
For an existing project, the Setup page in the sidebar tracks a canonical seven-step path. Each step shows its completion state and links directly to the page where you finish it.
| Step | What You Configure | Why It Matters |
|---|---|---|
| 1. Business Profile | Brand name, industry, category, website URL, description | Brand name is used for mention detection in AI responses. Website URL is used for citation matching. Industry and category feed the criteria engine and Query Universe. |
| 2. Define Your Brand | Who you are, what you do, how you qualify, whom you serve | The AEO foundation — an engine cannot recommend you for a category you have not defined yourself into. Runs the same automatic discovery flow as the onboarding wizard's Define step. |
| 3. AI Models & Surfaces | Chat models plus search-embedded AI (Google AI Overview, AI Mode) | Different AI models produce different answers, and search-embedded AI reaches users who never open a chatbot. Monitoring multiple surfaces gives you a complete picture. |
| 4. Locations | Country and language pairs | AI responses vary by region and language. Location settings ensure you capture responses relevant to your target markets. |
| 5. Prompts | Questions to ask AI models | The core of your tracking setup. The quality of your prompts directly determines the quality of your visibility insights. |
| 6. Capture Schedule | A recurring automatic capture | Schedules make analytics build themselves — no one has to remember to hit "Capture". |
| 7. Capture Your Baseline | Your first snapshot capture | Setup isn't done until the project has produced data. A configured project with no captures shows empty analytics. |
Note: API keys, themes, and tags are managed in Settings — they are refinements, not blockers for your first capture. See API Keys for the 3-tier routing system (Platform API, gateway BYOK, direct provider BYOK).
Step 1: Business Profile
The foundation of your project. The information you enter here determines how AEO Optima identifies your brand in AI-generated responses.
- Brand Name — Enter the exact name you want to track. If your brand is commonly referred to by multiple names (e.g., "AWS" and "Amazon Web Services"), use the most common variation here.
- Industry and Category — Where you operate and what you offer. These feed the criteria engine, Query Universe, and prompt suggestions.
- Website URL — Your primary website. Used to detect when AI models cite or link to your content.
- Description — A brief description of your business, providing context for analysis.
Step 2: Define Your Brand
Opens the discovery flow (see Discovery Autopilot below). On plans with Query Universe, the platform analyzes your own website and proposes evidence-backed definition items for review; on other plans you define the same things by hand. The step is marked complete once your project has business context (industry/category) or any building blocks.
Step 3: AI Models & Surfaces
Select which AI models AEO Optima should query when capturing snapshots — models are loaded dynamically from the platform's model registry and grouped by provider — plus the search-embedded AI surfaces (Google AI Overview and AI Mode, captured natively; Bing Copilot via the DataForSEO connector).
Each enabled model is queried with every prompt during a snapshot capture. Enabling multiple models gives you cross-engine comparisons so you can see where your brand is strong and where it is missing. At minimum, enable two models from different providers.
See LLM Configuration for the full list of supported models and recommendations.
Step 4: Locations
Configure the geographic regions and languages for your AI tracking. AI models can produce different responses depending on the user's country and language context. If you operate in multiple markets, adding your primary locations here ensures you capture region-specific responses.
See Locations for details on supported regions and languages.
Step 5: Prompts
Prompts are the questions AEO Optima sends to AI models on your behalf. Think of them as the queries your potential customers are typing into ChatGPT, Claude, Gemini, and Perplexity.
The quality of your prompts is the single biggest factor in the quality of your insights. Broad, customer-oriented questions produce the most actionable data.
Examples of effective prompts:
- "What is the best [your product category] for [your target audience]?"
- "Can you recommend a [your product type]?"
- "How does [your brand] compare to [competitor]?"
- "What are the top [your category] tools in 2026?"
Tip: Start with 3 to 5 broad, category-level prompts. You can always add more specific prompts later as you learn which angles matter most for your brand.
See Prompts & Rankings for a deeper guide on prompt strategy.
Step 6: Capture Schedule
Choose whether to run a recurring schedule (daily, weekly, biweekly, or monthly), with the estimated cost per run and per month shown beside the choice. Nothing runs unless you turn it on: the schedule is off by default, and creating one never starts a capture on its own — the first capture is the explicit baseline step that follows. See Schedules.
Step 7: Capture Your Baseline
Run your first capture from the wizard's baseline step or from Snapshots → New Snapshot. This is the only way a first capture starts — a schedule on its own waits for its first due time. Setup is complete once the project has produced its first snapshot.
Discovery Autopilot — the automatic Define path
Discovery Autopilot powers the automatic path of the Define step in both flows. Paste (or confirm) your company name and website URL, and the platform reads a bounded set of your own site's pages (homepage plus the about/products/services pages it discovers, up to 6 pages) and extracts the raw material of your project definition: who you are (brand and parent brand), what you do (industry, category, sub-category), how you qualify (certifications), whom you serve (segments), and the competitors your site names.
You then review the extracted items — each carries a confidence level and, where available, a verbatim quote from your own site as evidence; items without a verifiable quote are forced to low confidence — and apply only the ones you confirm. The confirmed items seed the project in one step: orbit identity, Query Universe building blocks, category/industry settings, and competitors. Declining everything writes nothing.
Discovery Autopilot runs as the recommended path in the onboarding wizard's Define step and is available any time from Setup → Define Your Brand. It requires the Query Universe feature (Pro-Individual and above); Free and Starter plans define manually instead — via the onboarding wizard's manual form or the guided Orbit Wizard. Each run counts as one AI analysis against your daily allowance.
Returning to Setup
You can return to the Setup hub at any time from the Setup Wizard link in the sidebar or from the Dashboard's Setup Progress card. Changes take effect on the next snapshot capture.