Every check follows the same three phases. Each one builds on the last.

Align it. Does your website tell AI who you are? Get concrete fixes to close the gap before you spend a single credit measuring.

Ask it. Does AI recommend you when your buyer asks? See every source that tipped the answer.

Track it. Watch your recommendation rate climb week over week. Run the same profile over time to measure the impact of what you build.

Enter your URL and Programate Lift goes to work. The tool reads your live site and builds the foundation of your workspace: the categories you compete in, the buyer personas relevant to your business, and the questions those personas would ask AI. You start from what the tool found, not from a blank form.

The more thoughtful and accurate your inputs, the more specific and useful your results. The tool gives you a starting point. You make it precise.

When AI reads your site, does it see who you say you are?

Alignment is not a check you write questions for, and it does not use your Discovery or Reputation scores. It compares your setup itself, your library identity, categories, personas, and goals, against what an AI actually extracts from your live site. Anything AI cannot find on your site is invisible to every downstream check. Fix gaps here before spending credits measuring.

What is actually happening. Programate Lift sends an AI reader to your live site and compares what it extracts against the identity you have defined in your library. Any field AI cannot locate is returned as a specific gap with a recommended fix.

Once you have reviewed your alignment check and are satisfied with what AI reads on your site, you can run a Discovery check, a Reputation check, or both.

Discovery. Does AI recommend you, unprompted? Discovery probes ask your buyer’s questions without naming you, then measure how often AI puts you forward. Run a Discovery check to score it. An Advanced check allows you to check individual persona questions against a selected group of AI models to get even more precise insight.

Reputation. What AI says when buyers ask about you directly. Check your AI reputation by completing your library against your target category: the market category where your business competes.

The first three steps live inside your workspace. Programate Lift builds them from your URL. Your job is to review what the tool extracted and make it more precise.

1. Build your customer personas.

Programate Lift surfaces the buyer personas it identified from your site. Review them. Refine them to reflect the specific person who would buy from you: who they are, what they care about, and how they search. The more accurate your persona, the more accurately the check reflects the real conversation your buyer is having with AI right now.

What is actually happening. This becomes what Programate Lift uses to construct the buyer conversation. It shapes the question the AI receives and determines how the engine interprets the search intent.

2. Define your business.

Review the target the tool identified from your site: the one brand, product, or service this workspace is measured against. Refine it to reflect the specific product, service, or brand you want AI to recommend. This is what Programate Lift measures visibility for throughout the run.

What is actually happening. Programate Lift uses this definition to evaluate whether your brand appears in the AI answer and how prominently it is positioned relative to any alternatives the AI names.

3. Give them a goal.

Review the questions the tool surfaced. Refine them, or write the real question your customer would ask an AI engine when they are looking for a solution. Not a keyword. The actual question a real buyer would type or speak.

What is actually happening. This question is the starting prompt for the AI conversation. The specificity of the question directly affects how the AI engine interprets buyer intent and which sources it consults.

Once your workspace is set, the tool runs the conversation and returns your results.

4. We run it against AI engines.

Programate Lift runs your buyer’s conversation across AI engines, measures whether your brand appears and how prominently, and captures the full response for analysis. Use advanced functionality to pick specific engines: OpenAI ChatGPT, Anthropic Claude, Google Gemini, Google AI Overviews, Perplexity Sonar, and xAI Grok.

What is actually happening. Each engine has a different knowledge base, citation behavior, and response pattern. The tool submits your persona’s question, records the response, and evaluates brand presence, position, and citation confidence, engine by engine.

5. See every question the AI asked on its way to the answer.

Before an AI engine gives a recommendation, it generates a series of fan-out queries as it works toward an answer. Programate Lift surfaces all of them. This is where you see how the engine thinks about your category, what it considers relevant, and what it is trying to resolve before it names a brand.

What is actually happening. AI engines fan out into further queries to arrive at a recommendation. Programate Lift captures and surfaces these intermediate queries, giving you a window into the decision logic that most tools never expose.

6. Build your page strategy.

Programate Lift identifies the content gaps between what AI needs to recommend your brand and what your site currently provides. The output is a specific list of pages to build, each with a clear rationale tied to the evidence from the run. Not a content strategy concept. Specific pages, specific reasons.

What is actually happening. The tool analyzes what the AI asked during the run, which sources it consulted, and what those sources contain that your site does not. The gap between them becomes your page list.

7. Find the deciding sources.

See which external sites were critical to the AI recommendation: the review sites, directories, publications, and forums the AI engine consulted before it named a brand. Programate Lift shows you which ones are within reach and what to do with each one.

What is actually happening. Programate Lift traces the AI engine’s citation behavior during the run to identify which external domains influenced the recommendation, ranked by their apparent influence on the outcome.

The alignment check tells you what AI sees about your brand from reading your URL, and what to change if that is not what you intend. The run tells you whether you are being recommended. The page strategy and deciding sources tell you what to build and where to appear.

On your owned pages you have the most control over your message. They are the best way to connect with buyers and differentiate your brand to both buyers and AI, and the quickest way to change market perception in the age of AI search. Getting listed on deciding sources takes longer but compounds: each new source you appear on feeds future AI recommendations.