Answer engine: definition and how it works

An answer engine takes a question in plain language and returns a written answer. It builds that answer from pages it retrieves, from what its model learned in training, or from both.

A search engine returns a ranked list and you do the reading. An answer engine does the reading and hands you the summary, usually with a few cited sources.

When the engine searches the web first, the answer is grounded in retrieved pages and can cite them. When it skips search, the answer comes from the model's training data, frozen at a knowledge cutoff. For a side-by-side view of what changes for publishers, see answer engines vs search engines.

Most answer engines follow the same five steps. The details differ by engine and change often.

1. Decide whether to search. Retrieval is decided per query. ChatGPT search searches at the model's discretion, and Google says AI Overviews do not appear for every query.
2. Rewrite the query. OpenAI says ChatGPT search typically rewrites your query into one or more targeted queries. Google describes a query fan-out technique that issues multiple related searches across subtopics.
3. Retrieve candidates. The engine pulls matching pages from its index or a partner's index.
4. Select and order. The engine narrows the candidates and decides which ones go into the model's context, and in what order.
5. Generate and cite. The model writes the answer and attaches citations. OpenAI's API returns the URLs consulted separately from the URLs cited, and it says the number of sources is often greater than the number of citations.

What a publisher can influence

• Crawler access. robots.txt rules are per bot. OpenAI's crawler page says OAI-SearchBot controls ChatGPT search and GPTBot controls training, and each setting is independent.
• Rendering. Vercel found that ChatGPT and Claude crawlers fetch JavaScript files but don't execute them. Content that only appears after JavaScript runs may be invisible to them.
• Eligibility. Google says a page must be indexed and eligible to show with a snippet to appear as a supporting link in AI Overviews or AI Mode.

Whether search fires, which queries run, how candidates are ordered and what the model writes all sit outside your control.

A page only reaches an answer if it passes every step. Most failures happen early and silently. A blocked crawler, a firewall that returns 403 to a bot your robots.txt allows, or pricing that loads only through JavaScript means the page never becomes a candidate.

The next move is to check the access steps first: robots.txt per bot, raw HTML before JavaScript, and index status in each engine's webmaster tools. Then look at which other sites the engines cite for your topic. Being named on those sites is associated with appearing in answers, which is the idea behind PR for AI answers. The free checker on this site shows whether engines name you for up to three buyer questions.

• "Optimize for the exact prompts customers type." Engines rewrite prompts into their own queries, so any prompt list is a proxy for an input you can't see.
• "All answer engines behave the same." Google says even AI Overviews and AI Mode may use different models and techniques. See whether tactics transfer across engines.
• "Special markup or files get you selected." Google says there are no additional requirements or special optimizations for AI Overviews or AI Mode. A critical survey of 45 GEO studies found no technique with a stable, cross-platform effect on discoverability.
• "Allow GPTBot so ChatGPT can find you." GPTBot governs training. OAI-SearchBot governs ChatGPT search.

• Grounding: tying an answer to retrieved sources instead of training data alone.
• Retrieval: the step where the engine fetches candidate pages.
• Reranking: reordering retrieved candidates before generation.
• The four stages of AI search: parametric, access, retrieval and generation, and who controls each.