SEO remains part of AI search optimization because many AI answers use information retrieved through search. The additional work is to evaluate the generated answer: which brands it names, which sources it cites, and whether its claims are accurate.
Search visibility and recommendations
The available OppAlerts report shows an association between search engine appearances and its LLM recommendation score. That is an observational comparison. It does not establish that a search position causes an AI recommendation.Available report data. Names the snapshot, fields, sample sizes, and limits of verification.
A strong search position can help a page be discovered. It does not guarantee a brand recommendation.
These comparisons show an association between search visibility and recommendations. They do not establish that every recommended brand must rank in Google or explain why an individual brand was omitted.
What transfers
Google says its existing SEO requirements apply to AI Overviews and AI Mode. That is guidance for Google’s products. Check other providers’ access requirements separately. The table below relates familiar SEO tasks to AI search work.Google's "AI features and your website" guidance, which points AI-features optimization back at standard Search best practices.
| Classic SEO activity | AI-search counterpart | What changes |
|---|---|---|
| Keyword research | Prompt and topic research | Customer questions and requirements supplement keyword research; see Prompt and Topic Research |
| Technical SEO | Crawler access for search bots plus AI bots | Same discipline, more user agents; see Technical Accessibility for AI Crawlers |
| On-page content | Writing for extraction | Passages must stand alone and answer directly; see Writing for AI Search |
| Structured data | Structured data, unchanged in markup | Use depends on documented platform support; see Schema and Structured Data for AI |
| Link building | Links plus unlinked brand mentions | Check the measured signal and its sample; see Backlinks and AI Search |
| Entity and knowledge-panel work | Consistent business information | Makes business identity and facts explicit; see Clear and Consistent Business Information |
| Rank tracking | Sampled share-of-answer measurement | Repeated runs and distributions replace positions; see Measuring AI Search Visibility |
The GEO paper tested content changes such as adding sources, quotations, and statistics. Its reported improvements varied by experiment and subject. Use the findings as research evidence rather than a promised improvement for a client."GEO: Generative Engine Optimization" (Aggarwal et al.), benchmarked on GEO-bench. The reported maximum improvement is specific to the paper’s methods and evaluation conditions.
What needs a different method
Generated answers can vary across repeated requests. Measure the frequency of a mention or recommendation across a documented sample. Keep the exact prompt, product, model, and test conditions in the report.
A product may generate several related searches from one prompt. Where logs expose those searches, record them. Where they do not, distinguish suggested query variants from executed queries. See Retrieval-Augmented Generation.
Buyer requirements can change the recommended brands. Test relevant customer segments separately. A broad category average can conceal differences between budgets, locations, or intended uses.Research sources and limitations. Identifies the available report data and the analysis still needed to support earlier draft figures.
An answer generated without retrieval depends on learned model parameters and supplied context. Publishing a page does not directly update those parameters. A searched answer can use newer information if the page is retrieved.
Changes to measurement and planning
Adjust the measurement method before changing the publishing plan. Repeat prompts, record answer accuracy, and separate search-enabled tests from tests without search. Use keyword research alongside customer questions. Choose a checking schedule that supports actual decisions.
Adapting existing SEO services
Existing services still apply: technical access checks, useful content, site organization, supported structured data, accurate business information, and relevant independent coverage.
Add prompt research, repeated answer sampling, citation review, and comparisons by customer requirement. Define those deliverables precisely so clients can distinguish completed work from measured changes in visibility.