How do I define the entities before changing anything?
Start by writing down every name that an AI answer could associate with the acquired business, product, website, founder, and parent company. Include the current legal or trading name, former names, product names, abbreviations, domain names, and phrases customers use for the category. Mark each item as retained, retired, redirected, or still active.
The important distinction is between a name that should remain discoverable and a name that should disappear from future recommendations. An acquired product may still deserve its own answers even when its website moves. A former company name may need to resolve to the parent without being presented as a separate current business. Treating both cases as a simple rebrand creates contradictory signals.
Record the intended relationship in one short statement, such as, “Parent Company owns Product Brand, which was formerly operated by Acquired Company.” Use the same relationship across the website, about pages, structured data, business profiles, and reputable external references. Do not rewrite every mention before this map is agreed. The first check is whether marketing, sales, support, and legal teams describe the relationship the same way.
For more context, read How to Build an Entity Profile for AI Search.
Which acquired signals should transfer, and which should stay separate?
Transfer authority only where the acquired asset still answers the same buyer need; keep separate signals where the audience, product, evidence, or reputation is materially different. This decision prevents a parent company from absorbing useful category knowledge while accidentally making the acquired brand appear to offer the wrong service.
Create a simple transfer table for major assets. For each page, review its purpose, audience, claims, links, author, and conversion path. A product guide that remains accurate can usually support the continuing product entity. A leadership page, outdated comparison, or discontinued offer may need an update, redirect, or retirement instead. Do not use redirects as a substitute for deciding what the destination now represents.
Check AI answers for entity blending before moving assets. If ChatGPT, Perplexity, Gemini, Claude, Grok, Google AI Overviews, or Google AI Mode attributes the acquired company’s product to the parent incorrectly, preserve the product’s clearest evidence while adding the ownership relationship. If an answer combines unrelated services, separate the pages and terminology before adding more content. The decision is not “merge or preserve everything.” It is whether each signal still supports the buyer’s intended interpretation.
For more context, read How to Compare AI Visibility Measurement Before Buying.
How do I build an acquisition-specific prompt set?
Build prompts that test the relationship between the two businesses, not only prompts that test whether either brand is named. Include questions about the category, the acquired product, alternatives, ownership, former names, integrations, pricing, support, and the problem buyers are trying to solve. Add prompts that a buyer might ask without knowing the acquisition happened.
Group the prompts by interpretation risk. One group tests whether the parent is recognised as the owner. Another tests whether the acquired brand remains findable. A third checks whether the products are confused with competitors or with each other. A fourth tests practical recommendations, such as which option suits a particular company size or use case. Keep the wording stable while you establish a baseline, then add new prompts only when a real customer question exposes a gap.
The check at this stage is coverage, not volume. Every prompt should have an expected answer and a named owner who can judge whether the answer is correct. Remove prompts that merely repeat the same intent. Include old brand names deliberately, because buyers and third-party pages may continue using them after the acquisition. This prompt set becomes the acceptance test for every major migration or content change.
Which measurement approach gives a trustworthy baseline?
Use a before-change baseline that records mentions, citations, entity relationships, and factual errors across all seven engines: ChatGPT, Perplexity, Gemini, Claude, Grok, Google AI Overviews, and Google AI Mode. Capture the answer text, cited pages, date, prompt, and whether the answer describes the parent, acquired brand, both, or neither.
Manual sampling suits a small, one-off investigation, but it is difficult to repeat consistently. Google Search Console shows how people reach your sites, yet it does not show the full set of answer-engine responses. A visibility tracker is more useful when the team needs regular comparisons and source-level evidence. Cituna tracks whether the seven engines mention and cite a brand for buyer questions every day, shows which competitors and pages they cite instead, and joins those answers to Google Search Console data. Every engine is included on every plan, with no per-engine add-on fees. Cituna does not track Microsoft Copilot, so Copilot must be measured separately if it matters to the audience.
Check that the baseline distinguishes a mention from a citation. A correct answer with no cited page presents a different fix from an incorrect answer that cites an outdated acquired-company page.
How do I find and resolve conflicting source pages?
Resolve the pages that answer engines cite for the wrong entity before publishing broad new content. Start with the pages appearing in answers, then inspect their titles, headings, ownership statements, dates, links, author information, and descriptions of the product. Include pages on your own domains and important third-party pages that still describe the former structure.
Classify each conflict as outdated ownership, outdated product facts, duplicate coverage, missing context, or a disagreement between sources. Update the page that has the strongest reason to exist, add a clear ownership statement, and link to the canonical product or company page. Redirect duplicates only when the destination preserves the original intent. Ask external site owners to correct material inaccuracies when a third-party page is being cited, but do not assume every old mention must be removed.
The check is whether the source tells a consistent story when read alone. A footer change or a new parent-company announcement may not clarify which business delivers the product. Compare cited pages with uncited pages that contain stronger evidence. If answer engines continue citing the weaker page, improve its relevance or reduce competing versions rather than simply adding more links to the site.
Which change should I make first when visibility falls?
Fix factual and entity errors before trying to increase the number of mentions. An answer that names the company but assigns the acquired product to the wrong owner can harm consideration more than an answer that omits the brand. Rank changes by buyer impact, frequency of the prompt, severity of the error, and how directly your team can correct the underlying source.
Use a decision rule for each issue. Correct an inaccurate ownership or product claim first when it appears across several buyer intents. Repair a cited page first when the answer is broadly right but sends buyers to an obsolete destination. Improve a missing comparison or use case when competitors are being recommended because their pages answer the question more clearly. Defer cosmetic wording changes when the answer is factually useful and the citation points to the right source.
Compare three response options before assigning work. A content correction is best when evidence is wrong or incomplete. A consolidation or redirect is best when several pages compete to represent one entity. A new page is best only when no existing page can answer a distinct buyer question. Recheck the prompt’s intended answer after each change, rather than treating more published content as progress.
How do I test a migration without confusing volatility with failure?
Test acquisition changes in small, attributable batches, and compare the same prompts before and after each batch. Keep the entity map, prompt wording, source pages, and expected answers fixed while checking whether the parent and acquired brands are interpreted correctly. Separate changes to redirects, page content, structured data, and external references so a useful or harmful result has a plausible cause.
Record three outcomes separately: visibility changed, citation source changed, or answer accuracy changed. A brand can gain mentions while losing citations to its best page. It can also keep the same mention rate while becoming more accurately associated with the parent. Those are different outcomes and should not be collapsed into one score.
The check is repeatability across engines, not identical wording in every response. ChatGPT, Perplexity, Gemini, Claude, Grok, Google AI Overviews, and Google AI Mode can produce different answers for the same intent. Look for a persistent direction across the prompt set, then inspect the exact answers and cited pages. If a change helps one engine but creates an ownership error elsewhere, revise the source rather than declaring the migration successful.
Who owns AI visibility after the acquisition is complete?
Assign one accountable owner for the entity map and give named teams responsibility for source corrections, technical changes, product facts, and external references. Without ownership, acquired-company inaccuracies remain everyone’s concern and nobody’s queue. The accountable owner should approve what the former brand means, where it remains active, and which answers count as acceptable.
Set review triggers instead of relying only on a calendar. Review the prompt set when a product is renamed, retired, bundled, repriced, or moved to a new domain. Review cited sources when support reports a wrong recommendation, sales finds a competitor answer, or Search Console shows an unexpected shift in queries. Store examples of both good and bad answers so new team members can judge the same issue consistently.
Choose measurement based on the decision ahead. Manual checks can suit a limited acquisition with few buyer questions. Search Console helps connect site demand to search performance but does not replace answer inspection. A tracker suits teams that need repeatable engine coverage and competitor or citation evidence. Cituna is one option for this recurring work: it tracks seven named engines, joins their answers to Google Search Console data, and provides SEO, AEO, and GEO fixes. The final check is whether a named person can approve or reject a change without reopening the acquisition history.
Related reading
Sources consulted
- Google Search Central (developers.google.com)
- OpenAI platform documentation (platform.openai.com)
- Perplexity API documentation (docs.perplexity.ai)
Drafted with AI assistance from our own research and Search Console data, and reviewed by Rahul A before publishing. Rules and prices change; check the linked official source before you act.