A Whitespark analysis of businesses with multiple historical NAP versions found that those with unresolved legacy citations averaged 34% lower local pack visibility than comparable businesses with clean citation profiles. This finding underscores that historical NAP debris is not a theoretical concern. It is a measurable ranking liability that accumulates over time as outdated records persist in data aggregator systems and propagate to downstream directories. The cleanup workflow that resolves these issues must address root data sources in a specific sequence to prevent re-propagation of corrected errors.
The Four-Source Priority Hierarchy for Citation Audit and Correction
Effective cleanup follows a strict source hierarchy dictated by how data flows through the local search network. Correcting downstream sources before root aggregators wastes effort because aggregator feeds will overwrite downstream corrections during the next propagation cycle. The hierarchy establishes a clear sequence that prevents this rework.
Tier one: root data aggregators. The three primary U.S. data aggregators, Data Axle (formerly Infogroup), Neustar Localeze, and Foursquare (which absorbed Factual in 2020), feed business data to hundreds of downstream directories, GPS services, and mapping applications. Acxiom retired its directory services at the end of 2019, removing it from the active aggregator set. Each aggregator maintains its own submission portal where businesses can claim and update their records. For businesses with historical name or address changes, the aggregator records frequently still show the original data because the business never submitted an update after the change. Correcting these three records first ensures that the next propagation cycle pushes correct data to downstream platforms rather than continuing to distribute outdated information.
The propagation timeline varies by aggregator network. Some downstream partners update within days; others take two to eight weeks to fully reflect changes across all platforms. This lag means that even after correcting the aggregator records, incorrect downstream citations will persist temporarily. The staged cleanup approach accounts for this delay by pausing before addressing downstream sources.
Tier two: primary platforms. Google Business Profile, Apple Maps, Bing Places, Yelp, and Facebook each maintain independent data records that do not rely solely on aggregator feeds. These platforms accept direct edits that typically take effect within days. For businesses with historical name changes, legacy listings on these platforms may exist under the old business name, creating duplicate entity records that compete with the current listing for ranking signals. Claiming and correcting or removing these legacy listings on primary platforms is the second priority.
Tier three: industry-specific directories. Legal directories (Avvo, FindLaw), healthcare directories (Healthgrades, Zocdoc), home services directories (HomeAdvisor, Angi), and other vertical-specific platforms maintain their own databases that may include outdated business information. These directories vary widely in their update processes: some allow self-service edits, others require support tickets, and some have no functional update mechanism at all. Prioritize the directories that carry the most authority and traffic in the business’s specific vertical.
Tier four: general web mentions and secondary directories. Manta, Superpages, CitySearch, and dozens of smaller directories round out the citation landscape. These sources carry the least individual authority but contribute to entity confidence through volume. Address corrections on these platforms produce minimal individual ranking impact, but in aggregate, cleaning a substantial number of incorrect secondary citations shifts the balance of correct versus incorrect data in the reconciliation calculation.
How to Identify and Suppress Legacy Entity Records That Compete With Current Data
Businesses that changed names or addresses often have orphaned entity records in Google’s Knowledge Graph. These orphans are remnants of the old business identity that persist because enough citation data supports their existence. The orphaned entity competes with the current entity for ranking signals, effectively splitting prominence across two records rather than consolidating it into one.
Identification method. Search Google Maps for both the current business name and any previous names. If separate listings appear for the same physical business at different addresses or under different names, the legacy entity exists. Check whether the old listing has reviews, photos, or other user-generated content attached to it. Also search for the old business name on Google Search and check whether a Knowledge Panel appears for the old identity.
Suppression approach. If the legacy listing is unclaimed, claim it through the GBP claim process and then mark it as permanently closed or request removal through the GBP support form. If the legacy listing has been claimed by a previous owner or a third party, request ownership transfer through the GBP ownership resolution process. If the listing represents a genuinely defunct business identity (the business rebranded and the old name no longer exists), requesting removal with supporting documentation (new business license, articles of incorporation under the new name) typically succeeds.
Merge considerations. In some cases, Google’s support team can merge two entity records, combining the reviews and user-generated content from the old listing into the new one. This preserves the accumulated prominence signals while eliminating the competing entity. Merges are not always available or appropriate, particularly when the old and new businesses are legally distinct entities rather than a rebrand of the same entity. Requesting a merge when the entities are genuinely separate (e.g., a previous owner’s business versus the current owner’s new business at the same address) violates Google’s guidelines and risks complications for both listings.
The diagnostic indicator for success is the disappearance of the legacy listing from Google Maps results and the absence of a Knowledge Panel for the old business identity. If the legacy entity persists despite removal requests, the citation data supporting it may need further cleanup before Google’s system will remove the record.
The Staged Cleanup Timeline That Prevents Entity Reconciliation Disruption
Simultaneous updates across all citation sources create a temporal data conflict that can temporarily reduce entity confidence rather than improving it. The staged approach spaces corrections across intervals that allow Google’s reconciliation system to process each wave before encountering the next.
Weeks one through two: aggregator corrections. Submit updated NAP data to Data Axle, Neustar Localeze, and Foursquare. Each aggregator has its own claim and update process, typically requiring business verification. For businesses with historical name changes, the update should include the current legal business name. If the aggregator record shows an address that the business occupied years ago, correcting it at this stage prevents that outdated address from continuing to propagate.
Weeks three through six: wait for propagation. Do not make additional citation changes during this period. Monitor downstream directories to observe whether the aggregator corrections are propagating. Some platforms update within days; others may take the full six weeks. During this window, the mix of old and new data across the network is expected and does not indicate a problem. Making additional changes during propagation adds noise to the reconciliation process.
Weeks seven through eight: primary platform corrections. Update Apple Maps, Bing Places, Yelp, and Facebook directly. Claim and correct any legacy listings on these platforms. If the business changed its name, check whether Yelp has both the old-name and new-name listings and request a merge or removal of the old listing.
Weeks nine through twelve: secondary and industry-specific directories. Work through the industry-specific and general directories, prioritizing those with the highest authority in the business’s vertical. For platforms that do not accept direct edits, submit correction requests through their support channels. Document which platforms cannot be updated and monitor whether the volume of corrected citations from other sources is sufficient to outweigh the remaining incorrect ones.
Ongoing: quarterly audits. Schedule quarterly citation audits to catch new inconsistencies introduced by data aggregator refreshes, directory scraping, or third-party data services. A citation that was correct three months ago may be overwritten by an aggregator feed that reverted to old data. Persistent monitoring prevents the cleanup work from being undone by the same propagation mechanisms that created the original problem.
Handling Unupdatable Citations and Measuring Cleanup Effectiveness Through Entity Confidence Indicators
A realistic cleanup workflow acknowledges that some citations will remain incorrect indefinitely. Abandoned directories, platforms with broken claim processes, and sites that scrape data from outdated sources all create persistent incorrect citations that cannot be directly corrected.
Volume strategy. The entity reconciliation system resolves conflicts by weighing correct data against incorrect data, factoring in source authority. If the business has 60 correct citations from authoritative sources and 8 incorrect citations from low-authority abandoned directories, the reconciliation resolves in favor of the majority. Building additional correct citations on new authoritative platforms shifts the balance further, effectively drowning out the uncorrectable incorrect citations.
Google-specific suppression. For citations that display materially inaccurate information (wrong address that could misdirect customers, wrong phone number that connects to a different business), submitting a Google Search removal request or using the “Report a problem” feature on Google Maps for listings that reference the incorrect data may prompt Google to deprioritize those sources in its reconciliation calculation. This does not remove the citation from the source platform but may reduce its influence on Google’s entity confidence.
Legal takedown requests. In extreme cases where a platform displays grossly inaccurate business information and refuses to correct it, legal takedown requests under consumer protection or defamation frameworks may be appropriate. This approach is rarely cost-effective for minor directory citations but may be justified when the incorrect information causes material business harm (e.g., a medical practice listed at a residential address, creating confusion about legitimacy).
Acceptance threshold. Not every incorrect citation warrants intervention. If the diagnostic checks show a clean Knowledge Panel, no duplicate entities, and no GBP suggested edits, the remaining incorrect citations are within Google’s tolerance. The cleanup effort should stop when the entity confidence indicators stabilize, not when every last citation is perfect. Pursuing perfection across hundreds of directories produces diminishing returns well before the final citation is corrected.
After completing the staged cleanup, practitioners need to verify that the work produced measurable improvement rather than simply assuming that corrected citations translate to better rankings.
Knowledge Panel accuracy and completeness. A clean, detailed Knowledge Panel that appears consistently for the business name search indicates strong entity confidence. Compare the Knowledge Panel against its state before cleanup. If previously missing attributes now appear, or if previously incorrect information now shows correctly, the entity confidence has improved.
Local pack ranking trajectory. Track local pack rankings across a geogrid for primary target queries. Entity confidence improvements typically produce ranking gains within four to eight weeks of the final cleanup wave. If rankings have not improved within 12 weeks of completing cleanup, the ranking suppression likely stems from factors other than citation inconsistency (proximity, review signals, category alignment).
GBP suggested edit frequency. Before cleanup, Google may have been suggesting frequent changes to the listing based on conflicting external data. After cleanup, the frequency of suggested edits should decrease as the reconciliation system encounters fewer conflicts. A complete cessation of suggested edits (aside from Google’s general feature prompts) indicates that external data sources now align with the listing data.
Duplicate listing recurrence. Monitor Google Maps for the reappearance of legacy listings that were previously removed or marked as closed. If an aggregator correction did not fully propagate, or if a downstream directory re-created a listing from cached data, the legacy entity may resurface. Catching and addressing these recurrences quickly prevents the entity confidence gains from eroding.
The measurement timeline should span at least 90 days after the final cleanup wave, as some propagation cycles and reconciliation refreshes operate on monthly or quarterly intervals. Declaring cleanup complete after two weeks risks missing delayed propagation effects that could reintroduce the original problems.
Should a business that changed its name also update its domain name, or can citations point to the original domain?
Keeping the original domain is usually the better choice if it has accumulated authority and backlinks. Update the on-site NAP data, structured markup, and title tags to reflect the new business name while maintaining the existing domain. Citations should point to the current domain with the updated business name. Changing the domain simultaneously with a name change doubles the reconciliation disruption and risks losing accumulated link equity that supports prominence.
How should a business handle reviews left under its previous name after a rebrand?
Reviews attached to the current GBP listing persist regardless of name changes and should not be removed. They contribute to review count, velocity history, and prominence signals. If a legacy listing under the old name has accumulated reviews separately, request a Google merge to consolidate those reviews into the current listing. Respond to old reviews referencing the previous name naturally, without attempting to hide the business history, as this demonstrates continuity to both customers and Google’s entity model.
Is it worth correcting citations on directories that Google does not appear to index?
Correcting non-indexed directories has limited direct ranking value, but it serves a defensive purpose. Data aggregators and scraping services pull from these directories and can reintroduce incorrect data into indexed platforms during future propagation cycles. If the directory feeds into a data aggregator network, correcting it prevents re-contamination. If the directory is completely isolated with no downstream connections, the correction produces negligible SEO benefit and resources are better spent elsewhere.
Sources
- Whitespark: The U.S. Local Search Ecosystem – https://whitespark.ca/local-search-ecosystem/
- BrightLocal: Citation Builder – Local Data Aggregators – https://www.brightlocal.com/citation-builder/local-data-aggregators/
- Pigzilla: Major Data Aggregators (Localeze, Acxiom, Infogroup, Factual) – https://www.pigzilla.co/seo/data-aggregators/
- Locl: How Agencies Scale With Citation Cleanup – https://locl.ai/blog/citation-audit-cleanup-agencies/
- Citation Builder Pro: Local Citation Audit – https://citationbuilderpro.com/local-citation-audit/
- Local Search Forum: Data Aggregators Still Worth It? – https://localsearchforum.com/threads/data-aggregators-foursquare-data-axle-etc-still-worth-it.61145/
