Core Update Wiped Your Website? Check Your Map Pack First
Google's December 2025 core update knocked out your website traffic, and every fix on offer is expensive. Google's own documentation points you to a content self-assessment. The recovery guide ranking beside it prescribes keyword fixes and a 3 to 9 month wait. Joy Hawkins at Sterling Sky gives you a harder choice: wait for the damage to lift, or start over on a new domain.
Start with a cheaper question: where do your calls actually come from? If the map pack still makes your phone ring, your job for the next few months is narrow. Protect your Google Business Profile, stop buying links, and use paid search to win back the question-stage searches you lost while your site is held down. The rewrite and the new domain can wait until your numbers say they are worth it.
The update flattened your website and left your map listing standing
Joy Hawkins published the clearest account of the damage on September 30, 2026, in a Sterling Sky post titled "Google's 2025-2026 Core Updates Are Destroying Local Businesses." She describes local business sites losing 90% or more of their organic traffic. Hawkins says the warning shots came in the June 2025 core update or the August 2025 spam update. The December 2025 core update finished them off. Her account is qualitative. She does not say how many sites her observations cover.
"One of the most dramatic algorithm updates I've ever seen impact local businesses was the December 2025 core update," she writes.
The mechanism is called a classifier. A classifier is an automated label Google attaches to your whole domain, and once it is attached, every page on the site ranks lower. Hawkins says these labels apply site-wide and come with no notice in Search Console. Without a notice, the first sign is usually a traffic chart.
Hawkins also found where the damage stops. In her words: "What's interesting is that algorithmic classifiers impacting standard organic rankings do not suppress Local Pack rankings." The local pack, or map pack, is the block of three businesses with a map that appears for searches like "plumber near me." Your website can carry the label while your Business Profile keeps ranking in that block.
None of the top-ranking recovery articles asks what the crash actually cost you in leads. Whitespark's AI Overviews study gives you a way to estimate it. Miriam Ellis ran it in May 2025 across 540 queries in Houston, Phoenix and Denver, covering six industries. The map pack appeared on 93% of local-intent queries. It appeared on only 6% of informational queries and 17% of hybrid ones.
| Search type | Example search (illustrative) | Map pack appears | AI Overview appears |
|---|---|---|---|
| Local intent | "emergency plumber Houston" | 93% | 15% |
| Informational | "why is my drain gurgling" | 6% | 92% |
| Hybrid | "how much does drain cleaning cost in Houston" | 17% | 97% |
Informational and hybrid searches are the question-stage searches, typed before anyone hires a plumber or books a dentist. In that same study, AI Overviews already appeared on 92% of informational results and 97% of hybrid ones. My reading of that appearance data: for many service businesses, much of what a classifier took was traffic Google's AI summaries were already pulling away before the update.
The arithmetic goes like this: list the lost queries whose pages produced leads, price them per click in Google Ads, and compare that monthly bill with the invoice for a rebuild. Local-intent searches are the ones closest to a booked job, and the pack appears on 93% of them. If your pack listing still ranks, my read is that the bulk of your loss was researchers, plus whatever local clicks the pack does not catch. That loss hurts. It does not justify burning a domain that still feeds your best lead source.
I have been in search since 2008. My 17 years cover the whole run of exact-match anchor link building, from the years it moved rankings to the years Google started punishing it. That history is now catching up with these sites. I have also run Google Ads for 10 years. That experience lets me price whether paid search can win back the lost queries for less than a domain restart costs.
Google's recovery advice never looks at links, and local owners pay for that
Google's own page on the topic, "Google Search's core updates and your website," was last updated on December 10, 2025. It states: "These changes are broad in nature, and don't target specific sites or individual web pages." After a large drop, it tells you to assess your own content. It never mentions links.
For a recipe blog or a news site, that advice can fit. Now take a roofer whose site still carries anchor links an old agency bought in 2020. A content rewrite treats the wrong problem. He can rewrite every service page and still carry the label that holds them down.
Google's page also sets your expectations on timing. A rewrite that misses the cause costs you the agency invoice first and the wait for another update second.
The recovery guide I read follows the same script. The May 24, 2026 guide by Chris Yadv at Branding Marketing Agency walks through finding the affected pages, analyzing keywords, fixing intent mismatch and adding E-E-A-T. E-E-A-T is Google's shorthand for experience, expertise, authority and trust. The guide puts recovery at 3 to 9 months and recommends earning editorial backlinks.
| Who is advising | What they tell you to do first | What it leaves out |
|---|---|---|
| Google Search Central (updated December 10, 2025) | Assess your content after a large drop | Links |
| A recovery guide (Branding Marketing Agency, May 24, 2026) | Fix pages, keywords and intent, add E-E-A-T, earn editorial backlinks, plan on 3 to 9 months | Links appear only as a growth tactic, and the map pack goes unmentioned |
| Joy Hawkins, Sterling Sky (September 30, 2026) | Diagnose in Search Console, then wait for the classifier to lift or move to a new domain | What the lost traffic was worth in leads |
| The lead-source split | Find which source produces your calls, protect the profile, stop buying links, price paid cover for the lost queries | A permanent fix. It buys time and tells you whether a rewrite or a new domain is worth paying for |
Hawkins' diagnosis is the sharpest on offer. Her recovery advice is a fork: wait for the classifier to lift, or start over on a new domain. She also reports, "So far, none of the sites we are tracking that got hit in 2025 have recovered." Waiting costs real money when your sales depend on the site.
A new domain sounds clean. It also starts without the links and history the old one earned. Google, the guide and Hawkins all skip the same question: how many leads did the website produce before the crash, and how many does it produce now?
Old anchor links sank the website, and the map pack barely weighs links
"We are seeing businesses get nailed for links they paid an SEO company to build five years ago," Hawkins writes. Those owners paid an SEO company for that work. The bill came years later.
An anchored link is a link whose clickable text is a keyword, such as "best roofer in Tampa," instead of your business name. In my early years in search, exact-match anchors moved rankings and were sold openly as a service. In Hawkins' cases, links built years earlier were still enough to trigger a flag.
Press releases get their own warning. Hawkins writes, "I strongly recommend you avoid using anchored links in press releases entirely." If a vendor still sells you distribution with keyword links baked in, cancel it this week.
The roofer Hawkins describes should worry every owner who has never hired an SEO. Competitors built links pointing at his site, and he got hit anyway. You cannot fully control who links to you. Knowing your own link profile is now part of running the business.
Where links carry weight depends on which channel you ask about. Whitespark's 2026 Local Search Ranking Factors survey, led by Darren Shaw and published November 6, 2025, asked 47 local search experts to estimate each factor's share. Treat the figures as informed expert opinion. Advice Local transcribed them on December 2, 2025.
Experts put links at 24% of local organic rankings and only 8% of the local pack. Your Business Profile carries 32% of the pack and just 7% of local organic. My reading: Hawkins' finding fits these estimates, because links count for little in the pack and your profile carries it instead.
| Ranking factor | Share of map pack rankings | Share of local organic rankings |
|---|---|---|
| Google Business Profile | 32% | 7% |
| Links | 8% | 24% |
| On-page | 15% | 33% |
| Behavioral | 9% | 10% |
| Reviews | 20% | Not in the cited figures |
In the experts' estimates, your Business Profile and reviews together account for 52% of the pack. I covered why reviews beat proximity in the map pack in a separate piece. If a classifier hit your site, those factors are where your remaining leads come from, and they deserve your attention first.
Split leads by source before you pay for any recovery
Run these in sequence. Steps 1 to 4 need Search Console and your Business Profile. Step 5 needs someone who can read a Google Ads account.
Branded searches are the ones that include your business name, and people who type them would find you anyway. Removing the profile's page strips out most of the clicks that come from the map pack. What remains is a cleaner read of your organic website traffic. A sudden cliff on one of those dates is the signature Hawkins' diagnostic looks for.
Picture a physio clinic that ranked for "why does my knee click when I squat" and "physio near me." Say the knee question fell off its Search Console chart after December while the "physio near me" calls kept coming through the map pack. That clinic should protect its profile, fire its link vendor, and put a small ad budget on the question searches that used to bring in new patients. A new domain would throw away the listing that still rings.
The fork after step 5 is plain arithmetic. Suppose the pack still produces most of your calls, and covering the lost queries costs less per month than a rebuild. Then protect the profile and buy the queries. If the website was your main lead source and the lost queries were local-intent searches, the domain question becomes real. At that point Hawkins' new-domain option deserves a serious costing. The split does not replace her fork. It tells you whether you ever need to take it, and what the wait costs you in the meantime.
While the classifier holds, keep the profile accurate: correct hours, steady review requests, and photos of real jobs. Cancel any retainer that still builds links, and get the full list of what was built under your name. Start the paid cover small, and only on question searches that produced jobs before December. A big opening budget burns cash before the account learns anything, which I unpacked in why starting big on Google Ads backfires.
If you have an agency, ask them this: "Which of our lost queries were local-intent, and what would it cost per month to cover them with Google Ads?" An agency that answers with a rewrite proposal and no lead-source numbers has skipped the step that tells you whether the rewrite is worth paying for.
Judge recovery by calls from the pack, and ignore total traffic
Total organic sessions will look grim for months. That chart will tempt you into panic spending. Track the numbers tied to revenue: Business Profile calls, direction requests, website form leads, and paid-search leads with their cost per lead.
Search Console and your Business Profile dashboard will disagree with each other, and the disagreement confuses owners. The reasons Search Console and Business Profile numbers never match have their own post. Read each number from its home surface:
Watch for vanity metrics. Impressions on informational queries can creep back while clicks stay flat, a likely effect of AI Overviews answering the question on the results page. An average-position line from a rank tracker can improve on keywords that never produced a call. A rising count of linking sites is a warning sign: Hawkins lists referring domains climbing while traffic falls among her flag patterns.
The same BrightLocal survey found the share of consumers who read reviews on Google fell from 83% to 71% between the two surveys. Your customers are spreading their research across more places, and my read is that part of the question-stage search you lost was heading there anyway. Your Business Profile is where the ready-to-buy calls get counted, and that is the number to protect.
Put the checks on a calendar. Check paid cover weekly for the first month, judging it by cost per lead. Review pack calls and website leads monthly. Re-run the Search Console overlay after each confirmed core update, since Google's own page says a stuck site may have to wait for the next one. Hawkins' tracked sites had not recovered when she wrote, and yours may take as long.
Frequently Asked Questions
Can my website recover from a Google core update?
Google's guidance says a site that sees no change after a few months may need to wait for the next core update. Joy Hawkins of Sterling Sky reported in September 2026 that none of the sites her team tracked after 2025 hits had recovered yet. Google points you to a content self-assessment, while Hawkins points to link history. Overlay the June, August and December 2025 update dates on 16 months of Search Console data to see which story fits your site.
Does a core update affect my Google Maps ranking?
The map results come from the Local Pack, and Hawkins found that the classifiers hitting local websites in 2025 and 2026 do not suppress Local Pack rankings. Expert estimates in Whitespark's 2026 Local Search Ranking Factors survey put links at only 8% of what drives the pack, against 32% for your Business Profile. Your map listing can keep producing calls while your website struggles. Check your Business Profile calls and direction requests month over month to confirm it for your own business.
Should I move my business website to a new domain after a core update?
Not as a first step. A new domain starts without the links and history the old one earned. Split your leads by source first and price the lost queries in Google Ads. A new domain starts to make sense when the website was your main lead source and the queries you lost were local-intent searches that paid cover cannot reach at a sensible cost.
A classifier on your domain is a real problem, and Hawkins reports it can last a long time. It is also narrower than the traffic chart makes it look. My read is that much of what it took came from people still researching, in results AI summaries already crowd. If your map pack never stopped ringing, ask how much of the website traffic you are mourning ever paid your bills.
If you want my eyes on your lead-source split before you commit to a rewrite or a new domain, book a call with me and we will price the gap together.
