DGTG logo DGTG

Google Algorithm Updates 2026: SEO Audit Guide for Agencies

A seven-step SEO audit to check if your agency adapted to Google's 2026 core updates. Spot intent drift, AI Overview gaps, and wasted spend before Q3.

Google Algorithm Updates 2026: How to Audit Your Agency's SEO Strategy — illustrative featured image
## The 40 Percent Traffic Drop That Wasn't a Penalty A B2B SaaS client came to us in January with a familiar panic: organic sessions down 41 percent since November, demo requests down 28 percent. Their previous agency had already sent the "Google is being weird" email. Three weeks and one proper audit later, we found the real story. Nothing was penalized. Two of their five highest-intent landing pages had been rewritten in September to chase a keyword that Google's core update had quietly reclassified as informational rather than commercial. The pages still ranked. They just ranked for people who would never buy. That is the shape of most 2026 algorithm damage. Not a hammer. A slow drift in how Google interprets intent, and an agency that never noticed because nobody built the audit into the retainer. If you manage an agency or run growth in-house, here is how to find out whether your SEO strategy actually adapted to this year's core updates, or just survived them by luck. ## What Actually Changed in 2026 Three shifts matter for commercial sites in the US, UK, and EU. **Intent classification got more aggressive.** Google now separates "research" queries from "buy" queries with far more precision. Pages that try to serve both get served to neither. If your agency is still building pages around head terms with mixed intent, you are paying for traffic that will not convert. **Helpful content signals merged into the core.** There is no separate helpful content update anymore. Quality assessment runs continuously. This means recovery is slower and prevention matters more. Sites that got hit in 2024 and "recovered" in 2025 are getting re-hit when they revert to old habits. **AI-generated content is not automatically penalized, but low-effort content is.** Google's position has been consistent: it rewards content demonstrating experience, expertise, authority, and trust. Thin AI content fails on those grounds, not because it is AI. Agencies that banned AI outright and agencies that used it as a volume play are both losing. ## The Audit: Seven Checks in Ninety Minutes You do not need a six-week engagement to assess whether your SEO partner is keeping up. You need access to Search Console, a spreadsheet, and the willingness to ask uncomfortable questions. ### 1. Segment your traffic by intent, not by page Pull twelve months of organic data. Split landing pages into three buckets: commercial (pricing, demo, comparison, "best X for Y"), informational (guides, definitions), and navigational. Now look at the trend line for each bucket separately. | Bucket | Healthy 12-month trend | Warning sign | |---|---|---| | Commercial | Flat to +15% | Declining while informational grows | | Informational | Growing or flat | Growing but zero assisted conversions | | Navigational | Flat | Volatile (usually a brand problem, not SEO) | If commercial pages are shrinking while total organic holds steady, your agency is optimizing for the wrong metric. Total sessions is a vanity number. Pipeline is not. ### 2. Check query-level intent drift in Search Console Open the Performance report. Filter to your top twenty commercial pages. Compare queries from twelve months ago to now. You are looking for queries that used to include commercial modifiers ("pricing," "software," "agency," "for enterprise") and now do not. When that happens, the page has been reclassified. The fix is rarely a rewrite of the whole page. It is usually adding comparison tables, pricing transparency, or proof elements that re-establish commercial intent. ### 3. Audit your internal linking for commercial paths Ask your agency for a crawl of internal links pointing to your money pages. In our experience, most sites have 60 to 80 percent of internal links pointing at blog content, because blog content is easier to produce and easier to link to. Your pricing page should be one of the three most internally linked pages on the site. If it is not, that is a structural problem no amount of content will fix. ### 4. Test whether your content demonstrates first-hand experience Pick five pages your agency produced in the last six months. Read them as a skeptical buyer. Do they contain original data, screenshots, named clients, specific numbers, or process detail that could only come from someone who did the work? If the answer is no, you have a helpful content exposure. Google's quality raters are not grading your prose. They are asking whether this page could have been written by someone with no access to the subject. ### 5. Review technical health against Core Web Vitals thresholds Run your top twenty landing pages through PageSpeed Insights. You want Largest Contentful Paint under 2.5 seconds, Interaction to Next Paint under 200 milliseconds, and Cumulative Layout Shift under 0.1. These are not tiebreakers anymore. They are table stakes, and they matter more on mobile, where 55 to 65 percent of your B2B traffic in Europe still lands. Ask your agency when they last ran this test and what changed as a result. If the answer is "we monitor it," ask to see the monitoring. ### 6. Check AI Overview and AI Mode visibility This is the newest gap. Query your ten highest-value commercial keywords in Google. Count how many now surface an AI Overview above the classic results. Then check whether your brand is cited inside it. Most agencies are not tracking this yet. The ones worth keeping are. Tools like [Semrush, Ahrefs, and Profound](/dgtg/blog/ai-marketing-tools-2026-buyer-s-guide-for-agencies-and-brands) now offer AI visibility tracking, and it should be part of any 2026 SEO retainer over roughly $4,000 per month. ### 7. Ask for the update log This is the simplest test of all. Ask your agency to send you a one-page log of every Google core update since January 2025, what they changed in response, and what happened to your traffic afterward. A good agency has this document already. A mediocre one will produce it in three days and it will be thin. A bad one will tell you that "these updates are unpredictable" and change the subject. ## What We Recommend If you are evaluating whether to keep, replace, or bring SEO in-house, here is our honest split. **Keep in-house:** technical SEO monitoring, Search Console hygiene, internal linking discipline, and content briefs. These require context your team already has. **Hire out:** competitive gap analysis, AI visibility tracking, and large-scale content production. Tools like Ahrefs and Semrush earn their keep at the agency level, not usually at the single-brand level. If you are spending under $3,000 per month on SEO, you are almost certainly better off doing it in-house with one good contractor. **Red flag agencies in 2026:** anyone who leads with "we will get you to page one," anyone who reports on rankings without reporting on assisted conversions, and anyone who has not mentioned AI Overviews in a client call since last summer. For agencies we would actually recommend evaluating: Siege Media and Animalz for content-led SEO, Foundation Inc for B2B technical work, and for European mid-market companies, Rise at Seven and Builtvisible have both adapted their reporting to intent-segmented data faster than most. ## FAQ ### How long should I wait after a core update before judging my agency's response? Four to six weeks minimum for ranking shifts, but you should see a documented response plan within ten days of the update rolling out. If your agency has no written plan by week two, that is the signal, not the traffic numbers. ### Is a traffic drop always a core update problem? No. Roughly a third of the drops we audit turn out to be tracking issues, site migrations, or seasonal patterns. Always rule out analytics breakage before you blame Google. ### Should I fire my agency if commercial pages declined but total traffic grew? Not automatically, but you should demand a revised reporting framework within one billing cycle. Growing traffic with shrinking pipeline means the strategy is misaligned, and that is fixable. If they cannot or will not reframe around commercial intent, replace them.

Frequently asked questions

1. Segment your traffic by intent, not by page Pull twelve months of organic data. Split landing pages into three buckets: commercial (pricing, demo, comparison, "best X for Y"), informational (guide

If the answer is no, you have a helpful content exposure. Google's quality raters are not grading your prose. They are asking whether this page could have been written by someone with no access to the subject.

5. Review technical health against Core Web Vitals thresholds Run your top twenty landing pages through PageSpeed Insights. You want Largest Contentful Paint under 2.5 seconds, Interaction to Next Pa

Four to six weeks minimum for ranking shifts, but you should see a documented response plan within ten days of the update rolling out. If your agency has no written plan by week two, that is the signal, not the traffic numbers.

Is a traffic drop always a core update problem?

No. Roughly a third of the drops we audit turn out to be tracking issues, site migrations, or seasonal patterns. Always rule out analytics breakage before you blame Google.

Should I fire my agency if commercial pages declined but total traffic grew?

Not automatically, but you should demand a revised reporting framework within one billing cycle. Growing traffic with shrinking pipeline means the strategy is misaligned, and that is fixable. If they cannot or will not reframe around commercial intent, replace them.

Related articles