Technical SEO is so back! In the early days of Google, a lot of SEO was about making sure search engines could actually crawl, understand and index your website and get links, get so many links! Then, through the 2010s and into the early 2020s, the focus shifted much more towards content, digital PR (links), UX, brand, entities and trust. Technical SEO never went away, obviously, but on a lot of established websites it became less of the problem. That seems to have changed again. I’m seeing more technical SEO issues this year than I have in the last few years combined. A lot of that is down to how websites, apps, and web assets are being built and teams operate now. We’ve moved from fairly rigid CMS platforms to headless builds (often not required), custom frameworks, internal tools, dynamically generated pages and now a growing amount of vibe-coded websites, tools and content assets. That gives teams the autonomy they need and removes developer bottlenecks.  It also gives them far more ways to break things. - JavaScript rendering issues. - multiple frameworks. - subdomain party central. - Content that exists for users but not for crawlers. - Infinite URLs. - Cloaking. - Wrong status codes. - Internal links that aren’t really links. - Content hidden behind interactions. - Caching problems. - Server header responses. - Resources accidentally blocked. - Staging sites indexed. - Production sites accidentally noindexed. You’ve got Googlebot, Bingbot, AI search crawlers, training crawlers and retrieval bots, all behaving slightly differently. Some clients want certain AI bots to have access and others blocked. Some bots handle JavaScript better than others. Some follow the same controls. Others have their own. So after 20+ years of SEO, we’ve somehow ended up asking many of the same questions again. - Can the bot (all of them) access it? - Can it render it? - Can it understand what it’s looking at? - Is it seeing the version we actually want it to see? - Have we accidentally blocked it? The tech and internet has changed. A lot of the technical #SEO problems really haven’t.

There’s been a lot of recent reporting and discussion around Steven Bartlett and The Diary of a CEO. Rather than add another opinion, I wanted to see what the channel’s own public data could tell us. So I pulled together the history of the YouTube channel and analysed the videos, titles, guests, topics, colours, expressions, views and thumbnails over time. Some of the questions I wanted to explore: • When did business stop being the focus? • When did the topics broaden? • Why is Steven in every thumbnail? • How have thumbnails evolved? • Which topics perform best? • Do celebrity guests perform? • How much is A/B tested? The research maps those changes across the history of the channel, including the evolution of its content mix, titles and visual style. I’ve also published the underlying data and methodology openly, so journalists, researchers or anyone else interested can explore it themselves rather than relying on my findings. A lot more data and findings at doac.audits.com

Felt cute, might delete later. A bit of fun at carlhendy.com.

Early days, but the recovery from this failed website migration is finally starting to get going. Hopefully, I’ll be able to talk about this one publicly one day. It’s a banger.

Has anyone actually bought something or booked travel using 100% agentic search yet? No Google. No browsing sites. No comparing options yourself. Just told an AI agent what you wanted and let it research, choose and complete the purchase or booking?

Updated my free StoreRatings.co tool. Compare 5 ecommerce competitors across 30+ signals, including Google Shopping, Core Web Vitals, Lighthouse, Knowledge Graph, trust checks, review momentum and more. Still 100% free. No signup. All public Google data.

I've rebuilt my free Knowledge Graph Search tool. Search any brand or person to see what Google knows about them, whether they have a verified Google profile, and how Wikipedia, Wikidata and Common Crawl see them. You can even compare two entities side by side.

@Cloudflare's new AEO Visibility product feels like Similarweb for AI discovery: tracking brand mentions, citations, prominence and share of voice across models, then adding crawler and referral data from the site itself.

Google has launched Search profiles for publishers and creators. So I built a tool to help you find Search Profiles from anywhere. search.audits.com/