The Visibility Audit.

Four years ago I wrote a short article about web search engine marketing mistakes. The central argument was simple: orthodontic practices spend a healthy monthly budget on PPC in an attempt to bring prospective patients to a chaotic and confusing website, and the only thing they achieve is a high bounce rate. The data supported it then and it supports it now: 94 percent of all clicks on Google go to organic results, and the top organic position captures 39.8 percent of all clicks on the page, paid results included. I compared the whole exercise to throwing a plate of spaghetti on the wall to see how much would stick. In 2026, the spaghetti is still on the wall. It just has more platforms to stick to.

The same mistake wearing new clothes

In 2022, the mistake was straightforward: practices poured money into Google Ads without understanding their patient demographic or geographic catchment, exhausted their daily budget chasing unqualified clicks, and hoped for results. The data has only gotten louder. First Page Sage, WordStream, and Seer Interactive have all documented the same pattern with current numbers. The average cost per click in paid search has climbed to $5.42 across more than 13,000 US campaigns tracked through March 2026. The top paid search result captures a 2.2 percent click-through rate. The top organic result captures 39.8 percent. That is not a rounding error. That is a structural fact about where the clicks go.

In 2026, the same mistake has metastasized. OpenAI now runs ads inside ChatGPT. Their latest product update introduces “conversion optimization with impression billing,” campaigns that bill on impressions rather than clicks while optimizing toward a conversion goal. They describe it as their recommended model for maximizing performance. Of course they do. The advertiser pays for delivery. The platform carries none of the click risk. And in a chat interface, the answer is already sitting above the ad.

Meanwhile, on Google itself, AI Overviews now appear on roughly 31 percent of search result pages, and they are eating organic clicks alive. Seer Interactive measured organic CTR falling 61 percent on queries where an AI Overview appears. Zero-click searches hit 45.6 percent in December 2025, the highest rate ever recorded, and the rising zero-click share is coming almost entirely out of organic clicks, not ads. The one defense? Being cited as a source inside the AI Overview. And 91.3 percent of those citations come from the organic top 20. Paid placements are never cited. If you are not ranking organically, you are not in the conversation the machines are having about your industry.

The technology changed. The mistake did not. Broad-brush, non-targeted waste is still broad-brush, non-targeted waste whether it runs on a search engine, a social feed, or an AI chat window. It is still no different than billboard advertising or coupon mailers everyone throws away. Stamps on trash.

The spaghetti is still on the wall. It just has more platforms to stick to.

The visibility radius you don’t control

Here is something most orthodontic practice owners have never been told. The geographic radius in which your practice appears in local search results is not a fixed fact about your practice. It is a Google output. Google’s algorithm considers how saturated the competition is in any given radius and adjusts the visibility boundary accordingly. A practice in a dense metro gets a tighter cone. A practice in a rural market gets a wider one. Same practice, same budget, same content, different reach. An orthodontist in suburban Atlanta competes against every other orthodontist, every DSO, every corporate chain in the radius, and Google decides where the fence falls.

That algorithm is unpublished, undocumented, and changes constantly. The only way to observe the boundary is to query Google from a grid of coordinate points and map where your practice shows up and where it drops off. You are inferring the fence, not reading it.

This is the first problem a visibility audit solves. You cannot make decisions about channels, content, or spend until you know where Google is allowing your practice to be seen. And you cannot know that without deliberately mapping it from the outside, because Google will never show you the map from the inside.

The page the map can’t see

But here is where most audits stop, and where the real problem begins. You can map your visibility radius down to the block. You can identify every ZIP code where your practice shows up and every one where it drops off. And none of that matters if your practice’s pages are not indexable.

A page that exists on your server but is not indexed by Google is invisible. It does not participate in the radius. It does not rank. It does not appear. It is publishing into a void. Approximately 94 percent of all website pages get no traffic from Google searches. Most of those pages have an indexability problem nobody diagnosed. And in 2026, the stakes are higher: if your page is not indexed, it cannot be cited in an AI Overview, it cannot appear in a ChatGPT answer, and it cannot surface in any of the AI systems that are increasingly deciding what gets found and what does not.

I wrote in 2022 that assumption is the mother of all web marketing mistakes. That statement has not softened with age. Why would any orthodontic practice spend a monthly budget on PPC, on social media, on AI-driven campaigns, in an attempt to drive patients to pages the crawlers cannot even reach? The assumption is that the practice website works. The assumption is that the treatment pages are indexed. The assumption is that the developer or the template provider did the job. And the assumption is wrong more often than anyone wants to admit.

A page that exists on your server but is not indexed by Google is publishing into a void.

What indexability scoring actually measures

This is why we built Indexability Scoring. Not as a one-time audit you file and forget, but as an ongoing scorecard that keeps the developer honest, keeps WEBPRO engaged, and keeps you informed the whole way. Every page gets a score from 0 to 100, and every score is tied to six specific, checkable conditions that decide whether a page gets read by the machines.

Think about the last time a doctor handed you a blood panel. You see triglycerides, calcium, a dozen numbers with ranges next to them, and you have no real idea what you are looking at. The report is accurate and it is useless to you on its own. A good doctor does not just hand you the page. They read it back to you, tell you what is high and why it matters, and prescribe the changes that actually move it.

We are not Semrush selling you an expensive subscription and a dashboard no front desk manager can read. We do the mechanical work, all of it, and we read the results back to you in plain language with a plan you can act on. You run your practice. We keep the machines able to read it.

The audit framework

A visibility audit is not a single event. It is a sequence. And it has to happen in this order, because each layer depends on the one below it.

Layer one: indexability. Can the machines reach, read, and trust the page? Every other number you chase, content quality, page speed, backlinks, rankings, only pays off after this is right. If the foundation is broken, the content is publishing into a void. This is why indexability scoring comes first, always.

Layer two: the visibility map. Once the pages are indexable, map the radius. Query from a grid of coordinate points and see where Google allows your practice to appear. Identify the boundary. Identify where competing orthodontists and DSOs are compressing your cone and where open geography gives you room to grow your patient base.

Layer three: content effort. Google has a number for how lazy your content is. The leaked Content Warehouse API confirmed an attribute called contentEffort, and it functions as the algorithmic proxy for the Experience and Expertise pillars of E-E-A-T. Featured snippets now achieve a 42.9 percent CTR, higher than traditional position one, and they reward pages that deliver direct answers, structured data, and clear specificity. A page built on stock smile photos, template “what are braces” copy, and AI-generated FAQ content does not earn a snippet. A page built on consent-cleared before-and-after cases, the doctor’s clinical voice, and local data like insurance accepted, financing options, and school-schedule-friendly appointment hours does. There is no shortcut, and AI-generated content is accelerating the problem: 58 percent of SEO marketers report a significant increase in competition due to the rise of AI. The floor is rising. Low-effort pages that coasted in 2022 are now competing against machine-generated content that is itself low-effort but prolific. The only defense is genuine effort, and Google can measure the difference.

Layer four: channel selection. Only after the first three layers are sound does the channel question matter. Google captures intent. Social media manufactures demand. AI Overviews cite the organic top 20, never paid results. But if the practice’s pages are not indexable, the map is invisible, and the content registers as low effort, no channel in the world will drive patients through the door. You are not marketing your practice. You are throwing spaghetti.

Final thoughts, updated

I closed the 2022 version of this article with a simple line: you would be wise to think about building a long-term organic foundation with a reputable SEO partner rather than a wasteful PPC deck of cards platform for your practice. Four years later, the average cost per click has climbed to $5.42, organic click share has compressed by double digits across major verticals, AI Overviews have rewritten the results page, zero-click searches have hit their all-time high, and ChatGPT has entered the advertising layer billing on impressions. The advice has not changed. The urgency is no longer optional.

Templates are still sloppy shortcuts, and orthodontic practice websites are lousy with them. Unseasoned designers, template providers marketed specifically at dental and orthodontic offices, cheap hosting companies, slow DNS, and old servers will still cripple your efforts in the end. And now, on top of all of that, the machines doing the reading have multiplied. Googlebot, GPTBot, ClaudeBot, Applebot, PerplexityBot, and whatever Amazon and Apple ship next. Each one with its own crawl behavior, its own rendering requirements, its own standards for what counts as a page worth keeping. Each one deciding independently whether your content is worth citing, worth surfacing, worth remembering. If your practice’s foundation cannot survive the scrutiny of all of them, you are not building on rock. You are building on spaghetti.