SEO Audit

Thank you for trusting us with your SEO audit. Somebody built this business and this website, and having a stranger measure it is not a small thing to invite. Everything in this report is measured, never guessed, and where we could not measure something we say so plainly rather than filling the gap.

What you are already doing right

Every line here is something we measured, not encouragement. These are also the things not to break while fixing everything else.

• Your site is genuinely fast on a phone, scoring 86 out of 100. Most small business sites do not, and it is the single thing visitors feel first.
• 33 different websites link to yours. Links are the slowest thing to build and you have a real head start.
• All 20 of your pages have a title and a description written for them. That is unusual and it is real work somebody did.
• You have structured data on the site (FAQPage, LocalBusiness, Organization), which is how a search engine knows what kind of business this is rather than inferring it from your words.
• Your site is served securely over HTTPS with a valid certificate, so no browser will warn a visitor away.
• You publish a sitemap, so search engines are told where your pages are instead of having to guess.

404 handling: Correct Missing pages return HTTP 404 and the page helps the visitor get back on track.

• The 404 page has no search box. Optional, but it is the most direct way to rescue someone who followed a broken link.

Measured by Lighthouse via DataForSEO on a simulated mobile device.

What each of these means, and why the server response time above looks so different

The two numbers are measuring different things, and both are correct. The server response time in the section above is how long your server took to hand over the HTML file, measured once from a fast connection. It is a plumbing number. Everything below is a real browser on a throttled mobile connection actually building and drawing the page: downloading images, running scripts, loading fonts. A site can hand over its HTML in a fraction of a second and still take many seconds to become usable, and that is exactly what a heavy page looks like.

Put plainly: the fast number is your host. The slow numbers are your page.

The right-hand column is Google's own published target for a phone, checked against their documentation in August 2026. Hitting it is what makes Lighthouse grade that measurement green.

Lighthouse is Google's own page-quality tool, built and maintained by the Chrome team. It is the same engine behind PageSpeed Insights and the Lighthouse panel inside Chrome's developer tools, so anyone you hand this report to can re-run it themselves and get comparable numbers.

It matters here for one specific reason: the Largest Contentful Paint, Cumulative Layout Shift and Interaction to Next Paint figures Lighthouse reports are the same Core Web Vitals Google has publicly confirmed it uses as ranking signals. This is not a third-party opinion about your speed, it is a measurement from the company doing the ranking.

Two honest limits. First, Lighthouse runs a simulated mid-range phone on a throttled connection, which is deliberately harsher than a modern phone on good wifi, so treat the scores as a stress test rather than as what every visitor experiences. Second, a single run varies: the timing figures move a little between runs on an unchanged site, which is why this report never treats a small change in them as an improvement. The score and the page weight are far more stable than the individual timings.

What these mean

The meta description is the two lines of grey text under your name in a Google result, and it is the only sales pitch you fully control at the moment somebody is choosing between you and four competitors. It is not a ranking factor and never has been, so writing one will not move your position. What it moves is whether the person clicks. Leave it empty and Google grabs whatever sentence it likes off the page, which is often a navigation menu or a cookie notice. It is also the text that appears when anyone shares your page on LinkedIn, Facebook or in a message. One or two real sentences under 160 characters, saying who you help and what they get.

20 of 20 pages read. Sampled evenly across the sitemap rather than taking the first few, so this is not just the homepage and the newest pages.

Why these four things matter

Your home page publishes 5 schema types: FAQPage, LocalBusiness, Organization, ProfessionalService, Question

Other pages add: BreadcrumbList, ListItem, Person, Article, Service, Product

Structured data is how search engines and AIs identify what your business is and what each page covers. It is code in the page, not text you type, so on a hosted site builder it usually means a code-injection block or a plugin.

Sitemap freshness: Clean All 20 of the 49 listed URLs we sampled load correctly and allow indexing.

We sampled 20 of the 49 URLs in your sitemap, spread evenly through the file rather than taken from the front, because the front is usually the homepage and the newest pages, which are the least likely to be stale. The sitemap publishes no last-modified dates, so there is no way to tell from the file alone how current it is.

No Google Maps listing matched the name "SEO Believer", which on its own is ambiguous: this search matches on the business name and a punctuated or abbreviated name can miss the index.

The map pack for that search is held by Boulder SEO Marketing (28 reviews), BSM - A Denver SEO Company (63 reviews), Hoopless Marketing SEO & AI Consulting (3 reviews). Those positions are free and they are won with a profile and reviews, not with ad budget.

Before anything else, find out whether a profile exists

This check could not confirm one either way, so the first job is five minutes on Google Maps rather than a list of settings to change. Search for your business name, then your own name, then your name plus what you do. If a listing appears, claim it. If nothing appears anywhere, create one. Creating a second profile for a business that already has one is the mistake that takes weeks to unpick, which is why this comes before every other suggestion in this section.

Confirmed against a live Google search for "seo consultants boulder" from United States. Your site did not appear in the first ten ordinary results for it.

Why this matters more than almost anything else on this list

For a business people look for near them, the Google Business Profile is usually the highest-value free thing available, and it sits above the ordinary blue links. When somebody searches a service plus a place, Google shows a map with three businesses first. Most people choose from those three without scrolling.

Three things follow from that. First, you can rank first in the ordinary results and still be invisible, because the map box is a separate competition with separate rules. Second, the map box is won with a claimed profile, correct categories and reviews, none of which cost money, so a competitor cannot outspend you on it. Third, the profile is where your hours, location and photos come from, so without one Google is answering questions about you from whatever it can find elsewhere.

It is also increasingly what AIs read when someone asks them for a local recommendation, so an absent profile costs you on both surfaces at once.

Whether this business is local at all was judged from its own pages, not from its industry: the site publishes a postal address; the site carries LocalBusiness-family structured data.

What "domain rank" means, and what it does not

Domain rank is DataForSEO's own estimate of how much authority a website has accumulated, on a scale that runs to about 1,000. It is built from how many other sites link to you and how strong those sites are.

It is not a Google score. Google publishes no such number. It is also not Moz's Domain Authority or Ahrefs' Domain Rating, so comparing it to a figure an agency quoted from a different tool is meaningless: the scales are different.

What it is genuinely useful for is comparison against the specific competitors in this report, measured the same way on the same day. Treat it as a relative position, not a grade.

One thing worth saying plainly for a local business: a modest number here is not the problem people assume. Being found near you is won in the map results and on your own pages far more than by out-linking a national organisation. Link building is slow, expensive, and rarely the first thing worth your time.

33 referring domains send 35 links to this site.

1 broken page found. No inbound links point at it, so this is tidying rather than lost authority. Each is named below.

The broken pages, and what each one is

A broken page with links pointing at it is authority somebody already gave you landing on nothing. Redirect it to whatever replaced it and that authority starts counting again.

These are the phrases somebody in Boulder would type to find a business like yours, with monthly search volumes from Google. They are deliberately different from the keyword list further down, which is mostly people searching your name because they already know it.

"Not checked" is not the same as "not ranking". Positions here come from the list of keywords this domain is already known to rank for, and that list only holds phrases with enough search volume to be tracked. Where a phrase shows no volume, its absence from that list tells us nothing about where you actually rank, so we say we did not check rather than reporting a zero we have not earned. Confirming those would take a live search per phrase, which is a real cost and is not part of this tier.

This site does not appear in the top 100 for any of the 10 local phrases below. That is the gap between being findable by people who already know your name and being findable by people who do not.

The strongest of them is "seo consultants near me" at about 1600 searches a month. In a single town that is a meaningful number, not a small one: those are people looking for exactly what you run, in the place you run it.

These are the sites that actually rank when someone searches "seo consultants boulder", which is the question a potential member types. Directories and social platforms are excluded because you cannot compete with them on their own terms.

2 of 2 competitors rank for more keywords than this site. The widest gap is boulderseomarketing.com, with 808 ranked keywords against this site's 2, about 404 times as many.

Nothing currently ranks in positions 11 to 20, so there is no page-two quick-win list to work from. Gains here have to come from new pages or new links rather than from nudging existing ones.

Page-two keywords are filtered to ones related to what you do, measured against "seo consultants boulder", "seo consultants".

builtincolorado.com also rank for these searches, and it is deliberately left out of the comparison below. That domain is more than fifty times the size of this site, so it is not a useful comparison for planning.

98 more of boulderseomarketing.com's keywords were left out because they are not related to what you do. A rival can rank for things that are none of your business, and chasing those is how a free afternoon gets wasted.

2 searches below are ones boulderseomarketing.com ranks for and this site does not. That competitor was chosen because it ranks highest for "seo consultants boulder", the search a customer actually types, so its terms are the closest thing to your own market.

Nothing currently ranks in the top ten, so there is no page-one position to defend yet.

What boulderseomarketing.com ranks for and you do not

These are terms the closest comparable site to yours ranks for and you do not. They are filtered down to things related to what you do, at a size you could plausibly compete for. Treat them as candidates for a page rather than a to-do list.

All major AI crawlers allowed

Your robots.txt does not block ChatGPT, Claude, Perplexity, or Google's AI crawlers. That is the correct setting for being quotable.

llms.txt file: present. llms.txt is an emerging convention for telling AIs what a site is about. It is not yet honored by the major AIs, so it is not currently costing you traffic.

This section measures whether AIs are permitted to read your site. Whether they actually cite you is measured separately in the AI Citation Share Index.

Google finds pages by following links, and the links your own pages send are most of how it decides which of them matter. A page nothing links to is one Google may never find and a visitor certainly will not.

Measured across the 20 pages we could read. That is the honest boundary: "nothing links to this" means nothing among those pages does, not nothing on the internet.

Pages nothing links to

These are reachable only by typing the address or finding them in your sitemap. They receive none of the authority the rest of your site has earned. Linking to each one from a relevant page is usually a few minutes of work.

Pages buried deep

How many clicks from your home page. Three or more reads to a search engine as "not important", and it is crawled less often as a result.

Pages with the fewest links pointing at them

The typical page on your site has 2 internal links pointing at it. These have the least.

Ranking on approximately 2 organic keywords nationally.

0 of 100 share of voice

We asked 8 buyer questions across 4 AIs for Boulder, CO, about "seo consultants". 31 of 32 answers came back usable, and only those were scored. None of the questions mentioned your name, so being named had to be earned.

Every one of these answers was produced with live web search switched on, which is what a customer's AI does when they ask it a question like this. That matters: an AI model answering from memory alone routinely invents local business names, so a memory-only test would tell you nothing about your actual visibility.

How these four differ, and the one we cannot test

Perplexity decides differently from the other three. It is a search engine first: it runs a live web search for essentially every question and then summarises the pages it found, with links. That makes it the most directly connected to ordinary SEO of the four, because where you rank in a normal search largely decides whether Perplexity sees you at all. ChatGPT, Claude and Gemini blend what the model already learned during training with whatever it looks up in the moment, so they can name a business without any page in front of them, and they can miss one that ranks perfectly well.

The biggest AI surface is not on this list. It is the AI answer Google now shows at the top of its own results, which reaches far more people than every chatbot combined simply because it appears in ordinary Google searches. We do not test it here: there is no reliable way to query it as a customer would, and this report does not report numbers it cannot measure.

Worth knowing how that one behaves, though, because it is the opposite of what people expect. Google's AI answers reach the most people and send the FEWEST clicks, since they answer the question on the page and the searcher never leaves. Most measured AI visits to websites still come from ChatGPT. So treat these two as different goals: getting named in Google's AI answer is about being the business it recommends, and getting cited by a chatbot is what actually brings somebody to your door.

You were not named in any answer. The businesses named instead are listed below, along with the pages the AIs read to find them. Some of them are local businesses like yours, which means this is a contest you can actually enter.

Your structured data does not name WHICH KIND of business this is, so these questions used a category worked out from the page copy instead. A LocalBusiness tag says you are a local business; a more specific type says what you actually do, and that is the part missing.

Every question we asked

The exact wording, so you can judge whether the test was fair. None of them mention your name: being named had to be earned. Each was asked with the location set to Boulder, CO, which is the town published on your own site, not your street address. AIs answer at town level, so a precise address would not change the answer.

The businesses the AIs named

These are not scraped out of the wording of the answers. Each one is a business the AI identified as an entity, with its own web address, so every row below is a real company you can go and look at.

Where the AIs went to answer

These are the pages the AIs actually opened before answering, most-used first. This is the practical half of this section: if you are not on the sites at the top of this list, you are invisible at the exact moment somebody asks an AI for a recommendation. Getting listed on them is usually far quicker than trying to make an AI model remember you.

Some individual answers did not come back: ChatGPT lost 1 of 8. Those are excluded from the maths rather than counted against you, and they are named here so a dropped answer cannot be mistaken for nobody mentioning you.

Not found in the Google Maps results

Your business did not appear anywhere in the 20 Maps results for this search. That is a real finding, not a failed measurement: for a local business this is usually the single most expensive gap on the list, because Maps is where people pick who to call.

You can get into this list, but a Google Business Profile is the entry ticket. Google Maps only ranks businesses that have one, so no amount of work on your website will place you here. Create or claim yours first, complete it properly, and then reviews and categories are what move you up. It is free, and it is the highest-value hour in this whole report.

We searched Google Maps for "seo consultants boulder" from Boulder, Colorado, United States, which is the kind of thing a customer types when they do not already know your name. That location is the town your site publishes, not your street address: Maps answers differently for somebody standing outside your door than for somebody across town, and a town-level search is the fair middle. Your structured data does not name WHICH KIND of business this is, so this search used a category worked out from the page copy instead. A more specific business type in your markup would sharpen it.

Who owns this search instead

Every "currently reads" line below is quoted from your page and was checked character by character against it before this report was written. The suggested rewrites are a copywriter's opinion, not a measurement, and they are the only part of this report that is. The copy review currently covers your home page only, which is why every finding below shares one address.

Ranked by how much each one is likely to be costing you, judged from what could be measured on this run. Effort is a rough estimate: S under an hour, M a few hours, L half a day or more.

The last column matters as much as the order. Some fixes are not available on every platform, and a task you cannot perform is a wall rather than a job, so each row says whether you can do it yourself. Anything marked "needs a developer" is worth quoting for rather than attempting.

Nothing here is generic advice: every row exists because a specific measurement in this report triggered it, and a row is absent when the thing it would fix was not found.

A paid report includes a CSV of every measurement, so the numbers stack into a history you own outright, and one free re-audit once you have made the fixes, which gives you a genuine before-and-after rather than a promise. Both are keyed to your own run, which is why they are not live links on this sample page.

See what each tier measures

Self-serve checkout is not live yet. Tell us which tier you want on the contact form and Annette sets the audit running by hand, with the finished report back to you inside one business day.

• The sitemap is an index, which is normal and read correctly by search engines. The page count above is the total across its 1 child sitemap.