Most SEO audits produce a long list of page-level defects and no answer to the question the business actually asked.
An attention asset audit checks whether the company owns durable access to a relevant audience, and what condition that access is in. It uses the same technical evidence a ranking audit collects, and then asks a different question of it: not "can this page rank," but "does the business still have this audience if the channel reprices tomorrow."
The technical work is not optional. A site that cannot be crawled has no search visibility to own. But crawlability is a precondition, not the asset. That distinction is the whole point of treating marketing as an asset class: the deliverable is not a defect list, it is a judgment about condition, ownership, and what to invest in next.
A ranking checklist audits pages. An asset audit audits access.
The two produce different artifacts and different decisions.
| A ranking audit asks | An asset audit also asks |
|---|---|
| Can search engines crawl and index this page? | Which durable buyer questions do we have access to, and which do we not? |
| Are the meta elements and headings correct? | Does this visibility reach the people who actually buy, or just people? |
| Is the backlink profile clean? | Would we still reach this audience if the algorithm changed distribution? |
| Are the errors fixed? | Who owns this asset, how is its condition measured, and when was it last maintained? |
| What is the score? | What did the audit change about where the next dollar goes? |
A ranking audit that finds nothing wrong can still describe an asset in poor condition: technically healthy pages, ranking for queries nobody with budget is typing. That is the failure the checklist format cannot surface, because it grades the page rather than the access.
Search visibility is one of five marketing asset classes, and it sits in the attention class alongside owned email, first-party customer insight, and recognized authority around a problem. Auditing it in isolation from the rest of the portfolio produces local fixes and no strategic answer.
Start with the four condition questions
Before opening a single tool, answer these. They decide whether the technical work below is worth doing at all.
- Can we reach these people without buying access again every time?
- Do we understand why they pay attention?
- Is the relationship becoming more direct and more useful?
- Would the audience insight survive if a vendor or an employee left?
If the honest answer to the first question is no, the site does not have an attention asset yet. It has an advertising channel with a website attached, and the correct next investment is probably not a technical audit.
Layer 1: Can the asset be reached at all?
This is the precondition layer, and it is where a technical audit earns its keep. Nothing else matters if this is broken.
- Crawl access. Server errors, broken pages, and anything blocked in
robots.txtthat should not be. Server errors are the emergency; broken pages are the same day; redirect chains and soft 404s can wait a week but should not wait a quarter. - Index status. Check what is actually indexed against what should be. Canonical tags pointing at the wrong page and conflicting index directives are the two failures that hide longest, because the page looks fine when you load it.
- Redirect hygiene. Every redirect should reach a live page in one hop. Chains waste crawl budget and dilute the signal they were supposed to preserve. Content that was deliberately removed and has no successor should return 410, not a redirect to a hub page, which behaves like a soft 404.
- Sitemaps. Google's own guidance is that a sitemap matters most when a site is large, when pages are poorly linked to each other, or when the site is new with few external links. Most sites being audited meet at least one of those conditions.
- HTTPS and canonical host. One protocol, one host, one trailing-slash convention, resolved in a single hop.
Record what you fixed and what you found already working. A layer that passes is evidence about asset condition, not a wasted step.
Layer 2: Is the asset worth reaching?
Technical health with no relevance is an expensive way to be invisible to the right people.
Match against intent, not word count. The useful question is whether the page answers what the searcher was actually trying to resolve. Length correlates with rank in aggregate because thorough answers tend to be longer, not because length is the mechanism. Writing to a word count produces padding, and padding is what gets outranked.
Audit for durable questions. An attention asset is built on questions buyers will still be asking in three years. Roundups, year-stamped lists, and tool comparisons decay on a schedule and have to be rebuilt continuously. That is rented visibility with extra steps.
Decide refresh or retire, explicitly. For every underperforming page: does it answer a question that still matters, to a person who still buys? If yes, refresh it, replace the stale references, and fix the heading structure so the answer is findable. If no, merge it into a page that does, or retire it. Keeping it costs crawl attention and internal link equity, and it makes the site look less maintained than it is.
Check for cannibalization. Two pages competing for the same query split the signal and neither wins. One search cluster should resolve to one canonical page, with supporting pages that deepen it rather than duplicate it.
Layer 3: Does it hold up on the device people actually use?
Google indexes the mobile version of a site. If the mobile experience is a degraded copy of the desktop one, that degraded copy is what gets evaluated.
- Core Web Vitals. Largest Contentful Paint, Interaction to Next Paint, and Cumulative Layout Shift. Note that Interaction to Next Paint replaced First Input Delay in 2024, so any audit template still measuring FID is out of date.
- Readable body type. 16 pixels or larger. This is an accessibility floor before it is a ranking consideration.
- Tap targets. At least 44 pixels, so the interface works for someone using a thumb on a moving bus.
- No horizontal scroll. Content wraps to the viewport rather than requiring a pinch.
- Interstitials that can be dismissed without covering the answer the person came for.
Field data beats lab data when you can get it. Lab scores tell you what a machine experienced once, in one place, on one connection.
Layer 4: Does the structure concentrate authority or scatter it?
Internal linking is the part of a technical audit that most directly shapes what the asset is worth, and the part most often reduced to "fix broken links."
Fix the dead links, then ask the better question. Broken internal links usually mean pages were removed without a redirect decision, which is a maintenance signal about the whole property, not an isolated defect.
Give every page a job. A page that exists because a channel expected a page has no job, earns no links, and dilutes the pages that do. Pages without a job should be merged or removed.
Find the orphans. Pages with no internal links pointing at them are invisible to both crawlers and readers. This is usually one template-level omission rather than dozens of individual mistakes. Fixing the template fixes them all at once.
Vary the anchor text descriptively. Link with the phrase that describes the destination, not the same two words every time.
Link from strength to need. Established pages should pass attention to newer ones deliberately, not accidentally.
Layer 5: Is the off-site signal real?
Backlinks. Most low-quality links are simply ignored rather than penalized, so a disavow file is a last resort and not a routine hygiene step. Google's stated position is that links acquired primarily to manipulate rankings are link spam and get nullified at scale. Plan on that being true. The audit question is whether anyone with a real audience has a reason to reference the work, and if not, what would have to be true for them to.
Measure the profile at the URL level, not just the domain. Domain totals hide the thing you need to know: which specific pages hold the equity, and whether those pages still exist. Retired pages holding real links are the highest-value redirect targets on most sites, and they are invisible in a domain-level summary.
Business listing consistency, where the business is local. Name, address, and phone number identical everywhere they appear, with an owned and complete Google Business Profile. Keep a single record of every place the details are published, because the moment you need it is the moment something has changed.
The questions a technical audit never asks
These are what make it an asset audit.
Who owns this? Not who works on it. Who is accountable for its condition and authorized to change it.
What is its condition, in a number? Coverage of the durable questions that matter. Share of qualified entrances rather than total sessions. Freshness of the pages that carry the most weight. Whether the trend is improving.
What is the concentration risk? If a large majority of qualified pipeline arrives through one channel, the business does not own an attention portfolio. It has a dependency, and the platform sets the price. This is the same portfolio logic that decides which asset to build next in the marketing operating model.
Does the audit change an investment decision? An audit that produces a defect list and no reallocation was a technical exercise. The output should be a ranked set of decisions with owners, not a PDF with a score.
What evidence would settle it? Agree before the work starts on what would count as improvement, so the review is not an argument about interpretation later. That discipline is the same one behind evidence before claims.
Maintenance is the asset, not the audit
A one-time audit produces a snapshot. An asset needs a condition check on a rhythm, and the rhythm is what keeps the value from quietly decaying between projects.
| Cadence | Check |
|---|---|
| Monthly | Index coverage, crawl errors, and the pages carrying the most qualified entrances |
| Quarterly | Content condition on the durable questions, internal link structure, redirect and retirement decisions |
| Twice a year | Off-site signal, competitive coverage of the questions you want to own |
| Annually | Full audit, portfolio reallocation, and an honest concentration review |
The schedule matters less than the fact that someone owns it and something changes because of it.
Where this fits
Search visibility is one attention asset among several, and it is the one most often mistaken for the whole marketing function. It belongs inside a website marketing strategy that also builds owned audience, conversion paths, and proof, so that no single channel decides whether the business gets found.
Score the whole portfolio before committing to a channel-level project. Take the marketing assessment to find which asset class is actually the constraint.
If the answer is that the work needs to be executed rather than explained, that is an implementation engagement, and it is what RedShift is for.