Tool Report or Real Audit? How to Tell the Difference

Quick answer: An automated SEO audit is a crawler’s list of everything it flagged. A real audit is a person deciding which of those flags matter for your store, proving each one with evidence, and putting the few that affect sales in order. You can tell which you were handed with five checks: is it prioritised, is each finding evidenced, does it show Shopify knowledge, does it say what to leave alone, and does each fix have an owner and a way to check it worked. On Shopify, “blocked by robots.txt” on the cart and checkout is usually safe to ignore once you have checked it, while a noindex on a page that sells is not.

Someone has sent you an SEO audit: from an agency you are weighing up, a freelancer you hired, or a free report that arrived after you typed your URL into a form. It has a score, a lot of red and amber, and an impressive page count. How do you know if it is any good?

Most pages on this topic compare automated and manual audits on speed and cost, and land on “use both”. That is true and not much help when you are holding a specific document. This post is about judging the deliverable: five checks you can run on any audit, then two real warnings on a Shopify store, one you can safely ignore and one you should never ignore.

It is written for the person receiving the audit, not producing it. For the audit steps themselves, see our Shopify technical SEO audit checklist. Every Shopify and Google page cited here was checked on 31 October 2026, and both change their documentation, so recheck before you act.

What a crawler can know, and what only a person can

A crawler works like a very fast, very literal visitor. It follows links and records what it finds: status codes, titles, headings, canonical tags, robots rules, redirects. That is real, useful data, and most good audits start from a crawl. A crawl presented as the audit is the problem.

What a crawler cannot know is anything outside the pages themselves:

• Which pages earn money. It has no access to your orders. A product that sells daily and a discontinued one nobody visits look the same.
• What you meant. It can report a noindex tag. It cannot know whether you put it there on purpose.
• What Google decided. It records what your pages say. Search Console records what Google did with them, and the two can disagree.
• What Shopify generates for you. It sees the output, not the platform rules behind it.

Even Google’s own tools say as much: its URL Inspection help says the live test does not check for all indexing issues, such as duplicates or alternate pages, or for manual actions. A third-party crawl needs a person to read it too.

So the question is not “automated or manual”. It is whether someone joined crawl data, Search Console, your sales and knowledge of how Shopify works into a decision. The five checks below look for that.

The five-check test

Run these on any audit, from any source. Each takes a few minutes.

Check 1: Is it in order?

Look at the first page after the summary. Is there a short list of what to do first, with reasons, or a table of issue types? My decision rule: more than about 20 findings and no ranked “do first” list of five to ten means you have an export. A real audit keeps the long tail in an appendix.

“High, medium, low” labels from a tool describe how its rule is written, not your store. A ranking you can trust says why: this affects pages that earn revenue, or blocks indexing of your main collections, or is a ten-minute fix. A score out of 100 deserves the same suspicion: I read it as how many of one tool’s rules fired, not how well your store sells.

Check 2: Can you see the proof?

For any finding that matters, you should be able to answer three questions from the document alone. Where was this seen? How can I see it myself? What else confirms it?

Take “42 pages have duplicate content”. That is a count, not a finding. A finding names the URLs, says what they duplicate, shows the canonical each one declares, and says which URL Google chose, which URL Inspection shows as the Google-selected canonical. Google says it might choose a different URL from the one you declare, and a report that never mentions Search Console cannot say which it chose.

Ask for the second source. If an audit claims pages are not indexed, the proof is in the Page indexing report, not the crawl. If it claims a page is slow, the proof is a dated measurement on that page.

Check 3: Does it know it is looking at Shopify?

This is the quickest check. Shopify’s help centre says your sitemap.xml and robots.txt files are generated automatically, and that canonical tags are auto-generated and added to pages to prevent duplicate content appearing in search results (as at 31 October 2026). It also says editing robots.txt.liquid is an unsupported customisation, and warns that incorrect use can result in loss of all traffic.

So some recommendations are tells:

• “Create an XML sitemap.” Shopify already did. A Shopify-aware audit checks that it is submitted in Search Console; it does not tell you to build one.
• “Add canonical tags to your pages.” Shopify adds them. A good audit checks whether any point at the wrong place.
• “Edit your robots.txt to block these URLs.” Possibly, but a good audit says Shopify’s default already blocks several URL patterns, that editing is unsupported, and why the change is worth the risk.
• “Install this plugin to fix everything.” Shopify’s own FAQ lists an app introducing title tag logic that does not apply to some page types among the common causes of a missing page title, so a recommendation to add an app should name what it adds and what it could affect. See what Shopify SEO apps do.

A report that could have been written for any platform, with your store name swapped in, is a template.

Check 4: Does it tell you what to leave alone?

A crawler has no judgement, so it treats every flag as a fault. A person discards most of them and tells you which and why. If an audit never says “this flag is fine”, either the author did not look closely or wants the list to look long. Two examples follow below.

Check 5: Does every fix have an owner and a test?

For each recommendation you should be able to tell who does it (you, your developer, the agency), roughly how big it is, and how you would check it worked. “Fix canonical issues” fails all three. “Remove the noindex tag from the collection template; check the page in URL Inspection afterwards; expect Google to take its own time to recrawl” passes.

Google says indexing after a request typically takes a day or so but can take much longer, so treat any fixed date for a fix to show up in Google as a guess. Comparing providers? Our guide to choosing a Shopify SEO agency in Australia covers what else to ask.

A worked example: what 187 warnings become

Here is the triage a real audit does. The numbers are illustrative arithmetic, not from any store and not typical. Say an export lists 187 flagged items. A person reading it alongside Search Console and the sales report sorts them into four buckets:

The audit you receive should name the 14 and rank five of them, give a rule for sorting the 38, and say plainly that the other 135 need no action. The export gives you 187 and leaves the sorting to you.

The cosmetic bucket is where Shopify’s guidance and Google’s differ. Shopify’s SEO overview recommends titles under 60 characters and meta descriptions under 160. Google’s documentation says there is no limit on either, but that long ones are truncated in results. A long title is a tidy-up for when you are editing the page anyway, not an error.

A warning that is safe to ignore: “blocked by robots.txt”

Why it appears. Expect to see this flag on a Shopify store. Shopify’s default robots.txt asks search engines not to crawl certain URLs. Its help page lists the cart, checkout, account and orders pages, collection URLs with a plus sign in them (filtered pages that duplicate collection content) and collection URLs containing sort_by (sorted pages). Shopify’s page on hiding pages gives the cart as its example and says it is unique to each customer and not useful in search results. A report that lists these as errors is reporting Shopify’s defaults, not a fault in your store.

Why it is safe. Nobody wants the cart or checkout in search results, and sorted and filtered pages are views of collections you already have. See our guide to filter URLs and index bloat.

How to confirm it is safe on your store. Three steps, ten minutes:

1. Open yourstore.com/robots.txt and compare the Disallow lines with the patterns above. The file on Shopify’s demo store on 31 October 2026 listed more rules than the help page does, such as cart.js and a recommendations path, so read your own.
2. Check that no blocked URL is something you want found: a product, a collection, a blog post, a page.
3. Check whether anyone has customised robots.txt.liquid. If so, the flag is no longer safe by default, because Shopify says a faulty edit can cost you traffic. Our Shopify robots.txt guide covers the file.

If all three come back clean, ignore the flag, and a good audit says so. If the report lists it as a “critical error”, read the rest of the document with more care.

The Search Console twin. The same logic applies to the Page indexing status “Alternate page with proper canonical tag”. Google’s help says the page correctly points to a canonical that is indexed, so there is nothing you need to do. My proviso, not Google’s: check that the canonical is the page you want indexed. We cover it in alternate page with proper canonical tag on Shopify.

A warning that is not safe to ignore: noindex on a page that sells

Now the opposite. The flag is “noindex”. Google’s documentation says the rule means do not show this page in search results. That suits an internal test page. You do not want it on a collection that brings in orders.

Why a crawler cannot judge it. The crawler reports a fact: this page has noindex. It cannot know whether you meant that.

How it happens on Shopify. Shopify’s help page on hiding pages from search engines describes three ways. You can add a noindex robots meta tag in theme.liquid, often inside a condition so it applies to specific pages. You can set a product to unlisted, which makes it discoverable only by direct URL and hides it from internet search, the sitemap, collections and recommendations. Or you can set the seo.hidden metafield to 1, which hides a product, page or blog post from sitemaps, search engines and storefront search. Each is legitimate and each is easy to leave behind: a tag added during a redesign, a metafield set while a product was being prepared. Shopify adds that if you cannot edit seo.hidden in the admin, a third-party app is using that key.

One caution on the theme method, which is my reading of two Shopify pages, not a Shopify warning. Shopify’s example condition checks a page handle with the Liquid contains operator, which Shopify’s Liquid documentation describes as checking for a string inside another string. A short handle in that condition could therefore match longer handles that include it. If your theme has a condition like this, view the source of neighbouring pages and confirm what it hides.

How to find out if it is hurting you. This takes two lists, and a real audit joins them:

1. List A: every URL carrying noindex. The crawl has this, and so does the Page indexing report in Search Console, which has a status for pages with a noindex directive.
2. List B: the pages that earn revenue, such as your top products and collections by sales over the last 90 days from Shopify.

Any URL on both lists is an action item. A crawler has list A only. A person with access to your sales has both. That is the whole difference between the two documents.

What to do. Decide whether the noindex is intentional. If not, remove the tag or clear the metafield. Shopify’s help says clearing the metafield value makes the item visible again. Then inspect the URL in Search Console to confirm the page is indexable. Google only sees a robots meta tag or header when it crawls the URL, so the change takes effect after a recrawl. Google says requesting indexing does not guarantee the page will appear, and that indexing typically takes a day or so but can take much longer, so do not promise yourself a date.

One trap: do not block a URL in robots.txt and also noindex it. Google says a blocked page’s noindex rule is never seen, so it cannot work.

The middle ground: 404s

Some flags sit between these two. The most common is “404 not found”. Google’s Page indexing help says 404 responses are not necessarily a problem if a page was removed without a replacement (and to use a 301 redirect if it moved), and its crawling documentation says it does not index URLs that return a 4xx. So a list of 38 broken URLs is not an emergency.

It becomes one when the URL used to matter: it had traffic, links from other sites, or was a product you still sell. This is common after a store move. Shopify’s FAQ lists a recent migration where the old site’s redirects were not implemented among the reasons a store may not appear in search results, and its SEO overview tells merchants to set up 301 redirects when moving or discontinuing products. A real audit sorts the 404s by whether they once earned something. Start with our Shopify migration SEO checklist.

What Google itself says to expect from an audit

Google’s guidance on hiring an SEO is worth reading before you judge anyone’s deliverable. Three points matter here (checked 31 October 2026):

• Estimates, not promises. Google says an SEO audit should give you realistic estimates of improvement and of the work involved, and that “No one can guarantee a #1 ranking on Google.”
• Tool output needs checking. Google says it does not evaluate or endorse third-party SEO tools, and suggests checking a third-party audit’s recommendations against its official guidance before big site changes.
• Read-only access at this stage. Google advises read access to Search Console, not write access, while the audit is under way.

Questions to ask whoever sent you the audit

A competent provider can answer each of these in a sentence:

1. “Which five findings would you fix first, and why those?”
2. “How did you confirm the top three, and can I see the Search Console view?”
3. “Which flags did you decide to ignore?”
4. “Which of these does Shopify handle automatically?”
5. “Who does each fix, and how will we know it worked?”
6. “Can I see the raw crawl export as well?” The crawl is an input to their thinking, so a good provider will share it.

Warning signs in the answers: a score as the outcome, “everything on the list” as the first priority, a promise of rankings or traffic, or vagueness about what Shopify does for you. Our post on SEO agency red flags goes further.

What to do with a tool report you already have

Do not throw it away. A crawl export is useful once someone triages it, and an hour gets you most of the way:

1. Sort by type. Group rows into robots-blocked, canonical, noindex, 404 and redirect, title and meta, and everything else, then cross out the by-design group using the checks above.
2. Run the noindex test. Overlay the noindex list on your top-selling pages.
3. Check Google’s view. Open the Page indexing report, filter by your submitted sitemap, and see what Google says about the pages that matter. Our guide to the Crawled, currently not indexed status shows how to read it.
4. Write your own top five. If you cannot rank them, take the document to someone who can.

Holding a report you are not sure about? Tell us about your store.

Key takeaways

• An automated audit lists what a crawler flagged. A real audit decides which flags matter for your store, proves them, and puts the few that affect sales in order.
• Run five checks: order, evidence, Shopify knowledge, restraint and follow-through. A crawler cannot see your sales, your intent or Google’s decisions.
• “Create a sitemap” and “add canonical tags” are tells. Shopify’s help says it generates both automatically.
• URLs blocked by robots.txt are usually safe to ignore on Shopify, because the default blocks the cart, checkout, account and sorted or filtered collection URLs. Verify it on your own file first.
• Noindex on a page that earns revenue is never safe to ignore. Only joining the noindex list to your sales list reveals it.
• Google says an audit should give realistic estimates, and that no one can guarantee a #1 ranking.
• Ask which flags the author decided to ignore. If the answer is none, nobody did the judging.

FAQs

What is the difference between an automated and a manual SEO audit?

An automated audit is software crawling your site and flagging anything that breaks its rules. A manual audit is a person reading that data alongside Search Console, your sales and the platform’s behaviour, then deciding what matters. Many good audits use both. What you can check is whether the result is prioritised and evidenced.

Is an automated SEO audit worthless?

No. A crawl finds broken links, redirect chains, missing titles and blocked URLs quickly, and it tracks changes over time. It becomes worthless only when presented as the finished audit.

Which Shopify audit warnings are usually safe to ignore?

On a default Shopify store: URLs blocked by robots.txt for the cart, checkout, account and sorted or filtered collection pages, the Search Console status “Alternate page with proper canonical tag”, and long titles or meta descriptions. Verify each on your own store first, because customised robots.txt or canonicals change the answer.

Which warnings should I never ignore?

Noindex on a page that earns revenue, important pages missing from Google’s index, canonicals pointing at the wrong page, and 404s on URLs that used to earn traffic or links. Each needs a person to check it against your sales and Search Console data.

Should an SEO audit promise rankings or traffic?

No. Google says no one can guarantee a #1 ranking, and that an audit should give realistic estimates of improvement and of the work involved. A forecast is fine if it says what it rests on. A guarantee is a red flag.

Not sure which warnings on your store matter?

Our Shopify SEO service includes a technical SEO audit covering Core Web Vitals, crawlability, indexation and schema markup, with every recommendation tied to revenue, never to vanity metrics. If you would rather talk it through, tell us about your store or book a free 30-minute diagnostic. We look at your store before we recommend anything, and we tell you honestly if SEO isn’t the right move. No pitch.