In September 2026, I audited my own site. Search Console showed copies of my pages on the www version of the domain, one of them with 642 impressions. It also showed 18 URLs that other sites still linked to, all returning errors, and four blog posts competing with stronger guides for the same searches. None of it showed on the blog itself.
A content audit is how you find that, and it works the way a closet clear-out does. You take everything out, look at each piece, and decide what goes back in, what gets mended, what gets combined, and what leaves. On a SaaS site, it also checks each page against today’s product, which changes faster than old posts do.
I’m a content strategist at Breaking B2B and a B2B SaaS writer, and I’ve run audits and content roadmaps for SaaS teams. Below is my method as of October 2026: a decision for every page, the product check most audits skip, my own audit with what happened after, and a free Google Sheet that suggests a decision for each URL.
What is a content audit?
A content audit is a page-by-page review of everything you’ve published, ending in one decision per page: keep it, fix it, merge it into a stronger page, or cut it. For a SaaS company, the review also checks each page against the product, pricing and positioning as they are today.
An audit starts from an inventory, and the two are different jobs. Think of the inventory as the list of everything in the fridge, and the audit as deciding what’s past its date. Nielsen Norman Group describes the audit as the step that evaluates the quality of the content you’ve listed.
A content audit isn’t an SEO audit either. An SEO audit asks whether Google can crawl and index your site. A content audit asks whether each page deserves to stay on it.
When a SaaS team needs a content audit
In audits I’ve run for SaaS teams, the trigger is usually one of these:
- You repositioned or changed pricing. Older posts still describe the old product to the people most likely to buy the new one.
- A product or feature was retired. Pages promoting it send buyers looking for something that no longer exists.
- You publish every month and traffic stays flat. New pages may be competing with old ones for the same searches.
- Two of your pages show up for the same search. Search Console’s Pages tab, filtered to one query, shows it.
- You’re about to migrate or redesign the site. Every page you cut first is one fewer redirect to map.
I’d run a full audit about once a year, plus a lighter pass after any repositioning, pricing change or product retirement.
Step 1: Build the inventory
List every URL that marketing owns. On a SaaS site, that’s usually more than the blog:
- Blog posts and guides
- Comparison and alternatives pages
- Integration and use-case pages
- Landing pages, including old campaign pages nobody links to anymore
- Gated assets and their landing pages
- Help docs, if marketing owns them
Start from your sitemap, then run a crawl to catch what the sitemap misses. The crawl is how I found the www copies on my own site, since they never appeared in the sitemap at all. A content audit tool or crawler does this in minutes for most SaaS sites.
For each URL, record the page type, the topic, the funnel stage, when it was last updated and who owns it. Those columns are already set up in the free template further down.
Step 2: Pull the data that decides
Four kinds of data decide most calls:
- Search Console clicks and average position. Pull them by page for the last three months and the same three months a year earlier.
- Conversions from GA4. Demo requests, sign-ups and any page that assists them.
- Backlinks. Pages that other sites link to need a redirect target if they go, never a dead end.
- The pages sales sends. Ask your sales team which pages they share with prospects. A page with 20 visits a month that sales sends to every late-stage deal stays, the way you’d keep the old jacket you still wear every week.
Judge pages on clicks, not impressions. Impressions are people walking past your shop window, and clicks are people who come in. AI visibility trackers make the window count worse, because they run the same prompts every day and each one registers as a passer-by. On my site, tracker prompts made up 53% of long-query impressions, with zero clicks.
Read weekly trends too, not single days. When I checked one of my own refreshes, a jump from about 1,000 to 11,000 impressions turned out to be one day. The weekly numbers showed a smaller, steadier rise, from about 240 to about 1,000 a week, which is the real result. I cover how to read those trends in spotting content decay in Search Console.
Step 3: Check every page against the product
This is the step SaaS audits most often skip, and it’s the one buyers notice first. A page doesn’t go stale when the calendar says so. It goes stale the day the product ships something that contradicts it, like a restaurant menu still listing a dish the kitchen stopped making months ago.
Check each page against five sources:
| Check | Where to look | What to do if it’s wrong |
| Features | The changelog and release notes | Rename, remove or update every feature the page mentions |
| Plans and prices | Your pricing page, monthly view | Update plan names, prices and limits, or link to the pricing page instead of quoting it |
| Positioning | The homepage and the sales deck | Rewrite the intro and the product sections so they say what sales says |
| Competitors | Their sites and pricing pages | Remove or update competitors that were acquired, renamed or shut down |
| Screenshots | The current product | Retake any screenshot of an old interface |
Third-party pages drift the same way. CoSchedule’s marketing team once emailed me corrections to my CoSchedule review, which they said was one of the pages AI assistants cite most about their product. Plan names had been retired since I wrote it. I checked each correction on their pricing page and updated the review the same day. If someone else’s review of your product can drift that far, your own posts from two releases ago probably have too, and AI answers keep quoting them until you fix them.
Step 4: Decide keep, fix, merge or cut
Once you have the data and the product checks, most pages sort themselves with four questions, asked in order:
- Keep. The page brings clicks or conversions and still matches the product. Give it light upkeep and check it again next audit.
- Fix. The page belongs on your site, but something is off. It’s out of date against the product, it’s losing clicks, or it sits on the first three pages of Google without reaching the top results. Update it on the same URL. If you need to work out why a page slipped, that’s a content decay diagnosis, and the fix list depends on the cause.
- Merge. Two pages answer the same search, or a thin page belongs inside a stronger one. Merging is like pouring two half-empty jars of the same spice into one: move the best sections into the stronger page, then redirect the weaker URL to it.
- Cut. I delete a page when it no longer serves a purpose. The two cases I see most are a page promoting a product that no longer exists, and an article so far from a purchase that it will never bring meaningful traffic. If it has backlinks or any visits, redirect it to the closest page instead of letting it return an error.
Usually, though, I merge or redirect rather than delete. A redirect works like a forwarding address when you move house: the links and visitors that still arrive at the old URL end up somewhere useful.
Some pages stay even when they look like they shouldn’t. On my own site, a group of off-topic posts about tools like Payoneer and to-do apps brought 833 of my 2,878 clicks over three months, or 29%. They don’t match what I sell, but I kept them and judge them on conversions, not on topic fit. Before you cut anything, ask whether it brings traffic, conversions or backlinks, and whether sales sends it.
A real audit, page by page: my own site
Here’s what I found on my site, and the call I made on each:
The redirects and folded pages were housekeeping. The merges were the real test, so I compared each surviving page’s Search Console numbers for the four weeks before the audit with the four weeks after it:
| Surviving page | Clicks before → after | Impressions before → after | Position before → after |
| /blog/seo-content-writing (absorbed one post, then refreshed) | 18 → 80 | 15,200 → 22,018 | 17.3 → 8.9 |
| /b2b-seo-consultant (absorbed one landing page, then refreshed) | 10 → 7 | 3,270 → 4,617 | 21.6 → 17.3 |
| /b2b-saas-content-writer (absorbed one landing page) | 19 → 19 | 2,045 → 2,401 | 17.1 → 16.3 |
| /best-seo-agencies-for-startups (absorbed one landing page) | 5 → 7 | 2,255 → 2,238 | 11.0 → 10.0 |
| /blog/hire-content-writer (absorbed three posts) | 5 → 0 | 1,191 → 18 | 8.6 → 5.1 |
The SEO content writing guide is the clear win. I merged a smaller post into it and refreshed it two days later, so the merge and the refresh share the credit. Its average position roughly halved and clicks more than quadrupled.
The landing page merges held steady. Impressions and positions stayed flat or improved slightly, and clicks moved by a few either way, which is what you want from consolidation: the surviving pages lost nothing by absorbing the weaker ones.
The hiring guide is the miss. It averaged 20 to 50 impressions a day before the merge, and almost nothing from four days after it. Its before-number also includes a three-day spike of 569 impressions on one search, so the normal level was lower than 1,191, but the drop is still real. A month later, Google was still showing the old, redirected URLs in its results instead of the guide they now point to. The page is indexable, so the next step is Search Console’s URL Inspection and a request to reindex it. It’s a useful reminder that a merge can take Google weeks to process, and that you should check every merged page by itself rather than the site total.
Step 5: Roll it out without losing rankings
Roll changes out in this order:
- Move the best parts first. Copy the sections, FAQs and examples you want to keep into the surviving page before you redirect anything. I moved the unique FAQs from each landing page before merging it.
- Use permanent redirects. Google says a 301 or 308 tells it a page has permanently moved, and that its indexing pipeline uses the redirect as a signal that the target should be the canonical page. Point each redirect at the closest page in one hop, not at the homepage by default.
- Update your internal links. Point links straight at the final URL. On my site, that meant repointing links in 25 posts.
- Update the sitemap and navigation. Remove cut and merged URLs so you’re not inviting Google back to them.
- Write down the date, then check four weeks later. Compare each changed page with its own four weeks before, the way the table above does.
Google treats deleting as the last option, not the first. Its guidance on core updates says “Deleting content is a last resort”, and adds that removing unhelpful content can help the good content on a site perform better. Merging and redirecting keeps what a page earned and still removes the duplicate.
Audit for AI answers too
Run your buyers’ main questions through ChatGPT, Perplexity and Google’s AI Mode, and note which of your pages get cited and which third-party lists mention you. In my experience, the pages AI answers cite tend to be the ones that already rank and answer the question near the top, so a merge that puts your best answer on one URL can help there too. I cover the rest of that work in getting a SaaS brand into ChatGPT’s answers.
Don’t judge a page’s AI visibility by its Search Console impressions, for the reason in step 2. The trackers inflate them without sending anyone.
What to tell leadership
Cutting pages can look like throwing away assets, especially to someone who approved the budget for them. Show what the library looks like afterward instead:
- Pages before and after. How many you kept, fixed, merged and cut.
- Clicks and conversions on the pages you kept. Usually most of them, which shows that the cuts removed pages nobody used.
- The pages now correct about the product. The number of pages fixed against the changelog and pricing page.
- The four-week check. Clicks on the changed pages before and after, misses included.
The Summary tab in the template below builds that view for you. For the numbers to report every month after that, I use SEO KPIs that tie content to pipeline.
Free content audit template
I built a free Google Sheet that runs this method. List your pages, add the data from Search Console, GA4 and your SEO tool, answer the product checks, and it suggests keep, fix, merge or cut for each URL, using the same four questions as the chart above. Make a copy of the content audit template.
It has four tabs:
- Start here. The steps, in order.
- Audit. One row per page, with the product checks, an automatic suggestion, your own decision and the redirect target.
- Rules. The thresholds for low traffic, falling clicks and the position range, which you can change to suit your site.
- Summary. Pages and clicks by decision, plus the before and after for the pages you changed.
The grey rows are made-up examples for a project management tool. Delete them once you’ve seen the format.
Content audit FAQs
How often should you run a content audit?
About once a year for a full audit, plus a lighter check after a repositioning, a pricing change or a product retirement. Sites publishing several posts a week tend to need a quarterly look at their newest pages.
Does deleting pages hurt SEO?
Not if the pages weren’t bringing anything. Google’s guidance treats deleting as a last resort and says removing unhelpful content can help the rest of a site. Redirect any page that has backlinks or traffic to the closest page instead of deleting it outright.
How long does a content audit take?
For a library of about 100 pages, usually a few days, most of it on the product checks and the decisions. The data pulls are quick. Larger libraries take longer mainly because more pages need a human read.
What’s the difference between a content audit and an SEO audit?
An SEO audit checks whether Google can crawl, index and understand your site. A content audit decides whether each page deserves to stay, and in what shape.
If your blog has grown faster than your pipeline and you want a second opinion on what to keep, fix or cut, book a call with me. I’m a B2B SaaS writer and strategist, and on the call, we’ll look at your Search Console data and pick the first pages to fix or merge.
