The short version: Before you plan a single new blog post, crawl your existing site, list every page you have, and work out what each one is for. Most businesses I audit have no idea how many pages they own, which ones rank, or which ones quietly cancel each other out. Fix that first and your content plan writes itself in half the time.
Why I stopped writing content calendars first
For years I did content planning the way everyone tells you to. Keyword research, a spreadsheet of topics, a nice colour-coded calendar for the quarter. Felt productive. Looked professional in a client meeting.
Then in 2023 I took on a client (a training company in the Midlands, I’ll call them the “skills firm” since they’d rather I didn’t name them) who had 340 pages on their site. Nobody there could tell me what most of them were. Not the marketing manager, not the founder. We ran a crawl and found 61 pages that hadn’t been touched in over four years, 14 pages that no longer appeared anywhere in their own navigation, and three separate blog posts all targeting the exact same phrase, “leadership training course UK,” written by three different freelancers over three years who’d clearly never seen each other’s work.
That’s not a content problem. That’s a map problem. You cannot plan sensible new content on top of a mess you can’t see. So now, with every client, mapping the site comes before any topic gets written down.
What “mapping your website” means
A sitemap.xml file is not a map. It’s a list Google might crawl, and half the time it’s out of date the day after someone generates it. A real map answers different questions:
- How many pages do we have, indexed and not?
- What is each page trying to do (inform, sell, capture an email, rank for a phrase)?
- Which pages get traffic and which get none?
- Which pages link to each other, and which sit there completely orphaned?
- Where do we already say the same thing twice, or three times?
Nobody enjoys answering these. It’s slower and less fun than brainstorming blog titles. But it’s the difference between adding content and adding noise.
The tools I use, and why the free ones are usually enough
You don’t need an expensive suite for this. I’ve written a full breakdown of the free website mapping and sitemap tools that work in 2026, but here’s the short version of my own kit:
- Screaming Frog’s free version, which crawls up to 500 URLs and gives you titles, meta descriptions, word counts, internal links, and status codes in one export.
- Google Search Console’s Pages report, to see what Google has indexed versus what you think exists.
- A plain spreadsheet, because visual mind-map tools look impressive and then you can’t sort or filter anything, which is the whole point of doing this.
For anything under 500 pages, that combination costs you nothing but time. The skills firm client had 340 pages, so the free Screaming Frog crawl covered everything in one go. If you’re over that limit, the paid tier is around £180 a year, which is still cheaper than one wasted blog post written for a page that doesn’t exist anymore.
The step-by-step I run
Here’s the process, in order, no skipping steps because step three feels boring:
- Step 1: Crawl the whole site. Get every URL, live and dead, into one list. Include status codes so you catch 404s and old redirects hiding in the navigation.
- Step 2: Pull the traffic data. Twelve months of Google Search Console clicks and impressions against every URL. This is where you find out that 70 pages get one click a month between them.
- Step 3: Tag each page by intent. Four columns, nothing fancier: informational, commercial, transactional, navigational (things like your contact page). This forces you to admit what a page is for, which most people have never written down.
- Step 4: Group by topic cluster. Not by date published, not by which freelancer wrote it. By what a searcher is trying to do. This is the step that exposes duplication.
- Step 5: Map the internal links. Which pages link to which. Screaming Frog’s export gives you this. Anything with zero inbound internal links is an orphan, and orphans almost never rank no matter how good the writing is.
- Step 6: Mark every page as keep, merge, kill, or update. Four buckets. Be blunt about it. This is where most people flinch, and it’s exactly where the value sits.
That’s it. Six steps, one afternoon for a site under 500 pages, longer if you’re a proper e-commerce catalogue. But it’s a finite job with an end point, unlike a content calendar which never really finishes.
The bit nobody wants to hear
Here’s what I’ve learned running this audit maybe forty or fifty times over the years, and it’s the part every other guide on this topic tends to skip past. A large chunk of your existing content is not an asset. It’s dead weight sitting on your domain, diluting the authority of the pages that could rank. On the skills firm’s site, 61 of 340 pages had earned precisely zero organic clicks in the previous year. Not low clicks. Zero. Those pages weren’t helping their SEO. They were competing with their own better pages for the same crawl budget and, in three cases, the exact same keyword.
Writers hate hearing this because it means admitting past work was wasted, and consultants hate saying it because “delete forty pages” is a harder sell than “here’s a shiny new content calendar.” But if you plan new content on top of a site carrying dead pages, you’re building a nice extension onto a house with a damp problem in the walls. It’ll look fine for a while.
We killed 38 of those 61 pages outright, merged 15 into stronger existing pages with proper 301 redirects, and kept 8 that were seasonal and would pick up traffic again. Within four months, the pages we kept and merged had climbed an average of 11 positions in Search Console, not because we’d written anything new yet, just because we’d stopped fighting ourselves.
Turning the map into an actual content plan
Once you’ve got the six-step map done, the content plan practically writes itself, because now you’re filling gaps in a structure instead of guessing at a blank page. Look at your topic clusters and ask, for each one, “what’s missing that a customer at this stage would need?” That’s your new content list, sorted by real gaps rather than whatever topic felt inspiring on a Tuesday morning.
This is also where I’d point you toward keeping your writing consistent once you start filling those gaps. A mapped site with fifteen different voices across fifteen pages still feels chaotic to a visitor even if the structure behind it is sound, and that’s exactly the problem I cover in why your AI content sounds like everyone else’s, and the one-page fix. Map first, then make sure whatever you add sounds like you and not like a template.
And be honest about what “success” means for each new page before you write it. I see this constantly: a business maps its site, spots the gaps, writes solid content to fill them, gets rankings within weeks, and then wonders why none of it turns into enquiries. That disconnect between ranking and selling is its own problem, and I’ve written about it separately in why your AI blog content is ranking but not selling anything. Mapping tells you where to build. It doesn’t automatically tell you how to sell from what you build, that’s a separate decision you make on top of the map.
What this looked like for one real client, start to finish
Back to the skills firm. After the audit and the cull, we had 260 pages left, tagged into eleven topic clusters, each cluster missing between two and five pages a genuine customer would need. Instead of a vague quarterly content calendar, we built a thirty-day sprint targeting the three clusters with the biggest commercial value: leadership training, apprenticeship funding, and workplace mental health training for managers.
I’ve written up the exact system we used to turn that into a running content engine rather than a one-off project in one client win, thirty days of content, the AI system that stopped me starting from scratch every week. The short version is that the mapping work did the heavy lifting. Once you know exactly which page needs what, writing new content stops being a creative struggle and turns into filling in blanks against a plan you already trust.
One more thing that came out of that same mapping exercise: three of the gaps we found weren’t blog posts at all, they were interactive tools their audience wanted, a funding eligibility checker and a simple training needs quiz. If your map turns up gaps like that, it’s worth knowing you don’t need a developer on the payroll to build them; I cover exactly how in how to create interactive content without a developer. A map tells you what’s missing. It doesn’t limit you to blog posts as the only answer.
A few honest warnings before you start
This work is unglamorous. Nobody has ever asked me for a case study on “the time I deleted 38 pages,” even though it moved the needle harder than most of the new content that followed. Expect resistance from anyone in the business emotionally attached to old pages, especially the founder who wrote the original “About Us” in 2019 and still quietly loves it.
Expect the crawl to take longer than you think if your site has more redirect chains than anyone realised, which is nearly always. And expect to find at least one embarrassing thing, a page still advertising a service you stopped offering two years ago, a pricing page with numbers nobody’s updated since before a price rise. You will find it. Better you than a customer.
Frequently asked questions
How often should I re-map my website structure?
Once a year at minimum for most small business sites, and immediately after any major change such as a rebrand, a new product line, or migrating platforms. Sites over 500 pages or with frequent content publishing should map every quarter, since orphan pages and duplication build up faster than most people expect.
Do I need paid software to map my site ?
No. Screaming Frog’s free tier covers up to 500 URLs, which handles most small business and mid-sized sites completely, and Google Search Console is free and gives you the real traffic data behind each page. Paid tools become worth it once you’re past 500 pages or need automated recurring crawls.
What’s the difference between a sitemap and a real content map?
A sitemap.xml is a technical file that tells search engines which URLs exist so they can crawl them. A content map goes further, showing what each page is for, how much traffic it earns, how pages link to each other, and where topics overlap or compete. One is a list. The other is a decision-making tool.
Should I delete old pages that get no traffic, or just leave them alone?
If a page has earned zero organic clicks over twelve months and isn’t serving a legal, seasonal, or navigational purpose, it’s usually costing you more than it’s worth by diluting your site’s overall authority. Merge its useful content into a stronger page with a proper redirect, or remove it outright rather than leaving it to compete quietly with better pages on your own domain.
Related reading: How to Choose an SEO Tool That Really Moves Your Rankings and What Software Options Exist for Running Facebook Ad Campaigns.
