How to make a content audit work for you

I’ve posted previously about how I think content audits should die.

But I’m going to expand that thinking here and give suggestions on how you can use data to improve content.

What are you auditing and why?

I understand the need for an organisation to want to know what it has now. However, a traditional content audit seems to stop at giving the organisation a snapshot of where it is now and a list of pages to work on. Traditional content audits don’t seem to tell you enough to make judgements on where the content should be or what the site should be doing.

By simply auditing what your site offers now, I think a number of people are missing a huge opportunity to turn their site from an organisation-focused site to a useful, accessible, user-centred site.

Content audit blog posts

I am seeing more and more posts supporting content audits – and they are saying the same thing. Which is perhaps great, the content community has a consensus!

But from the ‘pro-audit’ posts I’ve read this week:

  • 50% don’t mention any sort of data or analytics.
  • 80% don’t mention the user.
  • 100% don’t mention in-page search terms or user feedback mechanisms.

I find this a little odd. If you don’t know this information – how are you going to improve your content and who are you improving it for?

From what I have read, a lot of content audits follow this pattern:

  • get a sitemap of your site,
  • build a spreadsheet,
  • work out what you need/don’t need (some suggest only basing this on number of visits),
  • publish.

I find two of those points challenging.

The dreaded audit spreadsheet

I hate spreadsheets with a passion that’s almost holy. It’s not the software itself, of course it can be very useful and you may need it eventually – it’s the mindset some people go into when they have one open that disturbs me. I’ve seen people stare at a spreadsheet and stop thinking.

You may not experience this so let me explain: some people look at what they have and stay there. They improve what they have – which is wonderful. But if you are only looking at what you have, you may miss gaps or not be able to tie in a certain piece of content with certain user comments etc. It can be a very narrow view.

The blog posts I have read (and obviously it’s a small selection of posts out there) talk about the only metric being the number of visits. I don’t think that is enough.

As Louise Stone, product content lead for GOV.UK said: “Just because people visit the page, it doesn’t mean they get what they want”. I agree wholeheartedly. A visit number is the start of the journey, not an end point, to improving content.

Work out what you do and don't need

This is the one I struggle with the most. Some people appear to be advising to take the lowest performing pages and throw them away. Or see if the page still fits your organisation’s needs. Both of those ways are ok – but only if taken in context.

You may have a low-performing page because the title is wrong and no-one can find it, not that the page is no good. You should find out what your organisation wants to convey, but put it in a user-centered way or it’s vanity publishing.

I fully support prioritising improvements – particularly if you are short on time or people – based on your top-performing pages or search terms. But if possible, I think you need to go further.

What I would want to see in a content audit

As a content strategist, I want a place where I can see how to continually improve my content. So if I started work tomorrow on an existing site, this is what I would want:

  • page title,
  • what user need it was fulfilling,
  • unique visits (time span to be specified by me – and would be alterable),
  • time on page (criteria already set for success),
  • how many user comments have been sent in about that page – negative and positive (with a way of seeing those comments),
  • in-page search terms (highest number first),
  • search terms related to that page/topic,
  • related content on the site,
  • traffic trends in the year (if applicable).

I’d also want the following, but they are not as important so I’d want them at the bottom of the screen:

  • URL – so I can get to that page easily,
  • content type - this is more for reporting purposes.

If I had different content types, that is. For example, it would be good to see if all my long-form copy pages had high instances of ‘it’s too long’ in the user comments section.

Of course, you can add a stack more based on your organisation, like if there’s video on the page (and that it has a transcript. Please, internet, don’t supply a video without a transcript), or anything else you might want to report on. But to improve content, I want user-thoughts first, everything else later.

No more audit spreadsheets

I wouldn’t have this in a spreadsheet, if at all possible. I think the perfect place to have it is in the publishing system itself but I know how hard this can be. So perhaps just a database.

I think you can interrogate a database more effectively and you can cut out all the surrounding ‘noise’. You’d have one record, one thing to focus on when you were working. That’s a lot less distracting.

Audits are for life, not just for Christmas

Audits seem to be something some people do once a year. I’d want this information all the time.

Of course, it depends on how much you can automate this information or how much time you had. But in any place I work, I’d want to make this part of my day job.

For example: get in, make a cup of tea, look at the top performing, lowest performing pages, trends for this time, search terms from the past 24 hours etc. Then I could see what my users wanted now. Am I doing what they need now? Do I need to prioritise anything or be looking at trends? No? Fine. I’ll get on with the to-do list. But I’d like to see – every morning – what my users need.

I don’t think I would call it an audit. It would simply be a part of my job in creating and maintaining a user-centred site.

(Huge thanks to Chris McCarthy who helped structure my thoughts on this one.)

Sign up to our newsletter

Get content design insights sent straight to your inbox.




  • Choose what information you get: (required)