Five lessons learnt from technical SEO audits

Undertaking thorough technical SEO audits is essential and incredibly important to any business reliant on SEO as a revenue stream.

These audits can be a daunting prospect, particularly with older domains and sites with a lot of pages. The aim of this post is to share some personal lessons from our past technical SEO audits to help get you get on the right track.

1. Start with the logs

Start with an in-depth log file analysis, which will give you a solid bird’s eye view of how Google views your site. You’ll be able to identify exactly which pages Google is and isn’t crawling, and how often. 

You’ll almost always spot some interesting issues straight away. I won’t go into detail about what server logs are and how to analyse them here; David Sottimano and Tim Resnik have written fantastic guides over on the Moz blog.

2. Get pivoting!

Being able to manipulate Excel data using a simple pivot table is a great skill for an SEO auditor to have in their locker. It really comes in handy when you slice and dice the data from those server logs, trust me! 

Annie Cushing has compiled an awesome resource for managing big data with pivot tables over at Search Engine Land that will get you up and running.


3. Rel=catastrophe

A recent audit for one of our clients revealed we were implementing the incorrect rel=canonical tags on a large number of PPC pages. 

They had created lots of specific landing pages purely for paid traffic, but were borrowing content from their indexable organic landing pages to use on them. However, instead of referening these 'parent' organic pages, the paid traffic pages were using canonical self-referencing. This completely negated the tags, instead sending out dangerous ‘duplicate content’ signals to Google.

The takeaway from this is that you can easily shoot yourself in the foot if you aren’t using tags such as rel=canonical in the correct manner. These are designed to handle issues such as duplicate content in a smarter way, so do be sure you and your colleagues are implementing them correctly!

4. Noindexing doesn't solve everything

When dealing with a site that is suffering under Google Panda, it can be tempting to ‘noindex’ all low-quality and duplicate content. While this can be effective short-term, it can be dangerous long-term.

If you simply noindex pages, users may still have access to this content. But worse still, Google may be able to crawl and ‘see’ your noindexed pages. Having a high ratio of noindexed versus indexed pages can still send negative signals to Google. I once worked on a site that had 85 per cent of its 7,000 active pages noindexed!

So while noindexing pages to get them removed from the index might be a smart initial step to take, I would always recommend beginning the process of cleaning those pages up as quickly as possible: we’d suggest you either 404 redundant pages, or re-direct similarly themed pages to the closest live pages.


5. Document, check and test

This lesson is probably the most important I’ve learnt. Make sure everything you do is documented, checked and tested. Some top tips for this:

• Prioritise tasks into high, medium and low priority lists and target the high priority issues first.

• Make sure you've noted exactly what you've done, the date you did it and included any further action that might be required, such as removing a temporary XML sitemap a week or two down the line.

• Check all the changes you've made. Implemented a bunch of new re-directs? Run the URLs through Screaming Frog or check them manually. Updated your robots.txt? Download some log data and check Google is crawling what you want it to.

Hopefully this will help you get your site management back on track. If you’ve learnt any other important lessons from SEO audits, please share them in the comments below.