How to Perform a Technical SEO Site Audit

Learn how to perform a Technical SEO Site Audit without having to spend a fortune. Access to a Trello Board, copy it to your own personal Trello for free.

Learn how you can hack your way to the top with proven, data-backed tactics directly to your email every week.

I will through this out there. If you haven’t planned for a technical SEO site audit then you need to reconsider your entire strategy. There are quite a few things a forgotten noindex can’t destroy and you will be amazed by the number of clients that come through for a free technical SEO site audit.

Having said that, not all websites need a professional site audit. I always try to tell the same thing to all the leads that come through the site or that contact me through Linkedin. If your site, or the site that you are working on, is not a multinational eCommerce site the chances are that you can handle a lot of the work yourself and in the process save some money (or if you are an SEO Specialist make some).

The following steps are the ones that I take when I perform SEO audits. It’s by all means up for grabs but also you can take all of the things into consideration and then follow your own. If you prefer to just get the checklist and get it over with then here you go.

Trello Board: https://trello.com/b/4UWWpecB
Still under construction and will keep updating this regularly

Step 1 – Use your senses, Luke

Luke Skywalker training

I can’t stretch enough the need to get familiar with the website you want to perform the audit. No matter the industry or the niche, no matter the service or product, little to nothing will make a difference if you haven’t seen it with your own eyes.

Visit the website and scroll through the pages as if you are a user. Some of the issues will be immediately visible while others will require you to go through the entire process of adding some items in your card, trying to “buy” something.

Don’t just visit the site for a minute or two. Spend the entire day if you have to. Visit the site from multiple devices, multiple screen sizes, and different browsers. Try a different combination of paths. Try clicking on pages that you wouldn’t usually do so. Take some pen and paper notes.

No amount of tool usage will ever tell you the finding that this process will.

Step 2 – Keys, Wallet, Phone

Much like checking your pockets before you leave your house for the essentials, keys, wallet, and phone, I check for the basic implementations on a website.

You don’t have to go in too deep in this stage but familiarizing with the website means;

  • Where was the site built on?
  • Does it use a CDN?
  • Does the site have a Google Analytics set up?
  • Was the site verified with Google Search Console?
  • Does the site seem to have a canonical on every page?
  • Is there a robots.txt?
  • Is there a noindex present at the <head> section?

Again, don’t go too deep, all of those things will be more easily revealed in the next step but it’s also important to notice them sooner than later.

Step 3 – Screaming Frog is your best friend

There are plenty of crawlers in the SEO sea I am sure. Are any of them better than Screaming Frog (or cheaper for that matter)? From my personal experience, there isn’t one out there that can compete with the functionality of SF.

I won’t go deep on how to use Screaming Frog in this guide but I will write it down to my content ideas pad for a full explainer down the line!

Go ahead and punch your URL (or the URL that you are interested in crawling) in Screaming Frog and let it work its magic. Depending on the site you’ve asked it to crawl this might take seconds, minutes or hours.

Step 4 – Information Gathering

Step four could very well be done at the same time step three is taking place. As I mentioned, crawling a website can differ depending on the magnitude of it. In the possibility of taking hours, sitting there and watching the URLs populate (even though quite impressive) I would suggest that you move on to step four. The gathering of the fellowsh… information.

Tune your browsers to Google Search Console or even Bing Webmaster Tools and Yandex Webmaster Tools if you really care.

There are hardly enough manual settings that someone can mess up in those platforms but nonetheless you can break things from those dashboards too.

Questions you might want to ask when you are checking Search Console;

  • Are sitemaps listed?
  • Are there any issues or warnings listed in the Coverage section?
  • Are there any pages listed under excluded? (not always a bad thing)
  • Any pages in the Removal section which are obvious that shouldn’t?
  • What kind of Enhancements does Google list in the Search Console?
  • Is the security and manual action section clean of warnings?

Do the equivalent of the above for every webmaster tools. But make sure that since you are in there export some of those issues you find in a spreadsheet. Not all engines treat the same things as issues or warnings and not all search engines find all of them; export the relevant data, put them in a sheet and compare.

Step 5 – The logs don’t lie

Logs next to a snowy forest
Photo by Radek Grzybowski on Unsplash

Depending on the website you are working on this is something that you really need to consider doing. Google Analytics, Adobe Analytics, etc. I usually take what the say with a grain of salt. Very fine grain of salt… the finest. Well, put simply I don’t believe sh*t of what they say when it comes to large websites with multiple channels and various attributions.

I am not saying that they on-purpose false report but they are really mixed up.

If you can get your hands on the server logs then tools such as Sumo Logic can help you find every possible entry to your site. Keep in mind that if you are using CDNs such as Cloudflare, your logs will not be 100% true either since CloudFlare’s configuration (depending on how it was set up) will be keeping some bots or humans from even reaching your server.

Here are some questions that you might want to ask once you make your way through your log files if you are comfortable doing so.

  • Do engines encounter any kind of problem when they try to crawl the site?
  • Do users encounter any kind of problem?
  • Which pages get crawled and why some don’t?
  • Which pages take longer to load and why?
  • Are there any redirections that you might be able to solve easily?

Since tools such as Sumo Logic are not SEO tools, playing around with them in order to find the answers to the SEO questions above might take some time or even be a huge task for some. Tools such as Screaming Frog can help again since they offer a more SEO friendly Log Analyser. In fact, I have prepared a guide for SEO log file analysis that you might find useful for this job.

Step 6 – Evaluate

You knew this time will come. You have everything exported into spreadsheets and really depending on the site you have a whole lot of data to process. If you followed everything we discussed above then you have information about the following.

From Screaming Frog

Technical SEO Site Audit - Screaming Frog
  • Protocols
  • Response Codes
  • URL details
  • Page Titles
  • Meta Descriptions
  • Headings
  • Image Details
  • Canonicals
  • Paginations
  • Directives
  • Hreflang

From Google Search Console

Google Search Console Coverage Dashboard
  • Coverage Errors
  • Coverage Warnings
  • Excluded URLs
  • Security & Manual Actions

From analyzing the logs

sumo logic log files
  • Pages with no activity
  • Large and very slow pages
  • Orphan pages
  • Response Codes

Are you a Google Sheets or Excel wiz? Figure out how to make them pretty and move to the seventh step of hel… of the SEO site audit.

Step 7 – The Exams

It’s not just making them look pretty in a spreadsheet. You need to figure out what makes them fail and what makes them pass(able). All of the above have a right(ish) and wrong(ish) answer and they more you lean to the right the better it is.

Obviously the right answer changes slightly depending on the industry but you need to make sure that you apply your knowledge of “Getting Familiar with the industry” we talked at the first step. Nonetheless, I will offer my advice when I go through the above list.

Protocols

The protocol in this section refer to the Hypertext Transfer Protocol or else known as HTTP and the secure version of it – Hypertext transfer protocol secure else known as HTTPS.

Google has made numerous statements that the security of their products is an important notion for them. Aside from their products they consider the security of the websites they serve to their users through Google search of the utmost importance. Because of that, they made sure that HTTPS is “enforced” by making it a ranking factor in their algorithm.

There are free and paid certificates that you can acquire for your site. My opinion of the best of the free choices is Let’s Encrypt

Response Codes

Response codes refer to the HTTP status codes such as 3xx, 4xx, 5xx etc. You need to make sure that you have few to none 3xx aka redirections, few to none 4xx aka broken pages, and absolutely no 5xx aka server issues.

The reason why I am not saying you should have absolutely no 4xx pages like most of the articles out there is that there is no way that you will be able to succeed. And honestly, you shouldn’t really bother. Unless you cheat and I don’t want you to cheat. Make sure that the 404 pages that you have are manageable, fix them if possible, but the reality is that 4xx pages are a natural order of the world wide web.

Web pages come and go often and you should mark them properly.

5xx errors are server related errors. These types of issues can prohibit search engines from crawling and indexing your pages but also users from being able to enter your site entirely. They need to be dealt with as fast as possible.

URL Details

We don’t live in the caves anymore and thankfully CMSs are quite capable of beautifying URLs automagically. If you see URLs with underscores and special characters then you need to start fixing them as soon as possible.

Uppercases, underscores or URLs that look like the wall of china in length are really terrible for user experience and engines.

Titles

Titles are one of the most important on-page ranking factors you can easily optimize. It is also a required tag for all HTML documents. This is not the time to speak about how to optimize them but make sure that they exist in all HTML pages of the site you are auditing.

Keep an eye out for titles that are over 545 pixels (60 characters) or below 200 pixels (30 characters).

Interesting articles
https://web.dev/document-title/
https://support.google.com/webmasters/answer/35624

Meta Descriptions

Even though not a ranking factor, meta descriptions are considered to be a very important on-page tag. A meta description tag should inform the user of what a particular page is all about. Look at it as a one-liner pitch that you would use otherwise in order to convince the user that your page is what they are looking for.

There is no limit on the number of characters that you could use but engines truncate anything above a specific point in order to fit within the device width.

A few rules of thumb when you are checking the meta descriptions as followed.

  • Every page should have a meta description
    • There’s not a lot we can include here. Every page that you want indexed needs to have a meta description for engines to provide to the users.
  • Each meta descriptions should be unique
    • Unique meta descriptions are necessary for the user to make an informative decision when they are researching. If you can’t provide a unique description then Google might provide one for you and replace your meta description but this extremely lowers the user experience and minimizes the chances that someone will click on your results.
  • Include clearly tagged facts in the description
    • Contrary to what’s been said, meta descriptions do not need to be in a sentence format, and in some cases, there is more information that you could add using tag format. For example, even though a bookstore could use a book description, they could use the tag approach as well by providing a description that looks like this <meta name="Description" content="Written by A.N. Author, Illustrated by V. Gogh, Price: $17.99, Length: 784 pages">
  • Programmatically generate descriptions
    • Generating meta descriptions for small sites such as fourth-p.com is easy and should be done most of the cases manually. Generating descriptions, no matter the size, for larger websites can sometimes be impossible. In such a case it is encouraged to generate them using a formula. Make sure that the formula takes aspects of the page the meta description is added so that your descriptions are still unique and informative.
  • Use Quality Descriptions
    • Make sure your descriptions have all the above and then some. Make sure they are high-quality, unique, and descriptive and not a ploy to somehow grab the attention of people without providing what you say you do. Don’t overdo it with keywords. It will get you nowhere.

Headings

Headings hold less and less value every passing year, In fact, I don’t entirely believe that headings hold any kind of ranking factor anymore. What they are useful though is natural language and Google understanding the context of the page they are looking a lot better than they would if you didn’t have user-friendly titles.

Make sure there are headings in every page. At least an H1 and especially when there are photos, videos or products instead of content.

Don’t make them duplicate, don’t make them a full-on the description. A heading is a heading and is meant to help people understand what they are about to read or find below it.

Alternative Text – Image Details

Alt text is used within the HTML to describe an image using simple words.

You add an image, you add a description. If you find images with no alt texts then you make sure you write them.

One of the main functions of the alt texts is to provide a fall back for when the image doesn’t load properly and is a principle of web accessibility. Screen readers use the alt attributes of an image in order to provide a description to visually impaired users.

Are there images that weight more than they should be? I can’t find myself agreeing with the 100kb limit that Screaming Frog has but nonetheless use it. Check all the images that SF will return as more than 100kb and then make sure that you do everything that you can to optimize them.

Obviously don’t write a paragraph for the alt text. Keep it short and precise.

Pagination

Paginations are a pretty interesting subject. Especially if the site you are auditing is using an infinite scroll or a load more + lazy load implementation. Make sure there are links with proper anchors tags to the next and previous page.

Make sure there are no broken sequences. No broken links pointing to an imaginary page that you don’t have.

Directives

Directives are probably the first thing I check (not sure why I added it so low in the list). You will be seriously impressed on how many times someone forgot to take the noindex tag off when they finished developing a site or someone added the noindex to the whole bunch of pages by mistake.

Depending on the site you might want to check this guide on how to actually prevent Google from indexing pages that you consider the low-value pages in order to increase the indexation of the pages you do.

Canonicals

Canonicals are important stuff people. It can help you avoid a whole lot of misunderstanding in a pretty simple and straightforward way. Just make sure that all pages contain a canonical, to begin with, and then start checking the self-referencing canonical tags and the ones that point to another page.

Robots.txt

Simple put, robots.txt is a text file that gives directions to search engines regarding the pages that they are allowed and not allowed to crawl. You can understand the importance of checking the disallow rules and making sure that you are not blocking pages that are suppose to be crawled and the other way around.

Conclusion (not really)

The above are just a few things that you need to keep an eye out for a preliminary site audit. The difficulty increases (or decreases) depending the site you are working on but never think that because a site is small it can’t have enough issues to fix.

I would spend a good amount of time working on my technical audit. A good foundation is required for organic rankings. You don’t have that, you better be prepared to burn a lot of money without results.

Stay informed

Join hundreds of entrepreneurs, marketers and SEO specialists receiving a weekly data-backed, proven SEO tactic straight to their email.