Brighton SEO agenda was rich and full of enlightening talks. First time speakers were warmly welcomed as well as old acquaintances.
While a blog article doesn´t make justice to all talks nor to the amazing speakers this article wants to contribute to the knowledge sharing process by selecting few key practical takeaways focusing on technical SEO.
Improve performance metrics setting a process to optimize images
Myriam Jessier was a must to attend Brighton SEO as everytime she gets on the stage she catches all the audience attention and gives highly technical but accessible strategies.
Her talk focused on image performance budget: which implies defining an image budget and involving stakeholders on investing resources to ensure the images are optimized for a page. The reason for which this is a winning strategy when talking about it with stakeholders is that it is measurable since it´s based on performance metrics.
Furthermore, several tools can be used to automate the tracking of the budget and to flag and report if the budget is crossed.
The budget definition has 3 pillars:
- metrics (image weight, dimensions, total of https requests)
- milestone timing (time to first image, image load time)
- rule-based metrics (lazy loading, cumulative layout shift (CLS))
As a note on the CLS a good rule for it is that the image dimension should be specified otherwise the image moves leading to core web vital (CWV) issues.
Once the budget is defined the implementation involves the following processes:
- Define your goals to guide process
- Establish your baseline
- Set your budget: keymetrics and thresholds considering different needs for different website (the budget won’t be the same for an ecommerce than a photography site)
- Optimize your images
- Monitor performance (either with api tools or internal solution) proactively and reactively
- Iterate: revisit and adjust your budget as your site evolves
- Educate your team
So, as Myriam asked after the talk: when are you going to start implementing your image performance budget?
How knowledge graph can boost your internal linking strategy
Using knowledge graphs it´s possible to strengthen topic clusters and entities, as Silke Vanbeselaere explained.
SEOs should care about network analysis since Google does care and even has a dedicated team, the Graph mining team. However while knowledge graphs can look really fancy it is challenging to either define actions looking at them or show them to stakeholders.
Before stepping into the knowledge graph analysis for internal linking the previous step is creating a template:
- setting how you want your interlinking to look like
- define winning articles
- define what content should be connecting informational and commercial
Bridge articles allow you to connect informational articles with commercial ones: they are great as they enable you to achieve different goals: informational, provide high value content and they are great for interlinking content. Furthemore, a bridge article allows a user to move between sections of your website.
Once the template is created it’s the time for creating our knowledge graph. There are several tools to create it, the most known ones are Screaming Frog, Ahrefs, Moz, but Silke uses Gephi: an open source and free software.
Few technical terms are useful to look at any knowledge graph:
- nodes = circles, they represent one url on your domain
- edge = the line and internal link that connects 2 nodes
- modularity = for biggest nodes or URLs more connected
Different colors are then used for modularity groups to distinguish between different clusters.
Once you start looking at your graph you should ask yourself:
- Which articles could be the ones that would need more internal links?
- Looking at the colors, are there clusters topics that are less interlinked?
- Are the bridge articles fully utilized in your internal link strategy?
What is most appealing about this talk is that it opens a Pandora box of different ways on how to use an open graph exporting any other data from a csv file and merge data together to visualize it and set data. As Silke herself described this same strategy can be applied to competitors’ backlinks.
The opportunities beyond Log Files
Log file analysis can supercharge your SEO thanks to its multiple uses broken down by Sally Raymer and listed as:
- How Quickly Content Is Crawled And Indexed
- How Frequently Priority Pages Are Crawled
- Identify Opportunities To Expand Content On Your Most Crawled Pages?
- Testing
- Server Side Rendering (Ssr)
- Updating Schema
- Pages Visited By End User Vs Crawled Ones: Are They Aligned With Your Top Pages?
- Know If 4xx Pages And 3xx Still Get Crawled By User Agentes (Ua)
- Identify Why There Are 5xx Errors
- Identify Spam Generated Urls
As further practical tips Alex Wrights offers solutions to common issues faced during log file analysis:
- they are massive so as a solution they should be split into chunks
- since they include ip addresses there might be security issues for which it might be more difficult to get them, as a solution it´s possible to ask for logs without ip address
Alex digs deeper in what unusuals behaviors you might find with your analysis and dig deeper into it:
- UA: depending on which UA they might get different response code
- 304: this code means a user agent finds the same content
- discrepancies between crawl and log: URLs in log file but not in url data (can also depend by your UA)
- canonical vs. non canonical: have non canonical been crawled more? Is the non canonical considered more important than the canonical? If so, could it be depending on the parent page they are attached to?
- elements causing infinite urls
In conclusion if you get down and dirty with log files you could spot both technical issues as well as content worth to be expanded and updated.
Brighton SEO September 2023 key takeaways
There were many other very interesting talks and ideas and the technical SEO talks covered far more topics than the ones reported above: however the few selected here have been selected for the practical opportunities that each of them entails.
