So just to quickly recap, what I found was sites that had lost in the September 2023 and March 2024 Helpful Content Updates, the last one was merged with a core update, although I do think they’re quite distinct signals, even though they’re allegedly merged now. But anyway, the losers had pretty similar Domain Authority to sites that had been positively or not affected. But their Brand Authority was significantly lower on average, and as a result, their ratio of Domain Authority to Brand Authority was also very different.
Now interestingly, you’ll note that the ratio of averages, the average Domain Authority and the average Brand Authority are not a factor of two separate, but the average of the ratios is a factor of two. As far as I know, this is a mathematical inevitability. But if someone wants to post a proof or a disproof somewhere, I’ll give you an easy retweet. But that would be interesting to see. Anyway, I’m getting off-topic.
So what are we actually looking at here? So, for example, if I have a site where maybe my Domain Authority is pretty high, perhaps it’s 80, but my Brand Authority is pretty low, perhaps it’s 20, then I end up with a ratio of 4. A ratio of four suggests that I’m pretty likely, by no means certain, but pretty likely to be negatively affected by a Helpful Content Update, according to this research.
Now, obviously, neither Domain Authority nor Brand Authority are Google metrics. These are proprietary Moz metrics, but they are based on signals that Google has access to just as much, if not far more than we do. They’re based on data that we get from SERPs, from crawling the web to get a link index, from search volume modeling. Google has access to all of these same kinds of signals. So it’s not totally impossible that they could be doing something similar, and indeed we have seen them talk about it before, which I’ll come back to in a moment.
If you liked Google HCU: What Can You Do? — Whiteboard Friday by Tom Capper Then you'll love Miami SEO Expert
Check how much traffic comes from commercial and transactional keywords, then look for gaps your…
1. Consider specifying locations or personasSo one of the big ones that you could play…
1. SemanticSo number 1, semantic. This is obviously one we're very familiar with from conventional…
Avoid prompts that create false positivesSeparate prompts into three types: branded, comparison, and non-branded.Branded prompts…
So even brands that have made their name synonymous with a product will not be…
The model breaks visibility into four quadrants:Open areas known to your brand and customersHidden areas…