,
8 tweets,
2 min read
Read on Twitter
A few days behind this #SEO study written by @timsoulo: it's fairly short, so let's dive in!
ahrefs.com/blog/links-wit…
1/8
ahrefs.com/blog/links-wit…
1/8
Good usage of highlighting the potential pitfalls of their data and noting why its important. Honestly appreciate studies that start off this way.
2/8
2/8
We see the common use of Spearman Correlation which is disappointing given the topic of analyzing very similar metrics, but what I like is that the correlation talk is relative to the other factors. It's not about what has a H/M/L correlation, but which is higher among them.
3/8
3/8
Note: I'm not being pedantic about this btw, I think it is important to compare among the factors (or over time), even if none of the factors correlate even weakly given the low (<0.3) findings.
4/8
4/8
It does a good job highlighting an odd finding and provides what might be reasons for why that could be the case including the potential for their analysis not being complete enough. Given how often folks jump to conclusions, this is generally a good sign in an analysis.
5/8
5/8
There's no scatterplots sadly & while I'm generally not a fan of pie charts, a one-time look at the data is one of those acceptable uses. That said, would have been nice to touch on whether the result set meant you can rank without links OR if the data set was biased.
6/8
6/8
On the whole, it's a relatively easy read, doesn't try to claim anything grand, and looks within their own data to compare the metrics they have for interesting tidbits.
7/8
7/8
Lastly, I still do want/plan to do this () because the industry unfortunately is souring on analyzing factors imho because of studies that are too simplistic.
8/8
8/8
