SEOs - Stop Using Bad Data To Validate Algorithm Update Anxiety!

A new Google broad core algorithm update was confirmed for March 12th, dubbed "Florida 2." As with all Google algorithm updates, it has the SEO world in tizzy over anticipated impacts on both client and personal websites. Being a broad core algorithm update, it does not appear to be aimed at a specific niche or type of website, but instead impacting all websites globally.

Already SEOs are taking to blogs, forums, and yes, LinkedIn as well to air their grievances about traffic loss. In some cases, even celebrating what appears to be massive traffic gains. Such is the way of the world; some win, some lose, and some have no idea anything even happened.

This article isn't about the impact of the update. It's about how SEOs and analytics managers should be evaluating and reporting on in the impact (if any) this update has on their relevant web properties.

Your "Keywords Lost" Charts Don't Really Mean Anything

OMG! We've lost over 100 keywords! This is terrible! What is happening?? Google HATES our site now?!

Does that above reaction sound even a little familiar? Maybe it's a tad exaggerated, only in my experience it really isn't. I started as Search Manager for an agency last spring during the March 2018 algorithm update. My first day on the job, the guy I was replacing was having a complete panic attack.

He kept showing me SEMrush data of the sites he was managing SEO for. Some of the sites did show a completely tanking on the number of keywords they were ranking for. "We are being CRUSHED" he kept saying over...and over...and over again.

My eyes glazed over and I didn't quite understand why he was showing me this. "What does actual traffic in Google Analytics look like," I asked. He looked like a deer in headlights. "Keyword rankings are good to keep an eye on, but what matters is actual impact on website traffic. Have you checked Google Analytics yet?"

He hadn't. And he didn't seem to understand why he should. In his view, losing 100 keywords has to mean that the websites lost traffic. But no, that's not always the case. Losing keywords is a great leading indicator that something may be amiss, don't get me wrong. But you have to ask which keywords you lost. Certainly not all keywords are equal.

Losing 100 keywords that were ranking on Google's 2nd or 3rd search results page would likely have little to no impact on your traffic, since they are probably getting less than .5% of overall clicks for those search terms. Furthermore, gaining just one high search volume keyword in the top 5 results could completely offset or even increase traffic in the face of losing dozens or hundreds ranking in page 2 and beyond.

Another factor is search volume; how much potential for traffic did the keywords you lost really represent? Maybe you lost 200 keywords. But it's possible, even likely, that the bulk of those keywords were long-tailed search terms that were rarely, if ever, actually searched by a user. Which means every single one of those keywords could have been first page keywords, but if they had no search volume, they probably still represented a very small change in your overall website traffic.

Going back to my anxiety-ridden colleague who was sure that Google was beating us senseless, I was finally able to show him through Google Analytics that most of our client properties were not heavily impacted. A few did lose 5 or 10% vs. the previous week, but still showed strong YoY numbers.

Those Pretty "Estimated Traffic" Charts? Also Pretty Useless

All of this means those great, easy-to-access charts from Ahrefs and SEMrush that show your "estimated traffic" from organic search are really pretty useless when it comes to understanding algorithm changes.

Now yes, I do use them. I think they are great as a leading indicator that something might be happening. But that means that you have to take a deeper look into the real data, the actual Google Analytics, Search Console, or whatever platform you choose data that reports real numbers.

Estimated traffic data is based on a lot, a lot of assumptions. Estimated search volumes. Estimated ranking. Estimated click-through-rate. Industry benchmarks for CTR. Basically, it's based on assumptions, based on assumptions, that are based on assumptions.

The only source that can give you real information on your CTR, average position, and actual clicks for search terms is Google Search Console. I definitely recommend using this in tandem with Google Analytics to get the real data on how your website is performing in the midst of an algo-crisis.

Traffic Dropped. Okay. But What Does That REALLY Mean?

It's a worst case scenario. The estimate charts on your SEO tools were right; you lost 30% of your organic traffic overnight according to GA. You're unsure if traffic will rebound once the ranking volatility settles, and you aren't sure how to talk to your client about this.

All is lost. Quit your job. Change industries. Become a bar tender in Wisconsin to make ends meet.

Before all that, take a breath. Again, look at the data, and find out if your traffic loss is really as bad as it seems. The fact is that some websites, in fact I'd say most websites, get a lot of unqualified traffic. This is traffic that isn't really in the market for what your site is offering. They are probably high-bounce, non-converting users that make your website traffic look good, but nothing else.

Again, this goes back to which keywords did you lose rank on, leading to a decrease in website traffic. Were these high-quality keywords that drove a lot of conversions and engagement on your website? Or were these keywords that didn't really attract the kind of users you were looking for anyway?

Hopefully you have good goal tracking implemented on your website properties. If you do, then you're in a good position to figure out the real impact of traffic changes to your website. Compare overall website conversions post "catastrophic update" to the previous period or year. If you're not seeing any significant changes in conversions (whatever those might be defined as in you GA settings), then perhaps you just lost "junk traffic."You may also want to be more detailed and specifically filter to conversions from organic search, and see if there are any notable changes.

Similarly, compare other website engagement metrics, such as bounce rate and session duration, post-update and pre-update. If you see significant positive swings in these metrics, particularly in the organic traffic segment, then that again may suggest that you lost traffic that wasn't really qualified to be on your site in the first place.

Going back to my example before about an SEO manager I was replacing thinking the sky was falling on him and his client portfolio. When I dug deeper into the data, we found that even with the clients that did loose a significant amount of traffic due to the 2018 algorithm update, there was virtually no change in the amount of leads that were being generated from the site. Website metrics has actually improved pretty significantly. In short, we lost junk traffic that wasn't converting; and thats how we communicated it to the client. Since it didn't impact his actual conversions, he couldn't have cared less.

SEO Tools Have Their Purpose, But Always Validate The Real-World Impact

I promise, I'm not ranting against 3rd party SEO tools like SEMrush and Ahrefs. Their estimated charts and ranking data can be really valuable as a quick, easy way to check on overall website health. Significant changes in the data they provide can indicate something is going on that needs investigated. But take that just as a signal that it's time to do some deep diving into the real analytics on your web properties.

And also remember it's so important to report on the metrics that really matter to your clients. Traffic for the sake of traffic doesn't help anyone. Reporting on actions that matter for your client is what's most important, whether that be leads, views of important pages, or purchase conversions. Changes in traffic may not always impact those KPIs, so keep your eye on the ball!