The September 2023 Helpful Content Update has just finished rolling out.
Now that the dust has settled and SEO’s (particularly, niche content website owners) have finished collectively losing their sh*t, I had a question…
Was this update even about “helpful content?” The answer was…kinda, but also kinda not.
I decided to analyse 20 previous clients I worked with over the years for my own agency and while I was Head of SEO at a previous marketing agency.
Their results surprised me, but maybe they shouldn’t have.
How I Conducted this Helpful Content Analysis and Study
The first thing to say is that I’m not out here in an SEO lab isolating every single variable I can to show you the causes of why some websites got hit during this Helpful Content update and others didn’t.
In SEO, it’s very difficult to prove causation. You can look at some of my findings as correlations rather than causation.
There will be more thorough analysis of this update than mine online, with a bigger data set.
I’m working with a very small data set here and I’m sure plenty can and will pick holes in my very loose methodology here.
Heck, I’ll even do that myself at the end of this article so stay until the end to see the limitations of this analysis…but for now, let’s get into the meat and potatoes and see what I noticed.
Things I Looked at in The Analysis
I looked for the inclusion of a few different things on each site (and off site), looked at their domain authority and then looked at how their traffic and keywords were affected by the end of the HCU versus before the update.
Below were my documented findings:
You’ll see that I looked for the following things on and off their websites:
- GMB/KNOWLEDGE PANEL ON GOOGLE SERPS?
- ADDRESS DISPLAYED ON WEBSITE?
- PHONE NUMBER ON WEBSITE?
- EMAIL ADDRESS ON WEBSITE?
- SOCIAL MEDIA ACCOUNTS ACTIVE?
- PPC (GOOGLE ADS)?
- SERVICES OR PRODUCTS OFFERED?
- BLOG ON WEBSITE?
- DOMAIN AUTHORITY
Then I looked at the following changes before and after the helpful content update:
- ESTIMATED TRAFFIC (before)
- ESTIMATED TRAFFIC (after)
- Traffic % Change
- TOP 10 KEYWORDS (before)
- TOP 10 KEYWORDS (after)
- Top 10 KW % Change
The Results of the Helpful Content Update for These 20 Websites
From my analysis, I concluded that NONE of these sites got “hit” by the Helpful Content Update, even if they did decrease in traffic and keywords.
None of the losses seemed significant enough for me to claim they were “hit” (more detail on that a bit later)
Some quick stats that jumped out at me from my HCU analysis of these 20 websites:
- 70% of the sites increased in traffic after the HCU
- Only 15% had fewer top 10 keywords after the HCU
- Only 10% had BOTH fewer Top 10 keywords AND less traffic
- One of the two sites that experienced BOTH fewer top 10 keywords and traffic only had 3/6 “REAL” signals (REAL isn’t anything official. They’re just factors I associate with a real business. You can call it EEAT signals, but that feels like it has become an esoteric term thrown around by SEO’s without fully understanding what it means)
- Every other site had at least 4/6 “REAL” signals
- 74% of the sites had at least 5/6 “REAL” signals
- 100% of the sites that saw an increase had BOTH Google My Business and phone numbers displayed on the site
- 60% of the sites that experienced a decrease didn’t utilise Google Ads
- 77% of sites that saw an increase DID utilise Google Ads
- 95% of sites we looked at offered at least one of either products or a service..the remaining site still offered something of value – a free digital magazine subscription
- The 2 sites that gained the most absolute traffic (around 50k visitors a month each) have a TON of horrendous blogs (we’re talking 100’s. More bad, than good)
In fact, both of these sites were clients I managed for years at my previous agency and you can read the case studies yourself: One was an e-commerce furniture store SEO case study and the other was a tailor-made travel holidays SEO case study.
- There was no correlation between wins, losses and Domain Authority (second biggest traffic loser had a DA of 41, the highest DA in the analysis, and the biggest winner had a DA of only 13)
- Many sites had thin, unhelpful or badly optimised blogs. The clients who we did produce content for during their SEO campaigns, remained strong and almost all of these individual pages got a boost.
The thing that will immediately stick out to you about the clients that I’ve worked with is that they’re all “real” businesses and brands.
Yeah, yeah..I can hear niche and content site owners who got hit by the HCU saying “but my site IS a real business”. But try ticking off how many of the above factors you have included on your site.
I’m willing to bet 80%+ of these sites don’t have the majority of these elements included.
My Conclusion From This Helpful Content Update Analysis
While not one of these stats can be used to say “THIS is why my site gained or lost traffic”, I believe it shows that Google has in the past, really been struggling with identifying real businesses and real experts.
It seems that sites, even with fairly poor content in their blogs, fared really well in this update because they were “helpful” in general and provided something of value to their users. Whether a product or a service.
One thing I think I can confidently say is that having these “REAL” signals creates some kind of defensive “barrier” around your website from these types of updates, even if you have poor content.
What About the Sites in Your List that Experienced a Decline?
Sure, some sites on my list experienced a decline. 6 out of 20, to be precise. But only 2 of them experienced a traffic loss AND Top 10 KW losses.
One of these sites experienced a 4.3% decline in traffic (an absolute loss of 37 monthly visitors) and an 8.4% loss in top 10 keywords (an absolute loss of 4 keywords).
I don’t know what the threshold is for a site to be considered “HIT” by this update, but I’m confident that isn’t it.
The second one lost 23% of traffic (absolute loss of 2169 visitors monthly) and a top 10 keyword loss of 2.6% (an absolute loss of 6 keywords).
Again, arguably the traffic loss is a fair chunk, but the top 10 keywords remain similar, which doesn’t convince me it was hit at all, especially when I went through the sites keywords in more detail.
They had a pretty even amount of keywords declining and increasing. I also didn’t notice any patterns with specific keyword clusters declining.
Isn’t This Analysis Irrelevant For Content or Niche Websites?
Maybe, but there are some important things you can learn from it.
Is your brand or “entity” really something that Google knows much about and trusts?
What all of these businesses are doing is entity stacking and letting Google know that they’re a real brand.
They’ve got their social media accounts, they’ve got a Google My Business, they’ve got their NAP on their website and I assume, plenty of citations off-site in business directories and other places online too.
If your site is just sitting there churning out content, simply leveraging SEO traffic, without any of these other signals, is Google really to believe that your site is there to offer value and not just existing to try and make easy revenue?
Are There Other Things That Might Have Affected My Site in The HCU?
Yes, it’s likely there are plenty of other factors that were looked at in the Helpful Content Update.
Here are some of the other things that might have played a part:
Informational topics and searches seem to now favour UGC Sites
Niche sites are often heavy in informational content, so it’s no surprise that so many got hit by this update.
Especially when you look at the big winners like Quora, Reddit and LinkedIn Pulse…Google seems to think that users will find experts on these forums and would prefer to hear from the people who frequent them.
Time will tell whether that’s a permanent fixture.
My guess is that Google will look at user signals like CTR and their behaviour/engagement with those sites and adjust again (or not) from there.
Too Much Ad Code Above the Fold
Charles Floate just dropped a really thorough video analysing some other factors and one which really intrigued me was the amount of ad code a website has.
He found a correlation between sites that had a lot of ad code and scripts above the fold and sites that got hit by the HCU.
Yeah, I know…do as Google says, not as Google does, right? (we see you Google, with your 4 sponsored ads in the SERPS above the fold and all of your SERP features).
This finding by Charles actually leans in nicely to my analysis.
NONE of the businesses I looked at had ad scripts above the fold, as you would expect.
They don’t want revenue from ad clicks taking them away from buying their product or service.
Homepage Being a Blogroll of Your Latest Posts
95% of the websites in my analysis were well structured and linked out to the various categories of products or services that the business offered.
Most real businesses tend to do this well without even thinking about it.
Only one was a list of their most recent posts, which was a magazine and that makes total sense for that specific business, even though I’d encourage them to optimize their homepage structure too.
Have a look at your homepage, does it make it easy for both Google to crawl and easy for users to navigate your site and get to where they need to go?
Site Structure, Internal Linking and Semantic Relevance
This leads nicely from the last point. Something Koray Tugberk always talks about is reducing the “cost” for Google when crawling your website.
I won’t go into the complexities of Google’s bots here, but the first thing to do is to look at your site structure from your homepage, to your categories and inner pages.
Is every page on your website accessible from the homepage within 3 clicks? They probably should be. At least the most important ones.
Are you internally linking each page of your website, in a way that makes sense to send Google clues about which topics are closely related to others on your website, using descriptive anchor text?
If not, definitely make a start there.
So, Was This Really About “Helpful Content”?
Well, yes and no.
From my analysis of these sites and the complaints I’ve seen on Twitter and Facebook, a big part of this update was about whether you’re a real brand with real brand or business signals.
The part that DID seem to be about content was Google pivoting on what they believe their users perceive to be “helpful content”, which is why you saw Reddit, Quora and other UGC sites win in this update.
It seemed to me Google weren’t looking at whether a piece of content was helpful specifically, but whether there was enough evidence that the source of the content was helpful.
Quick Solutions and Conclusion For Recovery From the Helpful Content Update
To wrap this up, let’s look at some solutions from my analysis and others that I feel would work to recover your site.
- Add as many “REAL” signals around your business or brand as possible. For people who own a portfolio of pure content sites who think they can’t do this: In the UK, you can trade under multiple names for one company. There’s no reason you can’t use your registered company address for multiple different brands and apply for a GMB profile for each. You can also sign up for virtual phone numbers at really low cost, or sometimes even free, for example: Google Voice or textfree.us.
- It’s more work, but offer some more value beyond “content”. Even if it’s a free automated newsletter with a free lead magnet like an ebook.
- If your homepage just consists of your latest posts, think about making it a well-structured hub for all of your best content and category pages. Upgrade your categories too beyond just an archive of posts. If you need examples, Joost Boer has some cool homepage example templates that he sells:
Homepage 1 | Homepage 2 | Homepage 3 | Homepage 4 - Look at diversifying your traffic. Real businesses don’t just rely on organic traffic. They have a social media presence, may use ads and gain traffic from brand searches, whether direct or through search.
- Keep doing some of the things you should be doing anyway – like matching your content with the intent of the results on Google’s first page
- Finally, take off your rose-tinted glasses around your content and website. Google doesn’t care, nor does it rank content on “passion” or “emotion”. I know it’s difficult if the site in question is your full time income, but go through your site and content with a fine-tooth comb. Does it serve the correct search intent, firstly? Does it contain the entities and semantic topics Google expects to see? (you can use SurferSEO or NW to check) Is the content truly better than your competition? Do they present data in multiple ways like video, tables, infographics, reference high quality external sources etc?
Caveats and Analysis Limitations
- The traffic and keyword data is pulled from third party SEO software. As we all know, these can often over or underestimate traffic and don’t offer the full spectrum of long-tail keywords. They do provide a general direction of travel, though.
- It’s 20 websites. That’s not a huge data set.
- The factors I looked at were limited. I could have looked at more like technical health, internal link structure, homepages, about/author pages, backlinks etc.
- I looked at traffic and KW data from the end of August to end of September. There could have been natural rank and traffic movement in between the month before the HCU hit.
- Following on from that, I don’t know whether some of the clients I don’t work with anymore were adding content, building links etc, which also could have made a difference.
As I said, the impact can be analysed far deeper than I did but the goal was to look for patterns and I believe some patterns certainly did reveal themselves.
Overall, I’d be hesitant to say this update was ALL about helpful content and if you look at it through that isolated lens, that’s when emotions about any losses your site incurred can take over.
Zoom out and look at your site from a real business or brand’s perspective. Is your website behaving like a real brand or business would?
If not, I’d fix that before even thinking about how your content could be more “helpful”.
In cases where your informational content has been taken over by your Reddit’s, your Quora’s, LinkedIn’s, I’d actually argue trying to optimize your content is a complete waste of time at this point.
You’re just going to have to let Google ride their update out and see whether their decision to boost these sites results in the user behaviour (and revenue) improvements that they are hoping for.
