Has there been a Google algorithm update? September 2016
There’s recently been talk of further Google updates. Read our Google algorithm update here…
Penguin 4.0 became part of the core algorithm on 23/09/16 – read about the effect Penguin 4.0 has had on gambling and finance industries, as well as internal link networks.
There’s yet more mass hysteria as Google results experience unnamed and unexplained flux; all following the standard formula:
- Temperature gauges (like Moz weather, Accuranker, Algoroo etc.) display a spike in activity
- Search Engine Roundtable proclaims there’s a new update
- People deep-dive into their data, and spot some anomalous drops
- Some assertive SEO’s tweet Gary from Google asking if it’s Penguin
- Gary says no
- SEOs continue to shake their heads in exasperation, and hypothesise on what it could possibly be
Since the last ‘September 2nd update’ we at Pi have collected more data and seen some definite drops, and indeed peaks, across ecommerce sites:
September 2nd Google update 2016: More evidence
Major UK online bank experiences drops across key terms: Google September update 2016
Multiple travel sites: Slight sitewide drops across key search term groups: Google September update 2016
Major UK grocery store terms peak following Google September 2nd update 2016
Major Electrical retailer experiences drops across category pages – Google September 2nd update 2016
Google 13th-15th flux – A continuation of September 2nd update?
Major UK clothing retailers experience slight drops across category pages – Google September 2nd update 2016
Only a few search terms appear to be dropping out of the SERPs here, with lots of previous flux, but there is evidently negative movement. We will be collecting more data and will likely be able to see more evidence of the downward/upward trajectory of any URL within the coming days.
This site has also experienced drops on multiple dates (including the 15th August, 2nd September and 13th/14th).
Ecommerce category pages, product pages and faceted search affected by Google September update 2016
We’ve said it before, we’ll say it again, it seems that the pages affected by this update – according to Pi’s data – were content-light category and product pages. Yes, even those that experienced a peak in performance. This suggests to us, once more, that Google is in the process of testing thousands of results, which is the likely reason for all this arbitrary SERP movement.
More proof that Google is testing results: Image bar disappearing and reappearing in search results
Corey Thompson (Pi Datametrics Data Stategist): We’ve been collecting data to discover whether or not branded searches contain the image bar within the SERPs. The graph counts how many brands (from our 1,000 brand sample) contain the strip. There was a 160% drop in the appearance of image bars on the 9th September, an improvement on the 13th, and then an even larger 203% drop on the 14th.
See below for more examples of Google conducting local and knowledge panel testing.
He reiterated that Google is continuously updating its algorithms, and that the weekend flux was in no way related to Penguin!
The guttural cry of SEOs can be heard the land over as Google supposedly updates its algorithm on Friday, September 2nd (pretty cheeky of them to do so whilst a good percentage of us were living it up at Brighton SEO, don’t you think?).
There were rumours of:
- Penguin roll-outs (affecting Personal Blog Networks [PBN’s])
- Local pack result updates
- Knowledge panel updates
- Spam targeting
So, does anyone really know what Google’s playing at?
But that doesn’t stop us from taking to Reddit, Twitter and Disqus comment sections to discuss our theories and share our tales of search woe.
September Google update 2016: Reddit
Search Engine Roundtable
Many marketers have taken to the comment section of SE Roundtable to report on their ranking changes and traffic drops (166 people and counting so far).
Here are a couple of comments taken from the forum:
September 2016 search update: Search engine monitoring tools show SERP movement
It’s a bit of a chicken, egg scenario. Did the SEO’s discover the initial SERP flux, or were they buoyed up by the sites below?
Yes, many search engine weather and temperature reports showed some clear search activity, but did this just lead to a lot of panic and conjecture?
September 2016 Google update – Local pack results worst affected?
Glynn Davies (Head of Search Strategy): Moz SERP feature graph shows a big drop across local pack data 31/08/16 – 01/09/16, so there is perhaps some truth to the local pack hypothesis, with a bit of a corresponding jump in knowledge panel results too.
Since “Local pack results” seemed to be the buzz-phrase of this “update”, we decided to do some more testing to see if it has been affected as much as we all think it has.
The Pi Datametrics Platform is currently collecting local pack and knowledge panel data, so all of this is original and sourced first-hand.
Minor flux across local pack and knowledge panel for branded searches
Corey Thompson (Data Strategist at Pi): Testing the top 1000 branded terms in the UK, we detected a small switch in SERP localisation around this time; with an increase in local packs being presented. As local packs are linked to the type of knowledge panel (brand vs local), knowledge panel maps have shown a similar decline.
Knowledge panel header count is subject to flux or testing
We have been recently monitoring and collecting data on all different qualitative features of the knowledge panel, including headers:
Corey Thompson (Data Strategist at Pi): Around the time of the September 2nd Google “Update”, we realised that the number of headers presented in the knowledge panel across UK businesses showed a 7.6% drop. This could suggest that tests were being carried out on this particular area of the SERPs.
Pi Datametrics synopsis on September 2nd Google “Update”
We’ve monitored multiple industries to spot any potential drops or signs of a significant update, but found little to suggest something big is happening.
Aside from regular fluctuation and drops across typical problem sites (ecommerce: faceted search with poor canonical linking/landing page strategies, and general cannibalisation/stolen/ duplicate content/semantic flux) for a handful of search terms, we haven’t noticed any major issues or patterns. A bit of correlation isn’t necessarily causation.
At Pi, we think that Google is simply testing different areas of the SERPs, perhaps a little more rigorously than usual. This sentiment is conveyed by others across search too, including Google themselves (well, John Mueller anyway):
Nevertheless, we will be the first to hold our hands up if fluctuations intensify, and will continue to monitor multiple industries, as we always do.
If you’ve experienced any effects of this potential update, feel free to drop us a comment below.
We always welcome your input and feedback!
What’s more, if you have any questions about the local pack or knowledge panel data we’re collecting, get in touch here, or alternatively: