Google: Penalized Site? No, Fluctuating!

Yes, Google fluctuations are back. And with them, the despair of many webmasters who have seen drastic reductions in visits to their sites and, therefore, their income.

The strange thing is that owners of sites that benefited in the SERPs (search results) from this phenomenon have not shown any satisfaction. Indeed, there are only 10 positions available on the first page, and it’s natural that if someone drops, there’s always someone else who rises in their place. In this regard, I’ve made some considerations that could represent a hypothesis:

► Sites that have risen are spam-filled satellite sites, where positioning is not even tracked or analyzed, because these sites usually have a short lifespan and tend to rank for any keyword. In this context, spammers certainly won’t announce that their site has moved to the first pages of search results for a specific query. This would support one of the theories from March, namely the suppression of anchor texts (which also happened this month) to bring SPAM present in the SERPs to the surface and eliminate it (algorithmically/manually).

Now I would like to compare, schematically, the fluctuation of March with the one currently underway, as they indicate a mutation of the phenomenon probably due to the integration of new algorithms adopted by the Californian search engine.

Similarities

• The fluctuation is limited only to certain sectors or even keywords
• The sites affected by this phenomenon are from the most diverse sectors
• Changes in anchor text counting
• Concurrence with backlink export (detected by Webmaster Tools)
• Concurrence of export on Google Toolbar’s PageRank (a statement to be taken with a grain of salt, as it’s a pure intuition and supposition based on some anomalous update cases during the fluctuation period).
• Concurrence with the downtime. Indeed, this month, site drops were detected simultaneously on June 26, June 28, July 02, July 07, and July 12 (based on collected data, naturally)
• Return to SERPs characterized by ranking improvements

Differences

• The amplitude of the sinusoidal wave is much wider, ranging from 7 to 30 days (counted up to each day), whereas in March’s fluctuations it was practically on an hourly basis
• Alignment of all datacenters; unlike the previous one, this time all Google datacenters seem to be practically aligned downwards (although in the last few hours, misalignment seems to be occurring for a few hours)
• A large portion of sites experienced the SERP downturn only for secondary keywords, while the primary keyword (contained in the homepage in 90% of cases) remained in place. This excludes the possibility of site penalties by 80%.
• Anomalous site positioning in the very last search results, just before omitted results, for secondary keywords affected by the fluctuation
• Lack of apparent reasons: unlike before, this time authoritative sites with high PR (even 5 or 6) have suffered the same fate, along with small and relatively young sites.
• Revaluation of anchor texts; while in March the number of anchors for a specific keyword was equal to the total number of search results for that keyword, today it seems that affected websites have lost some links with anchors important for the selected keywords.
• Supporting the previous point, for some sites, Webmaster Tools has indicated fewer referring sites compared to the period before the fluctuation.
• The sinusoid affected the entire domain, even with completely different long-tail keywords (used by generalist sites).

Therefore, based on the observations made in recent weeks, the previous hypotheses must be revised and adapted to the current situation:

March 07
1. The possible cause of fluctuations is due to the change in how backlink anchors are calculated (in fact, during the waves, the number of anchor text results is exactly equal to the number of search results, which is mathematically impossible in normal conditions).
July 07
1. The modification of SERPs for anchor texts is not the cause but one of the effects of the actual reduction in link counting (which is the predominant factor in current rankings). By removing links to a site, it can only drop in the SERPs.

March 07
2. Based on the first hypothesis, it is plausible that Google can activate/deactivate this option at any time.
July 07
2. This statement is still current; in fact, it seems that Google is able to determine the timing and methods for the “temporary” elimination of links.

March 07
3. The most probable reason for this appears to be maintenance operations and, above all, the cleaning of SERPs from SPAM (in fact, spam pages rarely have a considerable amount of quality backlinks), and this theory is also confirmed by the return of most of the pages that disappeared from their SERP positions immediately after the sinusoidal event.
July 07
3. Indeed, in my opinion, it still seems the most plausible hypothesis. Although in recent weeks, the cleaning work might have been performed algorithmically rather than manually (this can be deduced from the wider sinusoidal waves, meaning the stability achieved by all datacenters allows for the creation of programming rules capable of automatically removing SPAM from SERPs).

March 07
4. Furthermore, Google seems to act at its discretion, targeting and analyzing sector by sector, keyword by keyword. In fact, while some sites in the same sector are fluctuating, others from different categories are static and show regular SERPs.
July 07
4. Situation unchanged.

March 07
5. The loss of positioning, instead, can be due to the reduced push in terms of backlinks caused by the exodus of pages and anchor texts (see lower data between before and after the sinusoid), perhaps due to increased severity of the anti-duplication filter and modifications made to Google’s algorithm to curb the Google Bombing phenomenon.)
July 07
5. As with point 1, this is not a cause but an effect of the backlink reduction for sites affected by the sinusoidal wave.

March 07
6. To avoid fluctuations, one needs high popularity to sustain the site in the top positions even in this delicate situation. In any case, it is advisable to promote the site globally, not focusing only on a specific keyword or keyword variations.
July 07
6. Unfortunately, this statement has been disproven, as highly popular sites with well-studied and differentiated long-tail keywords have been affected (the drop affected all keywords except the primary one). In fact, concerning this aspect, it seems like a diametrically opposite solution to the one obtained in March.

So what can we do for lasting organic positioning? Is it still worth writing and creating resources for users?

Yes, it’s the only certainty. On the internet, it’s not the search engine that converts; it’s the users who make the conversions. Creating tools and documents aimed at users always pays off, always.