One of the recurring themes I’ve been seeing in the past few months, comes from web publishers trying to confirm when and where Google updates actually happen. This type of thing has been going on for years and it’s not a new topic, what is new is that Google is now confirming that there is a lot of contradictory information out there, making things more difficult for everyone. Having been around for a few years, I’d like to provide a little perspective on the current situation.
What is a Google Update
There’s a phrase from Cool Hand Luke that sums up this situation, “what we’ve got here is a failure to communicate”, or put more simply, we aren’t speaking the same language as Google. To Google the word “Update” has a very specific meaning, most likely it means a deliberate change to the way things are ranked in the SERPs. What it doesn’t mean, is a change to the underlying data that is used in the algorithm, that results in a change in the SERPs. Google is constantly crawling the web, seeing new pages, new links, new social activity, new user data, link rot, disappearing content, and hundreds of other factors, and making ranking adjustments based on those factors. When that underlying data changes, its completely normal and expected that the SERPs would shift, however this doesn’t fit Google’s definition of an “update”. So like any good Philadelphia Lawyer, Google will say there was no update, and have no moral issue about having deceived anyone. Years ago we did get a peek behind the curtain, and learn that Google uses terms like “Data Push” or “Data Refresh” to describe these events, but they’ve gotten a lot better about keeping terms like that out of the public eye. To be completely honest, it’s in Google’s best interest to do so, the less we know about how and when things are happening, the easier it is for them to keep the system from being gamed. Would it be easier if Google worked with our definition of “Update”, absolutely, but it’s not something I ever see happening.
Different Kinds of Updates
Over the past 10 years the algorithm has gotten extremely more sophisticated, instead of thinking of it as one big formula to solve, it’s better to think of it as lots of little formulas that are built into one big result. We’ve learned the names of some of these parts, with Penguin and Panda being the most famous, but there are countless other parts that we don’t have names for, run without any public acknowledgement, but have just as big of an effect on the SERPs. On top of these different parts, they also run in different ways.
To explain how the different parts run, I’m going to use the analogy of an underground sprinkler system. Every year when you start up your sprinkler system, you’ve got manually run it for a few minutes to bleed the air out of the system, in the fall you’ve got blow all the water out of the line, so it doesn’t freeze and crack the pipes in the winter. The algorithm is the same way, from time to time, parts of it have to be manually updated, and these are the times we are most likely to get confirmation something has happened.
Unless you live in California, it’s likely your sprinkler system is set to run regularly without any intervention from you. The algorithm is the same way, lots of little jobs that run day in and day out, without any intervention from humans. As Google crawls the web, and gathers data, the SERPs are affected by these automatic jobs running. Sometimes these updates will be noticeable, sometimes they will be harder to spot. In your world this counts as update, in Google’s world it doesn’t, understand this is going happen regularly, don’t expect Google to confirm it.
If your sprinkler system was as advanced as Google’s algorithm, it would have some sensors built into it. If your sprinkler system was scheduled to run in the morning, but it rained the night before, your rainwater sensor would be “smart” enough to override it’s normal programming. If you’re sprinkler system is set to run every other day, but you have two hot days in row, your moisture sensor could be smart enough to water even though it wasn’t programmed to. The algorithm is going to have parts that don’t run at specified time intervals, but run in response to other signals. These changes may or may not be easy see, and again it may be an update for you, but it’s not an update for Google.
Welcome to the Hotel California
If you’ve been paying attention to the news lately, you’ve no doubt heard California is trouble with a severe drought. Continuing with our sprinkler system analogy, you may not know it, but you’re living in California, and you’re in a drought too, you’re just in an information drought. Looking back five to seven years ago, the publishing community had a much better relationship with Google, sadly those days are gone. Google has lost interest in that dialogue, and the amount of information they are willing to share has decreased. Additionally they are not very interested in feedback about what publishers do or don’t like. For example Google has increased the amount “one box” and “knowledge graph” answers that steal traffic from publishers. As Google moves towards it’s ultimate goal of becoming the Star Trek Computer, expect more of this down the road.
So to wrap things up, you need to understand Google’s ranking algorithm, has and will continue to become increasingly more sophisticated. It will run more and more intelligently responding to changes on the web. These changes can and will result in shifts in the SERPs, some more dramaticly than others, but for Google most these SERP changes will not fit the definition of an “update”, and they will be unlikely to confirm when they do or don’t run.
photo credit wikipedia