Author: Peter Meyers (Dr. Peter J. Meyers) - Researcher Moz.
We want to solve problems, and to believe that everything is under control.
When your home is filled with water, you take a bucket. If your roof is a hole, a bucket can help. But if the cause in a crowded sink the bucket only distract you from the real problem. When it comes to flooding, such distraction can be fatal.
When the traffic drops, easy to panic and to focus on what you can control. Traffic - it's not just something nice to have. He gives us food and shelter. Trying to solve the problem in a hurry, we are not always trying to establish its real causes.
Solution false problem - is at best a waste of time and money, and at worst - the aggravation of the situation.
In any crisis, especially global, the first question to ask is: what is the scale of the problem? Is it a local (related to a specific site) or global?
The answer to this question will not solve your problems, but it can save you from costly mistakes and help find a solution.
Let's start with a fundamental question:
1. Does the traffic down?
This question may seem odd, especially given the magnitude of the impact of the pandemic COVID-19. However, it is important to remember that traffic fluctuations - this is a natural process. There are things such as output, seasonality and so-called "regression to the mean." What goes up must fall, and the business, unfortunately, are not always moving up and to the right.
Let's look at a few ways to check the traffic fluctuations by using Google Analytics.
In the screenshot below - data for the four-week (1-28 March) on the site, which has been seriously affected by COVID-19.
Given the history of the epidemic (WHO declared it a pandemic 11 March), we get a fairly clear picture of traffic falling on the background of the well-known reasons.
However, in most cases to determine what is going to be much harder. Even here we have the influence of the weekend and daily fluctuations.
One of the fastest ways to get a clearer picture of what is happening is the summation of weekly data (make sure that the date range covers a full week, otherwise the data will be distorted).
Now, the overall trend looks much clearer. For two weeks the site has lost more than half of the traffic.
We have limited time range for clarity, but as you get more data to confirm this trend will be quite easy. The chart above covers all sources of traffic.
In terms of SEO - let's add another segment traffic from Google:
This chart contains only eight data points, but it can get quite a lot of information:
- We can clearly see the general trend;
- We see that this trend is almost identical for both Google traffic as well as for general traffic;
- We can see that this site is very dependent on traffic from Google.
Therefore, do not underestimate what you can learn from a small amount of data, if the data is correct.
Finally, let's also take a look, not whether this is a seasonal decline? The fastest way to understand if we compare traffic compared yoy.
Tip: Compare the full week so that you cover the same number of days off.
It's not the easiest schedule for perception, but the green and purple lines, we can see that in March 2019, and the total traffic and traffic from Google were approximately at the same level. Therefore, the current situation does look unusual.
Even if we do not know anything about the context and COVID-19, this rapid analysis allows to notice that there is something serious.
1.1. Does position fell?
Given the apparent drop in traffic from Google, the next question will be: Is your website lost position? (We will consider other possible explanations later).
The Moz Pro, you can quickly assess the overall weekly visibility in the search by clicking on the tab «Rankings» or «Competition». The second option also lets you see whether there are any changes in the industry as a whole:
In the screenshot below, we have simplified it a bit representation (and removed the name of the site and its competitors for privacy reasons), but the basic story is clear: none of the site in question or on its competitors is no reduction in visibility in March were observed.
For more advanced presentation we returned to the «Rankings» tab and choose from the «Rankings» drop-down menu (instead of «Search Visibility»). As a result, we obtained the following schedule:
This visualization is necessary to get used to, but it contains a wealth of information.
The columns show the total number of keywords / phrases that ranked site, and color blocks - ranking range.
Here we can see that the overall ranking has been relatively stable, even with a small increase in the segment # 1-3.
If your account is in Moz connected to Google Analytics, you can also put on a chart traffic over the same period, as shown by the dark gray line.
These charts seem rather complicated, but really this visualization clearly shows that there is a discrepancy between the fall of the traffic of the site and its position in the rankings.
2. I'm sorry if this Google?
Is Google algorithmic change the rules so that it had a negative impact on your site?
Major updates search algorithm can be monitored on the page Google Algorithm History at Moz site. Smaller / daily updates can be monitored via MozCast tool.
It is important to remember that while checking the major changes can be very helpful, the truth is that Google rankings - a real-time phenomenon, and it changes every minute. Only in 2018 Google made a search operation 3234 "improvements."
All SERP Google monitoring tools in one way or another based on the fluctuations in the rankings. In our example, we do not see such changes on the site.
Let's imagine that we have seen the fall of the traffic with a corresponding reduction of positions, and try to determine the reason for - to us or that something has changed at Google.
Below - MozCast schedule with data from our analysis of the January update of the basic algorithm Google:
In this case, we see quite a clear three-day period of the oscillation position. If our traffic fell during this period, it is not a 100% owned prove that guilty algorithm update, but it's solid and well-founded suspicion and useful starting point.
Let's look at the data in the two weeks when COVID-19 has been declared a global pandemic:
In the diagram above, we left the same scale and the 30-day average (taken from a relatively quiet period in the beginning of this year).
Please note that the algorithmic activity in this period was significantly higher compared to the time before and after the January update.
In this case one day (Mar. 18) is not entered in the diagram and exits out beyond it - according MozCast, in this day "temperature" in SERP was 104 ° F.
What does all this mean? There is a possibility that Google suddenly changes its algorithm to allow for more extensive changes in the world. However, we assume that in fact the world affect the performance of the algorithm.
Sites quickly updated, add and remove products and content, and many companies have temporarily or permanently closed. Against this background, we are witnessing an unprecedented shift in the behavior of search users and consumers.
Algorithmic changes may be useful answer to the question, what is the reason, in normal times, but now it only says that the world has been turned upside down. And although it's pretty accurate assessment, it is not particularly useful.
If the traffic is dropped, and no position, it is also possible that the changed user behavior.
We can get some idea about this by using Google Search Console.
Here's a graph of the total number of clicks to the site we are considering for the 1-28 of March (in accordance with GA data):
As expected, the total number of clicks on Google's search results are displayed approximately the same trend as that of the organic traffic from Google to GA.
Total clicks depends on two variables: the number of hits in the search and index CTR (CTR). Let's look at them individually.
Here's a graph of the total number of impressions for the same period of time:
On the chart we see a drop in impressions. This is due not only to the site in question, but also with the behavior of users. People are less likely to perform a search on the phrases that drive traffic to the site.
Now let's look at the CTR:
CTR also fell, and even stronger than impressions. And it's a little harder to interpret.
Based on the fact that we know people are probably less likely to click on search results due to a general lack of interest. This corresponds to the scenario COVID-19. People are less likely to seek the services offered by this website. On the other hand, it is possible that something has changed in respect of the site or the competitive environment, which led to a decrease in CTR.
If you see a drop in CTR without a corresponding drop impressions, review the most recent changes to the site - particularly those that affect what is displayed in the search results (including title tags and meta descriptions).
As for the site in question, in this case it is reasonable to assume that we are seeing a general decline in demand.
3. The world has gone mad?
Spoiler: yes, I come down.
Data Search Console, given above already suggests that we are witnessing a shift in the wider scale and the search behavior. If you want to go beyond your own data, you can learn what's going on in the world, using Google Trends.
In the screenshot below - search volumes on request [movie tickets] for the period from March 1 to March 28:
No wonder that on the background of the outbreak COVID-19 user interest in tickets has fallen sharply in the movie. People who do not go to the movies, do not look for showtimes and ticket prices.
According to long-tail queries Google Trends do not always show an accurate picture, but the data on non-brand queries usually help to understand whether the drop in traffic isolation and concerns only a specific resource, or is it a more general trend seen in the industry.
Last piece of advice: everything that was discussed in this article can also be used for research on the causes of increased traffic.
So, during the epidemic COVID-19 search volumes for some queries have grown substantially.
In the screenshot below - Google Trends data on request [how cut hair] for the same period (1-28 March):
In the graph we can see that people definitely show more interest in the subject.
Usually we do not delve into the reasons for the increased traffic. In our opinion, this is a big mistake. Understanding whether the traffic increase was due to changes you have made or the broader market shifts, it helps to know what has been done correctly, and can repeat this success.
Think more broadly
Over the past few years more and more people say that they do not care about the traffic, they are interested in conversions. Or that they do not care about the position at Google, as long as they get traffic from it.
Our gradual transition to the metrics at the bottom of the funnel makes sense - we are all trying to make a living. However, in extreme cases, we lose valuable information. Focus on conversions, certainly better than the emphasis on "hits", as it was in 1998, but no single metric does not tell the whole story.
Let's say you are tracking only Lida. Sales are growing, leads are growing, a good time. Fine. Suddenly - inevitably - there are any changes and lidam rates are falling. What are you doing? You give up your ability to "read" anything else, except for the last chapter of history.
You know it will all end, but do not know how to be in this situation.
Not realizing the whole chain down from of leads to transitions from the transitions to the positions and the positions of the through rate, you do not see the whole story and do not know at what point something went wrong.
Even in good times, it is a short-sighted approach. Oriented sales culture encourages celebrate and do not ask too many questions.
If traffic is growing, why is it growing? What content or keywords stimulate this growth? What industry trends affect traffic?
If you can answer these questions, then you are able to repeat the success. If it can not, then, when the holiday is over (and it always ends up - sooner or later), you have to start from scratch.
Probably, comfortable to know that the entire industry or the whole world suffer with you. Meanwhile, we hope that we have described the process will protect you from the wrong actions and costly mistakes.
Ideally, this process will also help to uncover areas that may have a tendency to grow and to focus time and money on what really works.