The Seotistics newsletter is written by Marco Giordano, a Data/Web Analyst with the goal of combining business and web data. Tired of the usual boring Analytics content without any business impact? Seotistics teaches you how to use Analytics, web data and even content in your workflow while helping you with Strategy.
Share
π 5+1 Data Mistakes That Destroy SEO Projects
Published 11 months agoΒ β’Β 8 min read
β
Use Data Or Be Used By Data!
β
The June 24 issue of Seotistics is here for you!
Everyone talks about data or even AI but random knowledge is worthless.
You need to be in the industry and know how to piece things together.
I'll walk you through the most common mistakes I've noticed in the past few years!
P.S. I will be in Berlin this week if anyone wants to meet.
Please move this email to your Primary inbox or reply to it. This is to prevent Seotistics goes into spam by accident. Gmail users can read this tutorial to do it.
The other common trap is not understanding sampling and the differences between data sources.
Downloading GSC data manually is NOT the same thing as using the API or the BigQuery bulk export.
Building a reporting system based on manual exports is the recipe for failure.
The solution to this is building a data dictionary, a magical document where you write all of the caveats to pull and combine the data + the interpretation.
N.B. Did you know that CTR and Avg. Position aren't available by default if you use bulk export? Now you know it.
Most of your SEO or Data Analyst job is about explaining technical stuff to laymen.
You'll see a lot of nonsense and repeated questions... so get good at documentation and communication!
#2 The Time Pitfall
In finance, you may have heard about opportunity costs.
If instead of updating an old cluster, you would write a new one, which would be the better option?
Many owners rush to publish new content and donβt update their old material because they think there is no gain in doing so.
Think again, does it cost more time (and possibly resources) to write new articles or to update old ones?
This goes case by case but you can say that in 90% of the cases updating content takes less time than writing from scratch.
Now I ask you, is it worth spending 4 hours updating meta descriptions?
Answer: probably not.
For low-adding value tasks, use automation, Python is your friend.
The opportunity cost of 4 hours can be invested into more profitable activities, so what moves the needle.
In SEO, it's the same, there are high opportunity costs:
Updating what doesn't move the needle
Focusing on observations (we have 20K Users, and so what?)
Obsessing over Tech SEO with a small website
β³ Optimizing Time
SEO means being torn by different choices and limited time/budget. Guess what? You need to choose.
The example below shows a very common use case:
Time it takes to perform an SEO task: 4 hours per week
Automation script: 10 hours to produce?
This is under perfect conditions, assuming you can craft something ready for use in only 10 hours.
There is a striking convenience in moving forward with the solution.
Those who think time isn't a KEY element in SEO are fools. Timeliness and execution speed are what make you money.
Now, let's see a more realistic use case. You need to build a solution to monitor SEO performance.
In many cases, you can get away with Looker Studio and call it a day.
If you are interested in a custom SaaS, things get a little bit different.
So you'd look at the following values:
Cost to develop SaaS
Time required to train someone
Price per hour
Total cost
Total Benefit
Many often ignore time but it's a crucial point in Analytics (and accidentally SEO).
Making faster important decisions is key to winning and slow decisions are often pointless in many instances.
#3 Statistical Problems
Why do most SEO case studies donβt make any sense? Think about it.
Why correlation doesnβt equal causation? This is also a popular one.
These are some of the most frequent questions in the SEO industry and why you need some foundations in Statistics.
Many often skip numbers altogether and only focus on software development.
This is perfect if you just want to build a SaaS or a tool but terrible for analysis.
The majority of SEO case studies are Observational, meaning that you can simply see if there is a correlation (at most).
Full explanation in the next issues! The above example is hard to grasp without further context.
Using GSC Data to do some good ol' Analytics is what you would define Observational.
Experimental studies take it to the next level and can be used to prove causation.
These are rare in SEO but an example could be testing the impact on rankings if we change title tags.
Oh wait, this topic was described by Giulia Panozzo in her talk. Do yourself a favor and read this.
In short, do NOT:
take case studies as applicable to your website
objective truth or ultimate evidence
invite to alter your strategy
explanations of why something happened to your website
Think they can explain the HCU or Core Updates
HCU hit a lot of websites indiscriminately, there is no scientific and ultimate proof it hit those with a blog roll, for example.
Now we have more evidence and clues but at first, people were claiming it was all about content quality (lol).
#4 Hindsight Bias
Why don't I read many Core Update analyses? Because of hindsight bias, one of the sneakiest biases ever.
"I knew this website was going to be hit, the content is bad"
Stating you were right and you knew it after something happened. You didn't know before, you validate your idea after the outcome is known.
Many bad websites don't get hit and you hardly see people commenting on them.
It's easy to evaluate an event that already happened and state your opinion but this is wrong.
The HCU showed how strong hindsight bias is.
β
Is it possible that everyone knew in advance which sites would've been hit?
Data analysis and content planning are 2 cases where hindsight bias does the most damage:
"I have always been sure this cluster would have performed greatly"
and when it doesn't, you blame Google.
Overrating your predictive abilities makes up for a great deal of problems.
Always think twice before evaluating an outcome and ask yourself if you knew before.
#5 Accumulating Debt
Crazy Python scripts won't save you from debts inside a company.
Some may say tech or SEO debt but I think that content debt is much worse.
Imagine having 1000 articles and no one cared to update them in 10 years (oh, this actually happened!).
Thin or outdated content hinders your organic performance but let's forget about SEO for a minute.
Yaaay, Content Decay destroying your performance!
To update articles, you have to:
Pay someone to do it
Spend time working on it (or you do it yourself)
Research, edit, proofread (extra time)
Be sure you repeat this process every once in a while
This should be one of your main interests in Analytics, not simply how to automate processes or to pull GSC data (which is obvious).
I will talk more about Content Management in the future because it's a key area where many companies still fail.
Needless to say, it intertwines with SEO and Data quite well.
If you want to skip a lot of time spent on research and mistakes, I have the solution for you:
[Analytics For SEO Course: Preorder]
I am already working on v2 which will contain something like 40-50% more content and some stuff that I need to add.
Come join the Seotistics Academy!
You will:
β Use GSC and GA4 Data to their fullest potential
β Learn Python/R/SQL for your needs
β Get a complete blueprint for auditing websites
β Learn how to 10x your productivity
β Learn BigQuery to work on large websites
I teach you what's needed to go from 0 to a professional Data Analyst.
Even if you leave SEO, the foundations are the same for other jobs!
P.S. (Ab)use the referral system and get additional discounts π
The Seotistics newsletter is written by Marco Giordano, a Data/Web Analyst with the goal of combining business and web data. Tired of the usual boring Analytics content without any business impact? Seotistics teaches you how to use Analytics, web data and even content in your workflow while helping you with Strategy.
Read more from Seotistics - Web Analytics + Business + Strategy
Use Data Or Be Used By Data! The May 26 issue of Seotistics is here for you! Knowledge graphs have been making their way through in the last few years. Small and medium companies have no clue but corporations are currently employing such technologies! So why am I even mentioning them? And what have these strange things to do with AI? P.S. This topic is huge, expect a more proper coverage in either my website or course. Please move this email to your Primary inbox or reply to it. This is to...
Use Data Or Be Used By Data! The May 19 issue of Seotistics is here for you! AI is making some nice advancements as of late and I am here to show you some ideas. I had to talk about Knowledge Graphs and Taxonomies but I will postpone it... These tools are quite relevant today and can massively help you. Mind you, Analytics is NOT only tools and tricks though... Please move this email to your Primary inbox or reply to it. This is to prevent Seotistics goes into spam by accident. Gmail users...
Use Data Or Be Used By Data! The May 13 issue of Seotistics is here for you! There is a lot of buzz around content and its future. I am more than bullish about it... you know why? Because many people haven't really grasped the essence of content. You will soon understand why content is so important. P.S. I've just launched my new course "Think Like a Web Analyst" yesterday. Please move this email to your Primary inbox or reply to it. This is to prevent Seotistics goes into spam by accident....
