Use Data Or Be Used By Data!

The February 12 issue of Seotistics is here for you!

Last time we talked about Google Cloud Automation for SEO.

I want to show you why processes are crucial and what's the true value of SEO data.

This is the #1 lesson to leverage solutions like Google Cloud or any other meaningful data investment.

Understanding the ROI of data is crucial in this century!

P.S. v7 of my ebook is out this week and expect a BIG announcement too!

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.

Read this in your browser

Quick Recap

In the last issues we have analyzed Google Cloud Automation and what's cool about automation in general.

That's not all as you need to define your outputs.

It's easy to work on cool stuff but it should be usable by your audience.

You may want to read first:

1. 🛠 Advanced Content Optimization With Analytics (Overview)
2. 🤖 3 SEO Processes For Advanced Automation

3. ☁️ SEO Automation With Google Cloud

The Output (What They Tell You)

Looker Studio is the preferred output to showcase many of our results, according to many SEOs.

But not the only one as you will see in the next section.

I am for minimalism and simplicity as you know because stakeholders won't bother checking all the details.

Something basic like this works (mind you, it's only the top!):

What I often include in the topmost part of reports... I am a fan of minimalism and effectiveness.

Yes, I use 3 filters and 4 scorecards because I don't care about getting too much info.

We will see dashboards in depth in the next Seotistics issues!

The claim is that you must have dashboards to show SEO results... which is partially incorrect.

Dashboards are often used to build cool sh*t that achieves nothing.

Beyond Dashboards

As I told you multiple times, you don't have to end your work with dashboards.

Many think that way and it's a way to devalue the quality of your work.

Dashboards are often misused and treated like the solution to all problems...

but there is more:

Google Docs/written reports
Spreadsheets
BigQuery (yes, you can use the output of your queries)
Web Apps

The opportunities are endless and I tend to use all of the above from time to time.

What matters is that you aid SEO execution and contribute to profit.

From Outputs To Actions

Many obsess over shiny objects and tools but forget about execution.

It took me years to understand this simple concept.

The actual ROI lies in swift execution.

Tie your data activities/products to processes and strategic business goals.

Why is your dashboard valuable? Is it even needed?

I explained it with this simple sketch:

We can take it even further and think about data in terms of how they help solve problems:

If you can prove that your content is rotting away, it's easy to prepare a process to optimize your pages periodically.

You can even quantify how much money you are losing because:

If you run Display Ads, you know the RPM
If you sell products, you can notice the drops in revenue
If you want leads, see if you notice drops

Again, the important isn't to be 100% accurate but to show that data will help you tackle a measurable problem.

Example: Enterprise SEO

"We don't trust reports" is probably one of the most common problems in large organizations.

Web data are all but intuitive and you may notice this even with simple metrics:

How would you explain clicks vs users to a non-technical person?
Why there are discrepancies?

This calls for a centralized approach that holds the truth so that no one can question the validity of your results.

Let's see a potential solution to this frequent problem:

Audit the current SEO dashboards/tools in place
Are they tied to an SEO process?
If yes, who's responsible for them?

Once you notice redundancies, cut down the reports/tools used and tie all of them to 1 or more processes.

The ROI here will be covered in the next issue in detail... I already gave you some strong hints.

Example: Many Publishers

The biggest issue for many publishers is creating order out of chaos or finding a signal through the noise.

You can say this is more of a content management problem.

Data help and you can usually do something like this:

There is a reason why I stress the importance of a content inventory...

it's because you can enrich your audits and avoid paying for expensive tools.

A publisher/big content website can't analyze every single page.

Now imagine the "convenience" of having solid systems and processes to find out where to focus.

Remember, if we talk about 20K articles, the numbers involved hit differently.

Think about it, how would you measure the ROI in this use case?

I'll tell you in the next Seotistics issue but please think about it!

What To Automate?

This question sparks many conflicts because some traditionalists will use any chance to justify against automation.

With the caveat that you always need common sense, here's which processes you could automate:

Repeatable
Stable
(Often) Low ROI impact

Adding meta descriptions is a low ROI activity and quite repeatable, although it's not always the same (stable).

Keyword Research requires some human intervention but it's now possible to automate it:

Source: https://seotistics.ck.page/posts/3-seo-processes-for-advanced-automation

Other activities that can be automated in SEO:

This doesn't mean you can use AI on everything, there are better methods with simple coding!

The Difference With Basic Workflows

Stuff running on your laptop is NOT the same as what can run the flows for an entire company.

I know this is obvious for IT people but it's quite common to see marketers analyze huge websites with basic scripts.

This is a BAD practice because:

Too many compromises
Tedious and error-prone
Most importantly, you can NOT analyze that much data locally!
Lack of security

You must ensure that what you do is solid, safe and most importantly efficient.

That's why automation is important and I stress the importance of reproducibility.

How many SEO audits can't be reproduced easily?

This is because there is not enough attention on documenting processes.

Takeaways

Today you've learned that adding value with data is harder than it looks like.

Coding or doing the job is relatively simple compared to orchestrating an entire project or extracting value.

Sure, most of you are SEOs so you don't have to manage a data project but...

if you deal with SEO data or sell services, this is your interest.

Some homework for your mind:

Find how many times you created a dashboard and it didn't get used.
How do you use data?
Have you ever measured the impact of data?

"Wait Marco, you are telling me that it was processes all along!?"

Always have been.

Processes are all and that's how you make the most value out of data.