Content is often relegated as something “soft” and absolutely far from data and operations.
Too bad, it’s actually the other way around.
Content Management on a large scale is not only for “creative” people but requires coordination and processes (Operations).
I will show you what you need to know to level up and abandon the old and outdated ways of producing and managing content.
Table of Contents
Old VS New World
The old way of creating content consisted in hopefully ranking on Google.
Social media posts were often disconnected and even unrelated to your website content.
The current “metagame” heavily involves distributing and repurposing your content across multiple channels.
“Say 1 thing in 100 different ways”
And for many reasons:
- social media platforms have lower reach than before. Only a fraction of your followers reads you.
- Your content should be aligned across channels.
- Creating content from scratch is expensive, especially in B2B.
- Content always decays, some channels (e.g. social media) make it happen way faster
The 0-click Scare & AI
Despite a lot of “convincing” arguments that websites are getting less traffic and that SEO is dead (a zombie since 2010), reality is different.
The normal professional dealing with normal websites (like us) doesn’t need to do anything extraordinary to excel.
The main issues have never changed and involve:
- politics
- not having systems
- lack of interest
This article will clearly cover the 2nd point as the other 2 don’t depend on you.
Having a clear roadmap and idea of what to do and failing at it is your problem!
For sure, it’s wise to measure your visibility and CTR over time to see if there was any real impact:
You can repeat the above analysis once in a while to see if your CTR by Position dropped over time.
Subscribe to my newsletter to receive the BigQuery Handbook with the code and a Content Handbook:
Then, you can simply build a curve:
Speaking of which, the real problem aren’t 0-clicks but the fact that you don’t know what brought you traffic from Google Search:
In this specific case, I know the 2 articles got significant nice traffic from anonymous queries but I have no clue for what.
Remember, industry data, even if correct, may not affect you at all.
You need to see what YOUR data suggests and avoid other people’s work as the ultimate source of truth!
The same applies to AI Overviews, you need to select which keywords matter the most to you and validate the results:
The code and all will be released soon.
Instead of seeing everything as a threat, measure your risk and develop new skills.
Content Management – Plans/Inventory
If you have worked with publishers, you know they have a content plan.
Tools like Airtable, Nocodb, Monday and ClickUp give you a unique place where you can list down your content, label it accordingly and keep track of your posting schedule.
Mind you, this isn’t your average spreadsheet to dump your info inside, it’s smarter.
You use views to have everything in one place and make it easier to editors and writers alike to work on your plan.
Spreadsheets aren’t databases at all, they are messy.
OK, you now have the option to create cool templates but the issue is that a spreadsheet is only suitable for basic use cases.
Forget complex data management, workflows and even proper order.
Content Management requires some good resources in actual scenarios… you need to invest to avoid drowning in content debt.
Labels, Taxonomies & Metadata
The real added value of content plans is that you can add labels and metadata to enrich your analyses.
You see, pages should be grouped otherwise it’s hard to give proper advice.
People don’t modify one page at a time unless it’s a small personal website (like this one you are reading).
The solution is to find a way to classify every page to analyze groups instead:
The best method is to simply label your data. Human-based, sometimes prone to error but the highest quality you can get.
The other approaches make sense if:
- the website is already there and you have a lot of pages
- the URLs kind of follow some patterns
So yes, in most cases you will rely on the latter approaches or use something like Fuzzy methods.
The best solution is to have labeling integrated into your CMS, for example:
- flagging pages as product or category or article
- author information
- tags/categories
This will save you a lot of money and time later.
ContentOps
A fancy word to say that you need processes and systems to manage and produce content.
Big publishers and professional content creators don’t simply hit publish and write once in a while.
There are a lot of hidden mechanisms that you can also use to improve your content efforts.
We’ve seen Content Management before, now it’s time to talk about Operations, what is needed to efficiently manage your content over time.
Automation
The current options are Make and n8n to automate most of the content workflows you see online.
They act as connectors and orchestrators for your content tasks.
n8n is optimal for developers and technical people.
Make caters to an audience of marketers and non-techies but I don’t find it intuitive at all, I struggle with it!
The real issue is that many people stop there and actually think automation solves every issue.
As explained in my other article about strategy, it’s not that simple.
AirOps is a nice premium alternative for SEO tasks and they have a lot of available options for integrating AI into your workflows.
Depending on the resources you have, you can even build internal tools and DIY.
You can also take inspiration for some processes to this article on Content Strategy and LLMs by Caitlin Hathaway.
Distribution & Repurposing
Content should never be relegated to one channel because it’s extremely risky (and you miss a lot of opportunities).
Channels have synergies with each other and you can recycle a lot of content with less effort and lower costs.
Imagine you are a B2B SaaS and want to create some content:
- you pay for a writer
- 50 articles are drafted
- you publish them and then what?
You spent a lot of money and now you need to wait for the magical SEO dust to work.
Don’t be that person and plan differently:
Social media give you a quicker feedback loop so you can easily understand what people resonate with.
At the same time, social content decays quickly and you need daily and consistent posting to stay on top.
Managing content isn’t only about doing boring archive tasks, it’s also about seizing opportunities and being extremely consistent.
Content requires a lot of work if you want to stay on top.
I myself heavily invest in Content Distribution & Repurposing as explained with the pictures below:
LLMs, AI & Stories Of Hype
LLMs have some great use cases actually!
My problem with the “AI hype” is that companies overspend in the wrong things.
LLMs will not create perfect content that people want to read, they are nothing more than a useful aid.
Chances are you aren’t missing out on anything… if you don’t have a solid understanding of content first.
But LLMs provide smarter methods to deal with text data (language models, you know):
- summarization of texts.
- brainstorming
- translation/proofreading
Some of these use cases look weak but they are actually extremely powerful.
I use Claude not only for coding but also to restructure ideas and get inspiration:
If you want to know other tools to use, check out my SEO Analytics Stack, it’s pretty much the same for every other use case I do in Web Analytics.
Yes, I will also write some more about LLMs and AI, don’t worry, more to come!
The Data Side
Read my other article if you want to know more about Content Auditing.
The gist doesn’t change, you can evaluate the success of your content with data.
HOW you evaluate content is what makes or breaks your content strategy. Too many companies use traffic metrics as the main reference and this guides them in the wrong direction.
Traffic is an indicator of general interest but almost never a KPI.
For sure, a publisher making money with Display Ads is a different story but even then, they’d want to optimize their RPM, pages visited per session and scroll depth.
I’ve developed several methods and workflows that I cover in my products:
If you want to learn more about all of these topics, I’ve prepared an Analytics for SEO course that covers both theory and practice:
SEO Content Is A Lie
You’ve heard of it, SEO content is used to refer to the slop content that populates search results.
Many believe that you can just create specific content to rank on Google and getting more traffic/visibility will be enough to solve all of your problems.
In reality, it’s the opposite.
I’ve consulted companies with tons of organic traffic getting $0 back.
That’s because they didn’t carry proper research and simply target the keywords with the highest volume.
This method of working has been ravaged by the Helpful Content Update (HCU) and by the increasing competition.
(Google didn’t exactly hit all the websites that deserved to be hit and did massive collateral damage but hey, they eliminated a lot of low-effort niche websites).
There is no such a thing like SEO content. You either create it and also rank on Google or you don’t do it all.
In the long-term it will save you a lot of headaches (and money)!
The “ROI”
I have always been against the ROIfication of work.
There are cases where estimating returns makes sense and you should do it:
- Performance Marketing (e.g. PPC)
- Data solutions (it makes sense to estimate cost and time savings)
For content, it’s hard to say forecasting and even ROI makes sense.
Usually these doubts come from organizations that don’t have enough budget or trust.
Avoid them like the plague!
You may be asked to provide an estimate to get some buy-in though…
it’s a necessary evil in those cases.
Once you realize that Web Data is pretty much “inaccurate”, your mindset will drastically change:
This is why Analytics is no longer optional if you work with data, you need to understand how data plays a role in your daily efforts.
The Basics You Can Measure
If you want a quick way to see if your content is working (in the long term), inspect your branded queries.
People can only search for you if they know you exist.
Once you are in their mind (top of mind), it’s done.
If you overcomplicate it and think that you need crazy tests… it’s going to be a big issue!
A website that merely gets organic traffic from Google or has a heavy social activity and doesn’t get branded queries is a bad sign.
Sure, you can believe in the tale of “free” organic traffic but you are just renting Google’s property.
The Reason Why Content Fails
As this article contains many uncomfortable truths, it’s time to double down and close with the best one.
Have you ever asked yourself why some companies can’t succeed with content?
Well, there are TOO many reasons:
- slow publishing frequency
- weak value proposition (the company offer is bad)
- lack of domain knowldedge (unqualified writers)
- no budget
- lack of processes and systems
Not every company needs to create the best content BUT it’s highly recommended to have a minimum social media presence on channels like social media.
A Bright Future Ahead
No, content won’t die and AI will not make everyone unemployed.
If you consider content as a sequence of strings without a meaning, you got it all wrong.
Content is much more than that and also needs people to be managed and coordinated, as well as preserved over time.
As you may have noticed from this article, content allows you to:
- Express your value proposition
- Get that mindshare for your customers, they will think about YOU
- Document your journey
- Build and authority and trust
Analytics and AI/LLMs are aids to help you in maintaining, updating and creating content but never replace your human component; they enhance it.
We live in the best era for expressing our thoughts, make that bold step and start writing!