When I sold some work around a site audit + per article recommendations around keep, rework, delete and categorization, I thought, “this will be easy, this is what I do now.”
But I just keep learning more and more on the job. Here’s me trying to herd the cats of my thoughts/experience here:
Start with “quality” to sort and take a first pass at what to keep, delete
But know it’s just a first pass. It releases the pressure valve to tag something “DELETE” and know it’s not a final decision.
Developing good taxonomies is an iterative process
Anything complex and nuanced that requires real effort to solve for is this way I think.
I don’t mean iterative in the sense of “optimization” but it requires an action reflection cycle, where you’re scouting and gathering decision making criteria blindly, reevaluating, then scouting more.
“Actually there are some patterns here that should get a tag name. Let’s go back and tag the patterns we started to pick up on but just now realize are patterns.”
Hm. I know there are at least 40 posts on x, but I can only find 6 that match the terms we have. Let’s find them and see how the expert talks about those ideas, because they aren’t saying “this post is about sales processes, pipelines, and forecasting,” even though that’s for sure what they’re talking about.
“Let’s create some categories we can automate reaching for through matching on a list of related terms from GSC, Gensim, TF-IDF. Oh, that netted 100 posts. That’s too many. Let’s try again.”
It’s chicken and egg, circular process.
There are three factors to explore in organizing website content based on quality or value
- content performance
- potential
- relevance
Performance is easy. It’s all the links, comments, shares, traffic, that demonstrates there is some value.
Potential is interesting. It’s what we’re really after. Anyone can make sweeping cuts with a large body of content. But how do we make sure what can get nurtured into performing does? How do you wrangle opportunity into actionables that get results.
Potential gets surfaced a few ways – identifying opportunity gaps, reviewing anomalies or outliers like high links low traffic, low word count high traffic but it’s intertwined with context, relevance.
Relevance sounds obvious. But I more mean when you have a comprehensive worldview on something, organizing website content is like world building.
You need infrastructure and the pieces need to all fit together. Maybe you linked out to some external source, then you write something that is a better fit for that link. Multiply that by 100 links that should be pointed internally, that would act as breadcrumbs for users to navigate your thinking around their needs.
To lay pipe for those user journeys, you need the raw materials. Often that is content that should be reworked you otherwise might have thrown away. So relevance is like context and the context is the lens for evaluating potential.
Maybe there is already a book in here
There are a lot of patterns that don’t fit in a neat clear edged taxonomy. These are the emails that are conversational, they feel more like letters, the key points are abstract but anchored in stories, analogies, or references.
Enough of those and you can deduplicate them into book fodder that you could probably throw at a developmental editor, and self-publish by the end of the month.
The nice thing there is you don’t have to just archive or delete a bunch of old content that is going to be too hard to organize into something you’ll get returns on. It’s getting made into something else useful, and so that’s a lens I’ll put on organizing content going forward – “where is the book here?”