Cover of masterclass

Website Management & Governance Training Course ($29.99)

Finding BIG problems on a website is easy.

Maybe the Information Architecture is a mess. Perhaps navigation is useless. Loading times might be a joke.

But big problems often need big solutions. That means lots of manpower, time, skills, tools and general capability.

Yet most web teams are catastrophically under-funded.

So, although it would be wonderful to plan proper UX sprints, do HEART analysis, dig deep into Top Tasks, etc—simply keeping the show on the road takes almost all your time.

Without a transformational change in resourcing, big problems are essentially unfixable.

Sad, but true.

So, we should just give up?

Nope. But instead of endlessly trying and failing to fix the biggest problems, focus your energy on those where you can make a meaningful difference.

Think of it this way.

You alone cannot stop wholescale industrial pollution. But you could clear the litter off your street every week.

A small change delivered with limited resources making a big difference in a specific area.

The same applies to web.

Maybe you can't fix navigation across your entire website. But you can radically improve the most popular content by focussing your team's limited resources in a specific area.

Data from Content Groups in Google Analytics is an incredibly powerful way to do this.

Analysing Content Topics with Content Groups

The main benefit of Content Groups is that they allow you to rank the Content Topics that are truly of most interest to your users.

This ranking (along with some extra analysis in Excel) tells you exactly what to focus on for best return—and, just as importantly, what to ignore.

I have analysed hundreds of Content Topics and tens of thousands of pages in this way and it always works.

Always. (I'm serious.)

Here's 3 ways to do it...

  1. Quick Start Prioritisation
  2. Progression Prioritisation
  3. Deep Dive Prioritisation

Note: The illustrated examples below are taken from my free Content Topic Analysis spreadsheet (XLSX 100KB). Feel free to download, modify and extend it as you see fit.

1. Quick Start Prioritisation

Quick Start is an effective way to "clear the fog" created by misleading traffic data. It exposes the content your team needs to prioritise at the highest level.

Even if you have no time to work on this content, the Quick Start list is still a powerful way to communicate what you "should" be doing to senior managers.

Step 1

Identify your website's Content Topics and set them up as Content Groups in Google Analytics. Watch my step-by-step explainer video for instructions.

When ready, export the data from Google Analytics into the free content analysis spreadsheet (XLSX 100KB).

Content Groups in Google Analytics

Step 2

Isolate the Content Topics that have the highest total number of sessions and the lowest average number of pages visited per day.

These are the best metrics for prioritising your team's activity. This is because focussing resources on topics with:

  • A large number of sessions = Maximal benefit to users
  • A low page count = Minimial "investment of effort" needed by your team
Spreadsheet showing average page count per day and average sessions per day

Step 3

Calculate the average sessions per page per day for each topic.

To do so, divide the total sessions per topic by the average page count per day.

Now, rank the results from highest to lowest.

Spreadsheet showing average sessions per page per day

Prioritisation results

A team with limited manpower and time should focus its attention on the topics with the highest ranking—and ignore everything else.

By working on these topics you will generate the greatest benefit for users at the least "investment of effort" by your team.

As an aside, it is worth noting that you may sometimes find the ranking that emerges from this system counterintuitive. Don't worry. That is a feature of the decision algorithm.

You see, the algorithm does not prioritise "big-ness".

Instead it prioritises content that a real team with limited manpower can actually work on. To do so, it gives most weight to topics with the fewest pages, not to those with the "biggest" activity.

That's what makes it so useful.

But, we can do more.

Step 2 Progression adds a useful extra layer for deciding which topics to work on first.

2. Progession Prioritisation

In terms of data from Google Analytics, so far we have used:

  • Sessions per topic (from Content Groups)
  • Count of pages visited within each topic (from Content Groups)

Now let's add:

  • Count of all live pages within each topic (i.e. not just the pages that are actually visited)

Re-run the algorithm as above and add Steps 4 and 5 below.

Step 1 .. 2 .. 3

As above.

Step 4

For each topic, calculate the average count of pages visited against the total count of all live pages within that topic. (The count of all live pages may be available from your CMS or other crawler.)

To do so, divide the average count of pages visited by the total count of all pages.

That will generate a percentage.

Spreadsheet showing total live pages and percentage of pages viewed

Step 5

Take the percentages from Step 4 and add them as a weighting to the results from Step 3.

To do so, multiply the percentage (from Step 4) against the average sessions per page (from Step 3).

When done, re-rank the values from highest to lowest.

Spreadsheet showing the weighted average sessions per page per day

Prioritisation results

Great! You now have even better insight about where to focus your team's limited manpower and time.

Applying the percentages from Step 4 has exposed the Content Topics that have the fewest number of unvisited (i.e. useless!) pages.

You should prioritise those topics.

"But wait! Shouldn't I focus on the topics that have the most unvisited pages?! Surely so many useless pages shows something is badly wrong."

Yes, something is badly wrong with those topics. But remember, your team has vanishingly tiny amounts of free time to spend on discretionary UX work.

You need to focus that time to get the best possible return from the effort you put in.

All else being equal1, the best way to do that initially is to work on topics that have the smallest number of unvisited pages.

The reason is that a small number of unvisited pages (likely1) suggests that those topics already have a coherent content design solution in place.

That means your team will not have to get bogged down in fundamental redesign work. Instead, you can focus on immediate enhancements (e.g. readability) that will deliver the best bang-for-buck for users.

Great!

But again, we can go even further.

3. Deep Dive Prioritisation

So far we have analysed and ranked Content Topics as discrete units.

But Content Topics are not fundamental. They are composed of pages2. Dozens, often hundreds of pages—and some are more important than others.

After you have isolated the main topics to work on, you need to deep dive into each topic to decide which to prioritise.

Usefully there are lots of ways to do this.

For myself, I prioritise pages based on user interest—using Page Views from Google Analytics as a proxy. I then add other page-level data to help triangulate where to spend my limited resources for best effect.

The good news is that small numbers of pages usually receive the overwhelming majority of user interest, i.e. 20% of pages get 80% of traffic.

To see how this works, let's re-run the algorithm above and add a new Step 6 and 7.

Step 1 .. 2 .. 3 .. 4 .. 5

As above.

Step 6

Starting with the highest ranking topic, export page-level data for that topic from Content Groups into Excel.

Rank all the pages within the topic using Page Views (from highest to lowest).

Calculate the percentage activity for each page against the total Page View activity within the topic.

Then allocate each page to one of two cohorts.

  • Pages in the top +80% of traffic
  • Pages in bottom 20%

On a limited team you should prioritise pages within the top 80% cohort—this usually contains the smallest number of pages.

Note: Remember to always exclude landing, navigation and wayfinding pages from your Content Groups in GA. These are not useful for deciding priority and will badly skew your data.

Spreadsheet showing page level analysis

Step 7

Finally, add any extra page-level data you have to help decide where to allocate your limited manpower and time. Typically I include:

  • Word count per page (e.g. from VisibleThread)
  • Readability score per page (e.g. from VisibleThread)
  • Average time on page (from GA)

Putting it altogether, my approach is to prioritise pages within the top 80% cohort that have a:

  • Low word count: I prioritise short pages as they are generally easier to improve quickly
  • Poor readability score: I prioritise pages with a low score as they have lots of easy to fix issues
  • High average viewing time: This metric is imperfect, but I keep an eye out for high average times

Focusing your team's effort on these pages will generate the maximum benefit for users at the least "investment of effort" for your team.

Spreadsheet showing advanced page level analysis

And that's it

You should now be able to see how a team with very limited resources (i.e. most teams!) can make real improvements by prioritising manpower and time in clever ways.

So take heart.

You can make a difference!

This article is Part 2 of my series on prioritising activity on under-funded web teams. Read Part 1 to discover why Content Topics are the best way to focus your team's energy and time.


Footnotes

1 This heuristic is a "rule-of-thumb" only. It assumes a certain quality of content design that may not be in place. Your professional experience and insight should override it as needed.
2 And documents and videos and sub-sections and panels and many, many other discrete content elements.