Prioritising content work

Our recent discovery research uncovered several new information needs. It was hard to know what to tackle first. Everything on the list could deliver value for disabled people and their families.

High-risk content

We had already labelled some content as ‘high-risk’. This helped us decide what was most important for:

• content creation
• content improvement
• promotion and advertising

We define high-risk content as where there could be:

• a financial impact
• a loss of access to health, social care or education
• a risk to safety if someone does not have that information.

This definition was helpful. But we needed something more than a binary ‘high-risk’ or ‘not high-risk’ classification. We needed a way of putting our content work in priority order.

A new method

This post describes our recent work to prioritise content needs. As usual, we’ll use a ‘test and learn’ approach.

We could have included lots more data sources and complex prioritisation algorithms. We could have spent almost as long working out priorities as producing the content! So we’ve tried to balance complexity with ease of use.

We used 4 factors to help us decide the order that we’ll produce the new content:

• importance to the user
• risk of not having the information
• dependencies between content
• flexibility.

We built a list of all our new content needs in Excel. We added columns to record and rank prioritisation data. We used this to work out and record the order of priority.

Importance to the user

We used search demand and code frequency as indicators of the importance to our audience.

Search demand

We use Semrush for our SEO work. For each item in our content list, we found the total keyword volume for the main keyword. This gave us a search demand ranking.

Code frequency

During our discovery research we built a database of evidence.

Evidence includes:

• interview transcripts
• feedback on existing content
• evidence from our helpline and online community.

We coded all evidence by topic.

This meant we could calculate the frequency of each code in the database. We ranked each item in the list by the code frequency for that topic.

Importance ranking

We added the 2 rankings together (search demand and code frequency). This gave us an ‘importance ranking’ for each item.

Risk of not having the information

We increased the priority of high-risk content. We did this so that ‘importance’ data did not de-prioritise high-risk topics inadvertently.

We weighted high-risk content as twice as important as other content. We do not know if 2 is the right weighting — it’s not based on any evidence! So, this is something else to test and learn.

We then sorted the content list by weighted score. This gave us a preliminary order of work.

Dependencies between content

Sometimes we must publish one piece of content before another so that a second piece can link to it. If the second content item is higher priority than the first, then we should change the order of the list. Otherwise, we block completion of the higher priority work.

We identified several instances where this was true, so we adjusted the order of our list. We noted the dependencies in a separate column so that we would remember why we made the change.

Flexibility

Sometimes content has a deadline.

Examples include:

• a new benefit becomes available
• a Scope campaign needs supporting advice content

Another team is planning a major campaign in April. They need content to support people who might contact them for advice. So, we moved our planned content about this subject up the priority list. We added notes to the spreadsheet so we would remember why we had done this.

In future

So that’s what we’ve done so far. We’ll see how this method works for us. And, there are extra things we’d like to consider in future.

Difficulty of finding information elsewhere

Information could be difficult to find because:

• it doesn’t exist
• there is too much information available — how to choose?

Should we use this as another prioritisation factor? How could we find out?

We could look at:

• competition for search terms
• if a trusted external organisation provides this content already

We have not tried to do this yet because search engine results and ‘competition’ research is a lot of work. We do not have capacity to do it.

We’d love to hear from you if you have found a simple way of doing this!

Create or improve?

So far, we’ve used this approach for new content. Is it more important to write a new page or improve existing content? Should we prioritise work on new and both types of content work together? We haven’t worked out how to do this yet.

Let us know what you think!