Oct 1, 2024
54 Views
Comments Off on An evidence-based approach to content reviews
0 0

An evidence-based approach to content reviews

Written by

How we use evidence to make sure we’re updating the right content in the right way for our users.

In our small content design team, our current priority is making sure the content we already have is the best it can be for our users. That’s not to say that we’re not publishing anything new. But we ask a lot of questions before we make a new piece of content.

Parent Talk, Action for Children’s online parenting advice service, launched in 2020 and has supported over 1.8 million families. As the service has grown and we’ve learnt more about the needs of our users, we’ve added more and more content to meet their needs. We now have around 240 pages of advice on a range of topics and a team of 2 senior content designers maintaining that content. We support parents and carers on a wide range of issues and it’s vital we regularly check our advice to ensure it:

is based on evidence of user needprovides accurate and up to date help and support

We look for evidence, insight, and feedback when we’re deciding what to review and what needs changing.

Illustration credit: Aira Borja from sparklestroke

How we prioritise content for review

Each time we publish or update a piece of content we set a review date and record it in our content library spreadsheet. Review dates are set based on the risk level of the topic — a scale we’ve agreed with our subject matter experts. So for example, an article that advises on child safety will need to be reviewed more frequently than a piece about activity ideas. Alongside this, we use data and evidence to determine where the most urgent updates are needed.

We prioritise content that meets one or more of the following criteria.

High traffic content with low user satisfaction

We use the tool Askem, to collect feedback on how helpful our users find our content. Users can vote yes or no to tell us if an article was helpful and provide additional feedback. This gives each article a satisfaction score out of 10 based on how many people voted yes or no. We use this data to prioritise reviewing content that:

is frequently viewed but the satisfaction score is below our averagehas shown a considerable drop in user satisfactionhas a particularly low satisfaction score or a piece of feedback that needs urgent attention, regardless of the article traffic

High risk and topical content

We focus more review time on themes that we have identified as high risk. These articles also go through additional safeguarding and subject matter expert checks. Sometimes we need to do reactive reviews. This could be because someone internally or externally has highlighted some advice that needs to be changed, or it could be because the subject is topical and we want to make sure our advice is still current.

Topics where we’ve lost search ranking

We recently commissioned an audit of Parent Talk’s search rankings. The freelancer we worked with helped us identify key terms where we’ve lost ranking since Google launched its Helpful Content Update last year. The criteria for this algorithm closely match our content principles — so by improving the content identified in this audit, we’re ensuring the people who need it can find it and that the content is as helpful as possible for them.

Content that’s out of style

We regularly review our style guide to make sure the guidance is in line with best practice for accessibility and findability. Some of our content pre-dates our latest style guide updates, so there are some inconsistencies in style across the site.

In a recent content audit, we looked at the wordcount and Hemingway readability scores of all our articles. We aim for a reading age of grade 4 or 5, so we want to simplify content with a higher reading age. Jack Garfinkel has written about the merits of writing to reading ages.

When we cross referenced this data with satisfaction and engagement scores, we also found that found that our shorter articles were less helpful. We’ve identified articles that are too short or have a high reading age as priority for review.

Related content in the user journey

We don’t look at a single article in isolation. When we’ve identified content that needs to be reviewed based on the above criteria, we also look at what else we have on the topic. We map out the articles to identify where there are:

topics that overlapgaps in informationpages that can be combined to cover a topic in more depthImage: screenshot of a content review mural. The round sticky notes show articles with linking themes related to child contact arrangements and/or coercive control. The surrounding notes show the article satisfaction scores, traffic, feedback and what the articles cover.

How we use data to improve content

When we review and update content, we use insights from our existing users and potential users to improve our content.

Article feedback on Askem

We review the feedback on our articles with our subject matter experts. This feedback gives us a good idea of what is and isn’t working but we need to make sure any suggested changes are correct and reflect the needs of other parents and not just one or two.

Themes from chats

When we review content on a particular topic, our parenting coaches provide us with data on the themes our service users are telling us they need support with. We can then review and rewrite the article with the coaches to make sure those needs are met.

Image: Pie chart showing data from chats that have been tagged with behaviour related themes.

Keyword data from Google search console

We use Google search console to understand the search terms that users are already using to find a piece of content. We need to ensure that any changes to the article continue to serve the needs identified in the search terms. So we’re careful not to remove content relating to those terms but instead look at how we can expand and improve on it.

Search volume research and social listening

The data from Askem, our chat themes and search console give us a good picture of what our existing users need from our content. But that only tells us about the parents and carers who have been able to find us. To get a fuller picture of the needs of potential users of our website, we want to understand what parents and carers are searching for or talking about. We look at:

search insight tools — Semrush, Answer the Public and Google’s ‘people also ask’ to find out what people are searching for around a topicforums like Mumsnet and social media channels to understand the issues parents and carers are talking about elsewhere and the language they use

We use all these insights to find ways to update and improve our content, then we rewrite it with our subject matter experts. The updated content goes through the appropriate sign off process for the topic risk level and a review date is set for when we need to revisit the content again.

Nikki Jones is a Senior Content Designer at Action for Children’s Parent Talk service.

Read more service design and content design blogs from Action for Children.

An evidence-based approach to content reviews was originally published in UX Collective on Medium, where people are continuing the conversation by highlighting and responding to this story.

Article Categories:
Technology

Comments are closed.