Six steps to a successful content transformation

While we're enthusiastic about ditching old design, many stakeholders want to hang onto irrelevant content for grim death.

If you have a mature content strategy process you'll already be managing reviews as part of the content life cycle. If you don't, your stakeholders are likely to be more attached to the existing content.

Breaking this attachment is the key to a successful content transformation. That's done by focusing on meeting clearly defined goals using a clearly defined process.

It's also important to set the scope: existing content only. You're not trying to fill any gaps, although you will identify them.

You should start your transformation as soon as you can. Get ahead of any website redevelopment, so you give yourself the maximum amount of time.

But before you start, you need to get stakeholders on board and have an up-to-date content strategy.

A content transformation is a people project.

Your stakeholders are integral to your success. It's an A-grade cliche, but you must sell the destination (the goals) and take stakeholders on the journey (the process).

We've found the most effective way to do this is to set up a stakeholders working group and meet regularly. Such a forum is a great opportunity to:

• share progress

• discuss concerns

• improve the process through constructive feedback

• identify who is responsible for each item of content.

A content strategy provides the criteria for your content audits. You cannot assess the relevancy of an item if you haven't agreed on your key audiences, themes, and messages.

If you don't have a content strategy, you must create one before embarking on the transformation. The good news is it doesn't need to be a hefty tome. A 1-page strategy is sufficient for a content transformation. We've got sage advice on how to create a 1-page content strategy.

A content transformation has 3 goals:

1. a collection of manageable, relevant, high-quality content

2. a revised and tested information architecture (ontology, taxonomies)

3. a content plan (for post-transformation activities).

Content often loses relevancy and quality because there's too much of it to manage. While your transformed collection will only contain relevant content, it must be a manageable size. Otherwise, you'll soon be back where you started.

What's the right size? That depends on resources, the proactiveness of content owners and how well you can automate ongoing relevancy checking and archiving.

You have your stakeholder working group, your relevancy and quality criteria, you know who is responsible for each item of content. You're ready to start.

Step 1: Run a relevancy audit

The relevancy audit removes irrelevant content from your existing corpus.

Subject matter experts assess each item against the relevancy criteria to determine its relevancy. While you can use a rating system, a straight yes / no is better.

Be brutal. Resist the urge to default to "not sure". Remember, you're aiming for a manageable body of content. It's a lot easier to add content post-transformation than remove it.

Analytics can help with the decision. Low traffic may indicate that your audience doesn't think the item is relevant.

A relevancy audit report summarises the outcome of the audit:

• the volume of content before and after the audit

• the rankings of the individual items

• which items you are discarding.

Now you have a body of relevant content, you can identify content that needs remediating.

Step 2: Run a quality audit

The quality audit identifies content that doesn't meet quality standards:

• Currency and accuracy - assessed by a subject matter expert.

• Readability - assessed by a content specialist.

Currency and accuracy are binary: the content is or isn't current and accurate. Record this separately.

Readability is usually a score. If you're manually assessing content, then clearly define the scale. For example, if you're using a 1 to 5 scale, describe the characteristics of each rank. More sophisticated approaches will use automated tools that check a range of characteristics to generate a ranking or score.

What's important is that you have a simple method to help prioritise remediation.

The quality audit report summarises the audit by:

• identifying the number of items in each ranking group

• estimating the size of the remediation task

• listing common problems and issues

• making recommendations for remediation.

Step 3: Remediate content

Before you start, you must decide if you can remediate all the content in the available timeframe.

If you cannot, then prioritise the quality criteria or the content:

• Currency and accuracy only - when you have a lot of out-of-date or inaccurate content, you only fix this.

• Most valuable content only - if you've ranked relevancy, you ensure remediate by priority.

• Poor content only - when most of your content doesn't meet a minimum quality standard (for example, 3 on a 1-5 scale), you remediate the lowest scoring content first.

Content that is not current and accurate must always be remediated (or not published) regardless of how you prioritise.

It's always a good idea to walk stakeholders through examples of remediated content, especially if you are making significant alterations or restructuring. You can explain the process and the rationale, address their concerns and identify sources of possible resistance.

Step 4: Model the topics

It is better to create a new information architecture (IA) from your improved, relevant corpus than updating an existing IA. Using your relevant corpus means that you are creating topics that already have content. It's also easier to spot missing topics when doing a gap analysis.

Unless your corpus is small, the best way to determine topics is through modelling techniques, such as community detection. This groups content based on common keywords, those keywords forming your topics. Some techniques even suggest a hierarchy, effectively drafting the IA for you.

Check that the draft IA correlates with the content strategy's key themes and audiences.

Step 5: Test and socialise the information architecture

Finalising the IA is an iterative process of testing, using methodologies such as tree testing, and socialising to get feedback from your stakeholders.

Verify all feedback against the content strategy before actioning it. This is another example of where you must set stakeholders' expectations in advance.

When the testing and socialising is complete, you can confirm the topic assignment for each item in your corpus.

The IA report:

• describes the structure

• explains the rationale behind it

• provides analytics on the number of items per topic.

Step 6: Update the content plan

The final step in the transformation is to update the content plan with how you will:

• fill any gaps in the IA (topics added to the IA that don't have content)

• complete the quality improvement.

The content plan becomes the 90-day and 1-year goals in your content strategy.

Once you have completed the transformation, you'll have:

• a manageable, relevant corpus of improved quality

• a relevant information architecture

• an audit trail of the actions taken to produce the corpus and the IA

• a plan for ongoing work to improve the content and fill gaps.

More importantly, you'll have built relationships with, and ideally educated, all the key stakeholders. This will dramatically improve the chances of maintaining the relevancy and quality of the content.

Almost certainly, this is just the beginning of a continual improvement process for your content.

Contact us to transform your content today.