Let the end-users have the last say on the intranet content structure

An eternal battle for the intranet team is the content structure.

Every information sender and stakeholder has an opinion about where a page should be in the navigation. Often, you hear things like “The HR manager thinks this page should be a new section tab!”, “We demand that you use exactly this word as a label and it must be put directly under ‘Self service’!”, or “I have decided this must be in the ‘Governing documents’ section!”.

It shouldn’t be like this, because we (intranet managers and team members) know an intranet must be built according to the end-user’s needs and mindset—not the sender’s and stakeholder’s. But often we have quite a hard time explaining this to the rest of the parties. They just don’t seem to get the point.

***

One part of the problem is that it’s really hard for most people not to have a sender perspective. Information senders and stakeholders often know a lot about something, and then it’s difficult to imagine how it is to not know. Kind of like riding a bike.

I think we need to make people aware of this difference in mindsets by giving them an example from another subject/field where they have no expertise. My favourite right now is talking about a classic case from the municipal external homepages in Sweden (always try to find a case not relating to your own organisation):

Dagis or förskola?

In Sweden there is a constant “linguistic battle” between parents, sending their children (0–5 years) to dagis (≈daycare or nurseries), and municipal staff, wanting to call this förskola (≈preschool).

Historically dagis is a short version of daghem (literally “day home”). Daghem was the correct name of this service from 1943 until 1972. That year the government changed the name to förskola, wanting to signal the pedagogical elements of the service.

Today, 47 years later, most municipal homepage search engines have a 60/40 split between the two words, förskola being the more popular word. But a big minority of the end-users still try to find some information about dagis.

People working in this service really prefer förskola, since it’s the correct word and it also convey professionalism. But every municipal homepage in Sweden toady also must take care of the word dagis in the search engine. Otherwise four of ten end-users will not find an answer.

When you work with people and create intranet content, an example like this is often a great eye-opener.

***

Another part of the problem is that we need to be better at explaining that good structure isn’t something everyone can build and that the intranet team really has a professional content structuring process making sure the end-users decide.

I have found that showing how many steps it takes to make it right tends to be an effective way. (In essence, this is the work process for an information architect in a UX team.)

process.001.png

1. Basic model

Every intranet needs to have a set of ground rules, a content model. When you fill the intranet you always follow the decided model. More about the Region Skåne content model.

2. Needs analysis

An intranet can’t and shouldn’t contain everything. Focus must be on the things most needed. A needs analysis tells us what content the end-users want us to prioritise.
More about the Region Skåne needs analysis.

3. Card sorting

Structure must work for the end-users if we want them to find anything. Through card sorting they tell us how they would sort the content.
Donna Spencer: Card sorting: a definitive guide.
Optimalsort—online card sorting software.

4. Tree testing

In order to double-check the future structure, we make tree tests with end-users. This way we can prove they find the answers they are looking for.
NNGroup/Kathryn Whitenton: Tree Testing: Fast, Iterative Evaluation of Menu Labels and Categories.
Treejack—information architecture validation software.

5. Web analytics

After the page is launched we continually check it with our web analysis tool. Do the end-users use the page? How did they navigate in order to get to the page? Is it in the right place or should it move based on actual usage data? Should it change name or have more keywords?
Wikipedia: Web analytics.

6. End-user observation

We also conduct end-user observation and task performance tests on the intranet. This validates what we built and gives input for improvement and additional content, which we don’t “just add”. Placement of all new content is first tested with card sorting.
More about the TPI test and other surveys we have done.

***

Talking about the knowing-too-much phenomenon and the content structuring process has helped me as an intranet manager. Hopefully this blog post has given you some ideas about how to better handle opinionated information senders and stakeholders, averting some foolish placement demands.

How to handle the most hardcore information sender, he who is impervious to arguments? Bonus idea: Trough power. Make sure you have an indisputable intranet steering committee decision stating that the intranet team is the sovereign owner of the structure. No matter how skilled you are as an intranet manager, you’ll need that statement as a last resort.

More than once.

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s