Sean McKay & Kade Schemahorn

User research is proving valuable and is on the rise. That’s why so many organizations are looking for ways to scale research.

One of the most common things they look for is a research repository—somewhere to store all their research “in one place.”

It’s a reasonable part of the process to ask, “Can we just build this ourselves?” Often, the answer is, “Yes, we can!”, especially if there’s a particularly tech-savvy and ambitious member of the team willing to take point.

The trouble is, there are risks involved that may be difficult to see before you spend time and money on trying to do it yourself.

Here’s what to be aware of.

Basic solutions are achievable but time-consuming

It is fairly straightforward to simply centralize research, taking inventory of the outputs from research and storing those items in a spreadsheet or database. You may want to track many elements, but here are the core outputs to include:

  • Research plan — What, when and who?
  • Data — What did we hear and observe?
  • Insights — What did we learn?

The first hurdle is structure. You have to decide how all this information will be entered into the system so that people can find it. This requires planning.

Once you know how the information will be entered, you have to enter it. This requires training team members on the system or providing clear documentation to ensure consistency and integrity of the data being added.

Now that you have the research centralized, people can find it, right? Well, maybe.

It really depends on the quality of your search. We’ve heard from many DIY’ers that search is where they are challenged the most.

“It’s all in one place, but good luck finding anything!”

A common theme when talking with DIY practitioners

The search capabilities of many storage or wiki solutions fail to deliver effective results, prompting the user to resort back to asking around, then digging through files and folders.

A basic DIY research repository can give you the ability to centrally access research. However, it can be time-consuming to plan and execute and has a high risk of under-delivering when it comes to search.

More robust solutions get much harder very quickly

The more that people learn about the value of user research, the more demand there will be for it. Unfortunately, a basic repository solution only goes so far in helping organizations scale their research efforts. A more robust solution is needed.

Tags

The next thing you might add to your DIY repository is a way to tag things. Tagging is valuable because it allows you to see the relationships between elements associated with different studies.

For example, if two insights have the same tag, you could find both insights using that tag.

Tags might seem simple enough, but any information architect will tell you what a challenge they can be. To serve your organization well, your DIY repository is going to need a way to…

  • Select from existing tags to avoid duplicates
  • Merge tags when duplicates are created
  • Manage typos when tags get entered incorrectly
  • Categorize tags when the list of tags gets too unwieldy

Tags are powerful, but can be a challenge to implement.

Traceability

Researchers using the repository are also going to want to show the evidence for their insights, to say which pieces of data support each insight.

This adds another type of relationship into the system that has to be created, displayed, and managed by users of the system.

Analysis

If your repository is going to support any type of data analysis, the system will need to facilitate the comparison of data across participants. There are a variety of mechanisms that could be built to do this (flagging, highlighting, color-coding, grouping, sorting, etc.), but each one adds more complexity, time, and cost to the system. Many choose to keep this separate, so the relationship to the source data or evidence is fragmented across the tool chain.

With great power comes great responsibility

Maintenance and support

No matter the degree of complexity, creating a DIY research repository means signing up for some level of ongoing maintenance and support.

  • Researchers will ask for more features.
  • Stakeholders will ask for reports that you don’t yet have the data for.
  • The tools you rely on will get updated and break your well-crafted formulas.
  • Documentation will become outdated and need to be rewritten.

Change is inevitable, and DIY means dealing with it yourself.

DIY solutions are dependent on the people that build them

Even a basic DIY research repository requires vision to get off the ground. Someone needs to champion the project, understand the needs of the team, translate those needs into features, and implement those features.

A DIY research repository needs to be treated like any other product to be successful.

Projects like this often have one person who has a vision and a few others that are passionate about supporting them. When teams lose those people, it can put the fate of the system in jeopardy. (BTW – the UX job field currently has a very high turnover rate at about 23.3%…so this risk is very real.)

Suddenly, there is no one who knows how and why the data relationships were set up or how to give new team members access. People start using workarounds, and everyone is right back where they were before the repository was created.

When it comes to research repositories, DIY at your own risk

Can you build your own research repository? Yes.
Will you run into loads of challenges and limitations? Yes.

The Handrail team knows this firsthand. Being practitioners ourselves, we share an understanding of the needs that teams have in storing, accessing, and sharing user research effectively. This deep understanding and desire to solve the problem long-term is why we shifted from consulting to a product company in 2018, after many years of incubating the product concept.

We are excited be helping other practitioners be more efficient and positioning organizations to succeed with research at scale.

Have ideas for what makes a research repository great? Let us know!

Leave a Reply

Your email address will not be published. Required fields are marked *