Purpose of Recommendations

Recall that the two goals for any report you deliver are to facilitate learning and evoke action to use that learning. Recommendations affect that second goal. A recommendation aims to connect the research to the world where your stakeholders or audience operate.

Recommendations connect your research back to the product. This means your recommendations have to be plausible and practical. You can't recommend your stakeholders start making spaceships when they make socks.

The worst part is when your stakeholders view you as an expert on all things "experience." You're not an omnipotent being that can interview five participants and propose recommendations to change the course of the company. Your skill set lies in conducting research and interpreting data, not being a product owner with flawless product intuition.

Sadly, you can’t hack your way to more practical recommendations. The longer you spend with your stakeholders and the experience, the more actionable and implementable your recommendations will be.

But what can you do if you're new to a team, experience, company, or industry? Your recommendations will probably be very generic and likely hard to implement. Before you start worrying, did you know that research recommendations are entirely optional?

‍

Recommendations Optional

Think about this scenario:

Fictional Reporting Scenario

• You were able to get five stakeholders directly involved in the research process. A few stakeholders took notes for you during interview sessions, while others helped finalize who to interview or help with your early qualitative data analysis.
• When it came time to report, you were able to present your findings to a larger audience, thanks to your stakeholders sharing their excitement about the research. And your stakeholders offered to explain certain quotes or case studies based on their first-hand experience.
• At the end of the report, the audience asked questions and some of your stakeholders were able to expand or add richness to your answers.

If you could see this level of investment and excitement around research, it doesn’t matter if you had a section in your report titled "recommendations. Direct and thoughtful stakeholder involvement is better than research recommendations you can suggest.

Getting your stakeholders directly involved in research is better because they're the decision-makers you’re trying to help. They know what resources they have or challenges they must face if they implement a recommendation. They also know who else they must talk to and work with to act on what they're learned.

Stakeholder involvement > research recommendations.

This doesn't mean not providing recommendations. If your stakeholders ask for recommendations, then include them. But don't automatically assume that you need to craft the "perfect" recommendations. In some situations, if you are new to the company or industry, your recommendations might signal to your stakeholders that you don't know how they work or what challenges are real for them.

If your report is shared with a larger audience, then try to budget at least 5 minutes at the end of your report presentation meeting to chat with your audience about recommendations.

Based on the report, what actions or decisions can everyone take to improve the experience? What decisions are not possible at all? What recommendations can be implemented in chunks or phases? Who else needs to be involved in how the research gets used?

If your stakeholders ask or even demand recommendations, you want to craft them in a structured and logical way. If you work with digital products and services, you should consider what element any recommendation will affect.

‍

Elemental Recommendations

One of the most important books you can read as an experienced researcher is "The Elements of User Experience" by UX pioneer Jesse James Garrett. In the book, Garrett does a fantastic job breaking down the various elements that come together to form the user experience. You might've seen the diagram online, displaying the elements as individual sheets or planes stacked together.

Why mention this book when discussing recommendations? Garrett's five elements are one very effective way to help you structure and craft recommendations. When considering a recommendation, first check to see what element it'll affect.

While you should read the full "The Elements of User Experience," the table below tries to distill the main ideas, so they're readily accessible. Please note that the elements work better for digital products and services than physical ones. With services, you should still be able to recognize which elements are easy or difficult to change.

Once you know what element your recommendation will be affecting, see if it's appropriate to make that recommendation based on the type of research you conducted. For example, if you've conducted tactical research, limit your recommendations to the first three or top-most levels (the product's surface, skeleton, or structure). If you've conducted strategic research, you can make recommendations that affect the scope or strategy element.

The flowchart below can help you figure out if you should make a recommendation or not.

Be aware that making recommendations that affect a product's strategy when running short-term, tactical research can have negative effects. If you were asked to see how unusable a prototype is, and your recommendation is not to build the product at all, then be prepared for negative emotions from your audience. If you also list recommendations outside of what you or your stakeholders can influence, then be prepared to work with additional teams to implement any changes.

You can also use the elements to help rank or order the recommendations you make. Recommendations that are somewhat easy to apply (like those affecting the surface element) would be listed higher, while recommendations that affect the strategy element would be listed lower.

‍

A Recommendation Framework

All recommendations can be grouped into three buckets: adding something, removing something, and improving something already created. But regardless of what the recommendation is, you can use the recommendation framework below to make them actionable. Use the equation to get started but feel free to customize or rework it to fit your needs.

Below are some examples using this framework.

Recommendation Framework Examples

• Consider making the menu button more noticeable to lower session bounce rates and task abandons. During research, 10/12 participants had to be given a verbal prompt to check the bottom of the page to notice the menu.
• Consider reducing the number of new changes shown in the change log pop-up after account log-in, to promote the discoverability of new changes without being overwhelming. During research, a few participants (3/8) spent 2-3 minutes just reading the new changes, losing track of the task.
• Consider moving when coupons can be applied during the check-out flow, to increase the frequency of coupons being used for online purchases. In the interviews, many participants (8/10) expected the coupons to be either added after selecting products or their preferred form of shipping.

You need to provide evidence for every recommendation. Luckily, you've run an entire study collecting evidence to improve the experience. If you need, you can list recommendations in a two-column table. The first column lists your recommendations, while the second column contains the evidence for it.

Make sure not to demand that your stakeholders implement your recommendations. If you work with a designer, an overly specific design recommendation might send the wrong signal and cause frustration. The goal is to frame the recommendation around what was learned without explicitly saying, "do this action." You want to work with your stakeholders to implement changes that don't cause other headaches or issues.

You can make more actionable recommendations once you learn the technical environment, design system, and product roadmap. If you can, try to review all three when you're drafting recommendations. It can help you see where the learning can be easily applied or be impossible to do.

Always ask yourself if any recommendations you make connect what was learned from research to your stakeholder’s day-to-day, the business, and the everyday world? If not, cut the recommendation or rework it to connect the research to your stakeholders and how they work.

One effective strategy is to have a peer read the recommendations and get their feedback. Do they feel the recommendations are somewhat plausible and practical? Do they feel the evidence matches or supports the recommendation? Make changes as needed to avoid negative or confused reactions from your stakeholders and audience.

‍

"The Recommendation is More Research"

Finally, you might run into a situation where your recommendation is to continue learning and run another study. As mentioned in Collection 1, Handbook 2, questions lead to more questions. While research is a continuous process, it doesn't mean studying something forever.

Frame any additional research as the continuation, deepening, expansion, or validation of what was reported. If you’ve planned a study with multiple rounds or methods, let your stakeholders know that you’ll be collecting additional evidence, findings, and insights in the next round of research.

A good strategy when aligning on what to study is to let your stakeholders or audience know that you could recommend additional research after a study is done. See how they feel about that and what concerns they have. You want to be ready for negative reactions to more research before creating your report

Sometimes, stakeholders are happy with your report and don’t ask for anything else.

Recognize that sometimes stakeholders might feel content with what they’ve learned and choose to move on to other concerns. Don’t expect to pursue every follow-up research question or topic after every study you run. But unless you’ve first done an effective job on your report, you might never get a chance to tackle those follow-up research opportunities.

‍

Handbook Closing Thoughts

The report you create should feel approachable and vibrant. After all of your predictions, decisions, discussions, and planning, sharing or presenting a report can feel powerful. The ability to translate data into something concise, applicable, and human for your stakeholders to use will serve you well as an experience researcher.

Try different reporting formats, make them dynamic, and watch as your stakeholders get excited to learn with you. You hold the power to have your participants' stories, emotions, and desires become an integral part of how products are built where you are.