Fighting the good fight with actionable user feedback

Fighting the good fight with actionable user feedback

User feedback is required if you’re going to make decisions based on data. In fact, one of the primary pillars of a data driven organization is customer insights. Unfortunately, the thing with user feedback is that it can be pretty unhelpful sometimes. If you set up your research objectives, methodology, and validation correctly you should be in a good position to gather helpful feedback from your users.

Research Objectives

Before you dive in and start asking questions take a step back and figure out what it is that you really want to learn.

• Are you looking for feedback about how this product resonates with a specific user group you’re strategically targeting?
• Maybe you’re hoping for specific feedback about a specific page that you changed?
• Validating a theory you have about a specific functionality?

Your research objectives should be guided by the strategic mission of the product – identifying how the product is related to the overall business strategy is a good start. Your feedback will likely not be actionable if it’s about something unrelated to the product MO.

Methodology

Once you’ve nailed down your research objectives you need to ask the right questions. Depending on what you want to learn you may want to use different methodologies.

For instance, qualitative research methodologies such as interviews are a great start for casting a wide net and hearing the major feedback your users have. You can ask open ended questions to allow the user to fully elaborate. Most importantly you can probe, probe, and probe some more to dig deep into the cause of the concern they’re raising.

Validation

Once you have a sense of what some of the main comments and concerns are from your qualitative research you can validate these directions with quantitative research methodology such as surveys. Surveys alone can often miss a lot of valuable feedback, but they’re cheap and quick so they do have their place.

With quant methodologies you can validate ideas to reach a level of statistical significance that allows you to move forward. A 95% significance level is standard for academia but probably not necessary for business. Your company should set the level of statistical significance that is required based off of their risk tolerance – and that will depend on a multitude of factors.

A Word of Caution

Despite all your best efforts to gather actionable user feedback some of it still likely won’t be useful. Make sure you keep an eye out for these situations.

If your user keeps talking about their very specific situation

For example, in one study targeting small business advertisers we conducted an interview with an author. She was talking exclusively about how the product should be modified for author-specific needs rather than from a small business perspective. Her answers veered so strongly towards her own specific situation that it skewed the data, and her feedback had to be thrown out so as not to favour the ‘squeakiest wheel’.

If your user keeps talking about what other people might want or need

If the user isn’t giving feedback about themselves and their input for the group, and you can’t get them to refocus with your questions, then this may be a red flag. Their opinion is probably not better than one of a subject matter expert on your internal team. You might as well just ask your product or design team.

Share itShare on FacebookEmail this to someoneShare on Google+Share on LinkedInTweet about this on Twitter