What is the role of a research professional in a research-fatigued world?
For a long time, surveys have been a primary tool used by researchers to gather data. However, now that almost every service provider you interact with has some kind of ‘voice of the customer’ program — we are bombarded with requests for feedback, and research fatigue has become increasingly pronounced. This is particularly true of younger and minority ethnic demographics who are less likely to fill in a survey than older respondents.
Surveys must also navigate engagement bias — respondents are most likely to only consist of engaged users. Like human relationships, those who are no longer interested, have most likely moved on from the relationship and show little inclination to revisit it. This is not necessarily a huge problem as long as that caveat is considered as part of your research design and analysis.
There is also a dominant narrative that human attention spans are getting shorter. I don’t personally agree with this but perhaps it’s true to say that our attention spans are increasingly ‘task-orientated’. If the task is not compelling, perhaps modern attention spans are more likely to abandon the task than they were previously.
For the reasons above, it is not unimaginable that a time will soon come where getting someone to undertake a 5–10 minute survey becomes extremely difficult, and as a result, achieving a base size that we can feel confident making decisions from becomes a rarity.
So what can we do?
A lot of the following suggestions are not claims to innovation, and in many cases, they are already being done to varying degrees within the research industry — this post is an attempt to explore these “varying degrees”, and consider the possibilities of how we can synthesise these innovations within market research and identify which skills a researcher will need to develop in order to do this.
Considering what data already exists or can be used as a viable proxy.
I recently worked on a project where we ended up with two data points around the same thing — the number of visits to a new product that was in development. One data point was self-declared via survey and one was captured via behavioural data on the platform itself. If I’d known we had the behavioural data at the beginning of the project, I wouldn’t have bothered asking it in the survey. That responsibility was on me and I should have asked better questions of my stakeholder at the beginning of the project to understand what existing data we had access to. It would have enabled me to work smarter and reduce the length of the questionnaire. Access to behavioural data isn’t always possible — but when it is, use it instead of relying on surveys.
Another type of question I get asked to include in primary research will be something along the lines of “Let’s ask what platforms our audience over 65 use to access news?” or something along those lines. There is always a temptation to put this in a survey but first it’s worth asking whether there is existing research out there that we can use as a reasonable proxy? In a lot of cases, the answer is ‘yes’.
As researchers we should be confident in using secondary research — it will often be conducted at a greater scale and at a more reliable level than if we were to do it ourselves.
Pick our moments.
Creating feedback mechanisms in places where people are already engaged is a great way to make feedback easy to submit, accurate and useful. Have a large social media following? Why not ask them a question on the platform in which they engage. Monzo recently posted on their Instagram account asking “What’s one tiny thing that annoys you most about the Monzo app?”. It got 509 responses, most of which were valid and useful data.
Consider the touchpoints which represent the best time to ask someone a question. If you want to find out how someone uses your website’s search mechanism, don’t conduct a survey by email to ask them about an experience they may not remember or misinterpret. Include a feedback mechanism so they can declare how happy they are with a search at the point in which it’s made. Most service providers now capture cancellation reasons as part of the onsite cancellation journey — why don’t we try incorporating this into other customer journeys as well?
Think of those smiley face buttons when we leave an airport toilet. People use them because they’re easy to hit at a point in time where someone is most likely to have feedback to give. We need to think of these moments for our digital journeys as well.
Communicate your findings with your respondents.
How many respondents are put off filling out surveys because they have a nagging feeling that “this data doesn’t go anywhere anyway” or “it doesn’t change anything”? And to what extent are we, as researchers, responsible for convincing them otherwise? That question is ultimately up to the individual and how much time they are willing to invest in it.
Data is often murkily and mysteriously used — and this is true within organisations, not just to those of us who sit outside them. Different departments may pick and choose metrics that work for them completely independently of others; they may decide which data they value based on existing biases and culture; they may discard data that doesn’t tell them what they want.
So ultimately, it would be great to go back to our survey respondents and say “60% of you asked us to develop x and we plan on building a solution for this in Q3” but this requires coordination and confidence between researchers, senior management, communications, and the product & tech teams. On top of that there will be privacy and compliance concerns that need to be negotiated, all of which becomes more difficult as an organisation grows in size.
And yet… despite the difficulties mentioned above, this is probably the most effective means of engaging and encouraging users to feedback. The reason being that it starts to feel like a genuine conversation rather than a shout into the void.
My advice would be to start somewhere small. Find some data that you are confident will be actioned, look to gain consensus across teams and senior management about what you’d be willing to communicate with respondents and then work with your comms team to deliver a short simple message to say what you’re doing as a result of collecting their data. Once respondents can see they’re being listened to, they are more likely to participate in future studies.
What will a researcher look like in the future?
In order to deliver on some of the methods discussed above, I envisage that the following skills will become ever more important for researchers.
There is a feeling that the line between researcher and analyst roles are likely to become more and more blurred. Though being able to navigate data platforms like Big Query, Tableau etc. would be a boon, even more important is simply understanding where behavioural data can better answer research questions and the ability to work with data confidently across different sources.
2) Having an understanding of best-in-class UX and Service Design.
To increase the number of respondents and recruit research participants, researchers will have to start to think like designers. How do I make this easy for the respondent? Is this question asked at the right time and in the right place? How can we recruit participants at a moment in which they’re most engaged?
Consider running a workshop with your Product & Tech team to discuss data points that would be useful to have and the best place to try and capture them.
3) Confident desk-researchers.
We often talk about how academia can / should be more like the private sector but we hear little about what the private sector can learn from academia. One thing that academia often does so much better than the private sector is to give proper credence to desk research. Understanding the wider context in which an organisation operates is going to become increasingly important and having someone who can draw upon multiple secondary sources across different formats will be increasingly valuable.
Two of the best research projects I’ve seen recently within my organisation have both done a great job of using desk research to place their own primary research within the wider context. They are both stronger pieces of work for it.
Perhaps not something that’s necessarily associated with researchers but vital in engaging research participants — understanding how your data is being used and actioned and the willingness of stakeholders to communicate those actions. Obviously not all data is appropriate for sharing, but where it can be, the researcher should play a key role in advocating the benefits of taking part in research to their research population by communicating what is being done with their data. This will involve talking with senior management and comms teams and looking to obtain consensus as to what they’re happy sharing with our research respondents. Where consensus is difficult to obtain, this can act as a prompt for the researcher to consistently review what data they collect, why they collect it and how it’s used.
Some of the above challenges are undoubtedly more difficult than others and may seem overwhelming in some cases and also highly dependent on other stakeholders within your organisation. However, it is worth starting small, concentrating on the controllable and moving in increments toward your goal.
I’m always interested to hear what other researchers are doing to combat research-fatigue so please post your ideas below.