2019 has not been a vintage one for physical retail. By the end of year, it’s predicted that more than 8,600 stores will have closed across the United States. The UK has also suffered, and with the recent announcement that Mothercare will be shutting its doors at the cost of 2,800 jobs just before Christmas, there doesn’t appear to be much holiday respite.
While many in the sector have been quick to point blame at the likes of Amazon, read around and you’ll find any number of well-versed arguments that we’re seeing is really just the full expression of latent problems that have been plaguing the sector for years.
A glance to the retail commentariat offers a variety of reasons in turn – Steve Dennis has recently pointed to the ‘unrelenting collapse of the middle‘ as an extension of the death knoll of ‘boring retail’. Richard Kestenbaum has suggested that consumers are increasingly driven to shop with retailers who we can identify with at at a personal level, a sentiment echoed by friend of the blog Andrew Busby in a recent entry to his Forbes column.
In-spite of the uncertainty and upheaval, not everyone is suffering (remember #not-all-retailers team). On this blog we’ve covered exciting examples of innovation on both sides of the Atlantic, while the runaway success of Primark shows that even the flagging fast-fashion sector seems to have life to it yet.
Introducing Non-Response Bias
Perhaps another explanation for the apparent slow-drift into retail-redundancy can be explained by a phenomenon more commonly associated with the world of surveys (but which is highly applicable to customer feedback) – non-response bias.
The term Non-Response Bias (NRB) has its origins in the world of elections, studies and polls and describes the event in which the results of a study are skewed by a disproportionate sample base. If 80% of respondents to your poll, come from a proportion that represents just 30% of your total audience, it’s likely that things are going to end up a little skew-whiff. Indeed, the potential of non-response bias to have a truly significant impact played out on the world stage in dramatic fashion during the last US election cycle.
HRC vs. DT – NRB in effect?
Cast your mind back to the 7th November 2016. Hillary Rodham Clinton is on the precipice of defeating Donald Trump to become the first female President in the history of the United States. Time Magazine has just published an article in which no less than seven highly-esteemed polling organisations predict a Clinton majority. Celebrations are already being planned. Yet as history was soon to reveal, the experts in this case had it wrong. Very wrong.
Where the pollsters fell down was in assuming their data set truly represented the national opinion. They failed to consider that a large (and influential) portion of the public simply hadn’t responded to their pre-election polls at all. They preferred to cast a different kind of vote – straight to the ballot box.
The Connection Between NRB and Retail
While the political landscape of 2016 might not immediately present itself as a direct analogy for the state of contemporary retail, dig in a little and the parallels start to reveal themselves. Current estimates suggest that in the last 20 years, the median response rate for feedback surveys for all businesses has dropped from around 20% to just 5%. At the same time, the rate by which customers are more likely to ‘vote with their feet’ is higher than ever.
96% unhappy customers won’t complain; 91% of those customers will simply leave and never come back.
While ‘putting the customer at the centre of everything you do’ has become something of a retail cliche, what do when a significant proportion of your audience is actively avoiding your attention?
Engaging the ‘Silent Majority’
The truth of it is that retailers and political pollsters both struggle with a common issue: a large majority of people today simply don’t have the time or inclination to engage with you. The tendency of ‘silent majority’ is to simply walk away, and this poses a problem for any business that wants to hear what their customer base thinks.
There’s only so many give-aways or gift vouchers you can offer before you run the risk of skewing a sample. And unless your customer data is coming in thick and fast, it’s very hard to effectively ‘operationalize’ that data in so far as using it as a tool to help with your daily decision-making processes.
Awareness is Half of the Battle
The good news is that a number of providers in the CX space are increasingly aware of the issues inherent with traditional feedback methodologies. By working with a provider that is transparent about these issues – and pro-active in terms of providing ways to counteract them – there’s an opportunity to turn non-response bias into a strategic advantage.
Here’s a selection of points to consider if you’re looking to build an effective customer feedback program:
1. Think about where your data is coming from
When looking for a partner vendors to support your CX efforts, consider where the data they’re using to support their recommendations comes from. How large is the audience being polled? Are you receiving enough responses to be actionable on a daily or monthly basis? Is your data at risk of being unrepresentative?
2. Consider whether you’re doing a brand or store-level analysis.
While many CX platforms are very good at focusing on feedback that allows for a ‘brand level’ analysis, building a data set that can be used at a store-by-store level isn’t always as easy. Look for providers that are able to give you both a brand overview as well as more granular store level drivers of behaviour.
3. Is your question set leading?
It’s important to remember the potential for ’emotive’ questions to lead to skewed data sets. While it’s invaluable to be able to uncover emotional drivers for customer behaviour, remember the lessons of ‘non-response bias’. Are you capturing feedback that enables you to understand the relationship between customer spend and sentiment? If not – how could you be doing that better?
Next time you’re considering whether your own dataset is fit for purpose, think back to those pollsters and the night of the 7th November. If something doesn’t feel quite right, perhaps it’s time to ask, am I missing the woods for the trees?
TruRating works with retailers to collect representative feedback at scale. By making feedback easy, we’re democratising insight for consumers and businesses alike, bringing the life back to ‘customer-centric’ retail.
Contact us today, if you would like to discuss how we may be able to help your business too.