Every retailer has, at some point, had this exact nightmare. They’ve just opened their new store or launched their new product and… not a single customer bites.
That ever elusive “new customer” has been the cause of many sleepless nights. We find ourselves needing to peer into the future, but usually struggle to even get a clear and concrete view of what’s happening right now.
We sort of know what’s working and what’s not working, but not in the level of detail we want and certainly not in the level we need.
In the early stages of a business, it’s easy to get carried away with trying to predict where your prospective customers will be and what they want. And that’s just as applicable to a retail start-up as it is to a grocery conglomerate looking to launch a new product – most people would rather try to pin down and predict an uncertain future, than thoroughly evaluate how they’re actually performing now.
So how do we do a deep dive into the nuances and intricacies of the market you’re operating in and take a look into the crystal ball?
Ultimately, it comes down to testing. Testing your assumptions, testing the performance of your products and messaging, and testing your overall business practices (don’t forget to test that your tests work!)
But there are many different types of tests, some more effective than others. How do you figure out which ones best suit your needs?
The traditional market research approach
We’re all agonisingly familiar with the traditional approach to this.
You do a big market research study before you launch, run some questionnaires and focus groups, and try to predict where the customers will come from and what they will want ahead of time. Then, see what happens.
And as events unfold, usually very differently to what you expected, you might do another big study to cover the ground that you missed. You may identify an issue that is holding back sales and run a survey or focus group around that specific issue, hoping to unlock some deeper insights that can inform your strategy going forward.
Unfortunately, more often than not, the predictions you get end up being only marginally more accurate than a magic 8 ball. And the insights are usually extremely narrow, bound by the parameters of the survey that you set out at the start.
The problem is that the approach itself is too inadequate for the current challenges that retailers face. The survey is designed on the back of conscious and unconscious preconceptions that don’t get properly challenged. While the research method itself only gets you the expressed opinions of your customers when asked – rarely does it give you a view into their actual behaviours.
It’s a painfully blunt instrument that can’t give you a live, real-time view into what’s actually happening right now in your business. And as a result it cannot give you a glimpse, let alone a prediction, of what the future holds.
You should never place all of your faith in one big piece of pre-launch market research. If you’re doing market research, you need to complement it with a proper testing infrastructure.
How to build a modern testing infrastructure
Testing is all about data. You need to capture and process as much of it as possible. What your customers are doing, which messages are landing and how all of this is affecting your margins.
Fortunately we live in a time where we have access to exponentially more data that we have ever had before in human history. And the distinction between online and in-store is increasingly fading away, with techniques increasingly working across both types of store.
There are three components that you need to build a testing infrastructure:
The ability to collect data
This means installing tags on your website; or using technologies like Internet of Things devices, smartphone tracking and loyalty cards to track in-store customer behaviour.
A data processing platform
A platform where you can gather all this data, see what the patterns are and find the relevant insights.
A testing toolkit
A list of tests that you can carry out, to see how different factors affect the results.
With these three elements, you can see, analyse and test any aspect of the customer experience to see what works and what doesn’t work – and get the results as fast as your data processing platform can handle.
And when it comes to developing your testing toolkit, there are many options. Some of the most popular tests that retailers carry out include:
Audience segmentation and analysis: breaking down your audience into groups, and exploring the unique needs of each.
Tracking the customer journey: taking the customer’s perspective to see how you can make their experience easier and more seamless
A/B testing: examining which messages work more effectively by showing each variant to different parts of your audience and seeing which has the best results
Stress testing: calculating how your store will cope with a sudden increase in traffic or footfall, such as around Christmas
Security testing: making sure that your online or offline store is secure against the latest threats
Aligning everything with your business needs
Naturally, your testing infrastructure needs to be designed around your business needs and you need to make sure you are getting enough data from the test to make it a truly applicable learning. And often you will need to adapt as those needs change and new priorities emerge for your organisation.
We worked with women’s lifestyle and smart shopping brand TBSeen on this, back when they were first breaking into this cluttered and highly competitive market place.
When you’re entering a new market, you’ll find that your priorities change rapidly as you figure out where your new customers are, what approaches work best and how to sustain your growth.
Setting up a rigorous testing infrastructure was critical for this. The structure meant we were clear on what we were trying to learn at each stage, so that TBSeen could move forward quickly in building the business.
It gave us a solid foundation that enabled us to dial up different audiences, messaging and buying approaches depending on the current KPIs. And we were able to quickly adjust to drive the growth of registrations, shopping activity or traffic, delivering what the business needs at any given time throughout the year.
Because we had that real-time perspective, supported by a rigorous testing infrastructure, we could adapt quickly and effectively. And that is what you need to survive in today’s retail market.
Lyndsey Best is director at Altair Media