How You Achieve Customer Satisfaction Despite COVID-19 Restrictions

Thoughts on restart challenges for retailers

With containment and counter measures showing effect and the corona virus’ reproduction rate dropping below the critical value of one¹, governments timidly start to relax the crisis rules. The success of the economic restart will greatly depend on retailers’ abilities to implement efficient and customer-centric process adjustments to operate within the allowed parameters. From a process consultant’s point of view, I’ve seen some epic failures, two of which I am going to analyze in this article before proposing some ideas on how retailers could deal with the current situation.

As food and everyday consumer goods are more critical than clothes, electronics, etc., supermarkets remained open throughout the crisis. This is why I observed the first well-meant but badly implemented “COVID-19 process” at a local supermarket:
To prevent infections (and probably to reassure customers), an employee was placed at the supermarket’s entrance to disinfect the shopping carts’ handles. As becomes apparent in figure 1, the supermarket obviously hadn’t thought the measure through on a process level.

Figure 1: The shopping cart disinfection failure at a local supermarket

Placing the employee at the supermarket’s entrance instead of the shopping cart house may have had it’s reasons². However, in the original set-up, this was not the ideal place for disinfecting the carts: The customers picked-up the shopping carts themselves and drove them to the supermarket’s entrance where they would be disinfected. This means, that every customer touched their shopping cart’s handles before disinfection, thereby possibly picking up germs the previous customer might have (unintentionally) left there and rendering the whole disinfection process useless.
The next time I passed the supermarket, they had seemingly realized their mistake and deployed a set of shopping carts next to the entrance (as shown in figure 2) that the employee disinfected as soon as a customer left the supermarket and brought back their cart. This way, the next customer would get a freshly disinfected shopping cart — well done!?

Figure 2: The improved yet still imperfect shopping cart disinfection process at a local supermarket

Unfortunately, the supermarket hadn’t closed off the shopping cart house, so customers unfamiliar with the new process (or simply unwilling to follow it) still picked their cart up at the shopping cart house or brought them back there after loading their shopping into their cars.
Had the supermarket closed off the shopping cart house, the second, improved version of the disinfection process would have been a simple yet effective measure. By limiting the number of shopping carts next to the entrance and requiring customers to use a shopping cart no matter how much they intended on buying, the supermarket would also have effectively implemented a Kanban system, limiting the total number of customers within the supermarket at any time — without the need to actually count customers.

The second example of a badly implemented “COVID-19 process” is more recent and aims at controlling the customer number and “flow” at Apple’s technology temples³.
On first sight, Apple has set up a fairly thought-through system as schematized in figure 3(every person displayed within the figure is an employee, the dashed lines show, where the customers would queue):

Figure 3: The queuing process failure at Hamburg’s Apple Store

There are two queues outside the actual store and two more within the store. The first two queues differentiate between customers who visit the store for repairs or service orders on their Apple products purchased on a previous occasion on the left side (Queue 1) and customers wishing to (potentially) purchase products on the right side of the entrance (Queue 2).
At the moment, there are rules in place in Germany that limit the number of allowed customers per store space, which makes letting customers wait outside inevitable for popular and highly frequented stores such as the Apple Store. So that’s quite alright.
Once allowed to enter the store, customers from Queue 1 get serviced while customers from Queue 2 are asked what exactly they are looking for and then sent to the left queue (Queue 3) or the right queue (Queue 4) within the store, depending on their answer. So the secondary queuing system differentiates between product categories, which also makes sense as not all shop assistants may be trained for all product categories.
Now what I consider the failure within the system is the point between the primary queuing system outside the store and the secondary queuing system within the store: Before Apple lets customers enter the store, they have to answer four questions aimed at identifying people’s risk of being infectious and allow Apple to measure their body temperature. Again, this may be a sensible measure. However, what is appallingly stupid about it, is that Apple happily lets people wait for about an hour in the primary queuing system outside the store before deciding whether to let them enter or not. This means two things:

1. If a customer answers one of the four questions in the “wrong” way or has an increased body temperature (which doesn’t necessarily mean they contracted the corona virus), they are made to wait in the queue in vain before being told that they cannot enter the store. This is not only inefficient, but may also lead to major frustrations and thus churning through customers.
2. Should a customer really have contracted the corona virus, he might infect other customers (who will then enter the store) while waiting in line. The risk might be limited, as Apple ensures that customers keep their distance, but all the same, the risk is definitely higher than if they just sent those people straight away.

So, as proposed in figure 4, both for commercial and health reasons, it would be much smarter if the two security people managing the queues’ ends (they are already there, no additional staff would be needed) were the ones asking the four questions and taking the customers’ temperature.

Figure 4: How to improve the queuing process set-up at the Apple Store

Most retailers are not used to having customers queue up in front of their stores. So there is a lack of experience and mistakes are bound to be made. However, we will most probably see a lot of queuing in the weeks to come and next to avoiding sheer process stupidity, retailers would be well advised to learn from industries that have a lot of experience with queuing such as amusement parks and also to be creative about it.
My general advice is:

• Reduce the queuing time as much as possible. Easier said than done, but nobody likes queuing (except maybe the British) and you want your customers to be happy when they think about their shopping experience at your store.
• Make queuing as comfortable as possible. If you have to do something you don’t like, you’d rather be comfortable while doing it, wouldn’t you agree?
• Make queuing as interesting as possible. Queuing is boring as hell! Eventually even customers who initially read stuff on their smart phone to pass the waiting time get the feeling they have “read the whole internet”.
• Make queuing as fair as possible. There’s nothing more annoying than dutifully queuing for ages only to see another customer jumping the queue — and there’s a good chance customers will unconsciously blame you as much as the queue-jumping asshole-customer!
• Try to extract value from queuing time. Or — as they say — make strengths of your weaknesses. Your customers will never again spend more time in front of your shop windows!

Here are some more specific ideas to ponder — some of them are pretty straight forward, others are a little more controversial and may not be the right measure for everyone, but controversy is always good for change:

• Try to reduce queuing time by working with appointments. Inform customers about estimated queuing time when they join the queue (and on the internet/in your brand’s app) and offer the opportunity to return for an appointment (maybe even with their favorite shopping assistant) without queuing time. Don’t eliminate the option of queuing as an alternative to appointments.
• Display waiting time estimations for the whole queue and if possible also for “sections” of the queue. Let people know what they are in for and assess beforehand, if they can spare the necessary time. They will get really frustrated if they queue up and later have to abort their “mission” because it takes too long. You run the danger of churning customers this way.
  Have you ever been to an amusement park? They always have signs telling people how long they have to wait at certain points in the queue before they will reach the ride. It is obviously easier to estimate waiting times for a roller coaster ride that always takes the same time than it is for customers wishing to browse a shop. But then people don’t always fill all the seats on a roller coaster because they want to ride with their friends so that process is not entirely deterministic either. However, on average and with growing experience, estimations tend to be quite good.
• Work with (maybe even digital) “ticketing” systems. Let people draw a number when they “queue” up and let them enter in the order of their numbers. This will automatically make queuing fair, allow people to remain in one place instead of moving with the queue (which makes seated waiting much more feasible) and if you display the last called number and the average waiting time per number, this may also allow customers to run a quick errand for something else between drawing a number and entering your store.
  A very basic and easy to implement digital version that does not require customers to download an app could be a QR code that can be scanned by customers and opens a simple website that displays a waiting number which increases with every call from a new IP address. On the other hand, requiring customers to download your brand’s app to be able to draw a number might be an excellent opportunity to boost your app’s user base.
• See if you can provide a roof and shelter against wind and rain for people waiting outside your store and provide seats if waiting times are exceptionally long to make people as comfortable as possible. You could also offer coffee or other refreshments.
• In case of very long waiting times, providing a bathroom (or at least a porta potty) and thinking of a way for people to hold their place in the line while relieving themselves might also be a good idea.
• This is a tough one as it follows a very narrow path with respect to perceived fairness. But consider providing a separate queue for VIP customers (e.g. customers that are high up in your brand’s loyalty program) or letting them jump the queue. This could make people want to join your “club”. Depending on your store’s attractiveness, you could also allow people to jump the queue by paying for it. That way the saying that time is money becomes reality and customers can choose money over time according to their individual preferences.
• Use queuing time for customer engagement. Let them estimate the number of customers that visited the store the previous day and award a prize to the customer who expressed the best guess. Oh my — doesn’t that mean that customers have to leave some means of contact information that enables you to tell them if they won? Doesn’t sound too bad now, does it? …just make sure you do it in a GDPR-compliant way…

These are just some ideas and I’m sure there are thousands more. I actually came up with all the ideas above while waiting in that queue at the Apple Store. I guess a last piece of good advice for all retailers might be to join a queue themselves to observe the process from a customer’s point of view…

[1] The reproduction rate is the average number of people an infected person spreads the virus to. So at a reproduction rate of one, the average number of infected people at any point in time remains relatively constant (small fluctuations will occur as some people recover faster than overs), while it will grow for any value above one and decrease for any value below one.

[2] The employee was also making sure that no one entered the supermarket without a shopping cart as a physical measure to ensure a minimal distance between customers.

[3] I do not know, whether Apple handles the process similarly in all its stores. My observation was made at the Apple Store in Hamburg, Germany.