How to win customers & prove ROI with the IoT
Using live data to personalize content
A brand like McDonald’s might create an up-sell offer because they want customers who buy every day value items to buy more premium items. So when a new product range is launched they may target a promotion to customers who’ve already expressed an interest in similar products; marketing a premium Clubhouse burger to someone who usually buys a Quarter Pounder.
We can incentivize people to upgrade to that more premium burger with a free drink, and we can base that free drink on the weather. So we might offer a cold drink in hot weather, or a hot drink when it’s cooler.
What we’re doing is combining buying preference with local weather to create a more personalized message that resonates with the customer and makes it more likely for them to convert — which will actually start to increase transaction value across the board.
Live data also helps you optimize any discounting budget you may have, because there are periods of low foot traffic you can correlate to bad weather. If you are going to offer discounts, then being able to target those discounts at times when you know foot traffic will be low is a better use of that budget than handing out discounts when people are coming in anyway.
Reaching out to customers on mobile
In a QSR, which is high frequency / low transaction value — $10 a few times a month — engaging people frequently is OK, especially if you’re providing them with specific value. However there are different levels of engagement — from simple badge notification on the app to indicate new offers, to a push message that pops up on the screen. We try to limit those interrupting push messages to no more than one every couple of weeks.
There are appropriate ways to encourage people — for example when they walk into the store, a push message is absolutely relevant as people expect to be engaged by the brand when they enter a store. If there are specials on products or particular offers you want to highlight when people walk in, you can use a beacon to do that. It’s OK to do that every time, as long as you’re providing value. If it’s just something you’re pushing to people in a broadcast manner, then it really is important to limit that.
There are also certain things you need to build in to your platform to accommodate blackout periods — which you need to do across different time zones — to make sure you’re not sending messages in the middle of the night; and you need to account for different countries and languages. Having a platform that can support all of those things, and that you can monitor to make sure people aren’t inundated is really important.
And it’s dependent on vertical to some extent — you’d be happy to receive a notification from a QSR every couple of weeks, but that wouldn’t hold true for a car dealership. In that kind of space you’re much more likely to move into pestering territory if you ping customers too often.
This is something some retailers are really struggling with; they’re trying to be helpful without being creepy. There are different tolerances for different verticals and unwritten rules: you don’t want to ping someone more than twice when they’re in-store — once when they come in and maybe again on the way out to remind them to pick up a dessert. Or in a big-box retailer you could ping them in a particular department that has a special they might be interested in.
There’s a fine line between useful and creepy. The people we’re reaching out to have opted in, so they’ve allowed us to contact them and trusted us to take care of them. If you push them too hard or abuse that trust they’re going to turn Bluetooth off or delete the app and you’ll lose them altogether.
Making sure your app gets used
The first step is getting them to actually download it. The biggest asset QSRs have is their real estate — the restaurants and all the foot traffic that comes through there. That’s where you want to draw attention to the app; even down to putting it on the tray mats and coffee cups. You should be providing incentives to download the app right up to the point where customers are making a purchase. You can also train staff to encourage people to download the app by reminding them of the benefits. Some chains do this very well — if a customer comes into purchase something and doesn’t have the app, they’re made aware of it and the benefits it can offer.
The first week is key
When people download the app it’s important to get them to use it within the first seven days. People who don’t use it within the first week tend to drop off, and the percentage of those people who will go on to be high value mobile customers is far less than for those people who use it immediately.
The average app loses almost 80% of its users in the first 3 days — your mission is to prevent this happening to you.
That’s why companies like Uber provide a discounted first ride or purchase — they want you to have that mobile experience to see how easy it is. And once you’re on board they know that you’ll become a long-standing mobile customer of higher value to them than the average customer. Taking a leaf from that book, in that first seven days it’s important to offer something of value that will encourage people to come in and go through the process to get that value — whether it’s a discount or early access to a new product. Give them that during the first seven days, then follow up on it.
Don’t stop marketing
You see a lot of brands put out a mobile app and forget to promote it, and then the consumer will forget to use it. Do what you can to bring back those lapsed users, but also be sure your brand doesn’t stop promoting that app — in-store or via email notifications. Keep on marketing that app as much as possible.
Once you’ve got a group of people using the app, it’s important to measure the frequency of usage — separate users into heavy, medium and light users, and to build a program that increases the frequency with highly relevant messaging and offers.
Prioritizing IoT investment
When prioritizing IoT, it’s important to be able to demonstrate ROI as early as possible. The best way to do that is by integrating data from your PoS — your transaction-level data — into a cloud environment, and then integrating data from the app into that same environment. You then have a unique way of identifying a mobile offer when it’s redeemed at the point of sale and it’s easy to split your PoS data into those transactions that included a mobile app and those that didn’t.
Then you can work out the average transaction value for those that include mobile offers vs those that didn’t, and what we typically find is that transactions that involve mobile are typically 30–40% higher value than those that don’t. The reason is that when people use mobile, they come in with what is typically an upsell offer — an incentive to drive spend by adding another product, or by buying a more premium product with higher margin. It’s actually very quick to establish — even with just a month’s data — a trend that transaction value with mobile is higher.
Once you start demonstrating that, getting investment in mobile is easier as people start to see the returns. People see mobile customers are worth more, and understand the importance of investing more to create more mobile customers. We invest more in getting people to download the app. We invest more in the engagement program, to get lapsed users coming back and using the app more often. All of which contributes to more of those high value mobile customers, which increases average transaction values right across the business.
Future-proofing your IoT
You need to choose technology and components that can easily be built out further, and that contribute to the longevity of the systems. It’s a good idea to get as much of your data as possible into a modern flexible system; your first step should be shifting your PoS data into an environment where you can also include data from the mobile app which is linked back to a device or customer ID. So even if you have legacy systems, like old PoS, look at how you can get that data into a new cloud based platform — that way you future-proof that data without necessarily having to update your entire PoS system.
When looking at hardware systems, it’s important to make sure they fit the bill too. New doesn’t necessarily mean it’s going to be entirely future proof. Look for open systems with flexible APIs, configurable databases, flexible reporting — these are critical for systems that will play nicely with others and will continue to do so into the future.
Taking beacons as an example, because we’re hearing a lot about them lately: Bluetooth as a protocol has been around for a long time, and BLE has provided a new and more affordable way to power those beacons and put them in a retail environment. The way beacons have changed — lower energy and reduced power consumption — will continue to improve over time. You’ll see advances in physical size too. You’ll get smaller beacons with bigger range and longer battery life, using the same protocol but becoming more practical to use. There is a perception that Bluetooth is a drain on the device battery, but we’ll see that improve in the phone hardware as well.
And wearables will encourage more beacon use; as people start using more wearable tech that requires Bluetooth to connect, you’ll see an increase in the number of people with Bluetooth enabled on their phone, and the use of beacons for in-store marketing will make even more sense.