The aggregation of marginal gains

There is another reason why details matter: the total customer experience is really just the sum of the smaller interactions. If you improve everything you do, even just by a little bit, these tiny gains add up to something big, an approach the legendary GB cycling coach David Brailsford calls ‘the aggregation of marginal gains’.²

In my experience, clients who want to improve their customer experience start by kicking off a project, and the higher their aspirations the bigger this project gets: more stakeholders, bigger changes, timelines and budgets. It starts with a grand, strategic vision and ends in tears. To start with it takes too long to agree on the scope of the project, then when it moves into a feasibility assessment they realise the whole technical infrastructure of the company needs to be replaced which doubles the workload and creates more dependencies. Requirement documents swell from a few pages to a few hundred pages. In short, it becomes an unmanageable mess. Nothing gets delivered, or if it does it’s usually so late as to be irrelevant, or so expensive as to offer no return. Have you ever looked at a company’s products or website and thought to yourself ‘Why don’t they just fix it?’ This is most likely the reason why they haven’t. They aren’t trying to fix small problems. They are trying to change the entire universe and failing.

One alternative is to look for ‘quick wins’, which is a more pragmatic approach, but most often ‘quick win’ really means ‘slow fail’. In practice it is usually something that can be bodged easily, rather than a small issue we can do a thorough job on. Quick wins make things far worse later on, and often don’t really make things better in the short term.

Clients most often succeed by delivering smaller projects. Little and often works best. You want the lowest number of the best possible people, working to a clear and achievable objective. If they are smart they can manage themselves. If they are committed you don’t need to keep an eye on them: just give them the brief and let them get on with it. Most projects would be more successful with a smaller scope, shorter timelines, smaller budgets and fewer people.

Stages and steps

In Chapter 4 we discovered how great customer experiences reflect the customer’s identity. This helps us improve the experience at a brand level. In Chapter 5 we saw that great customer experiences satisfy our higher objectives. This helps to get the experience right at a product or service level. This chapter completes the hierarchy by looking at the customer experience at an interaction level, which gives us a complete structure to work within when trying to identify areas to improve.

You can’t infuse the experience with the right brand values if you don’t know what they are upfront, and you can’t evaluate whether features or functions are useful without relating them back to the customer’s higher objectives. But both these things – the brand promise and the higher objectives – are delivered through each and every interaction the customer has with the business. To create a great customer experience we need to identify what these interactions are.

The temptation is always to charge in and start designing stuff or bouncing exciting ideas around, because this is the most fun part, but before we get into that we need to know exactly what needs to be designed. I find it easiest to do this by breaking the customer journey down into two levels that I call stages and steps. I do this for four reasons:

1. Distinguishing between broad stages made up of small steps allows us to look at the experience at different levels of detail. It’s useful to have a bigger picture to refer back to because it’s easy to get lost in the minutiae when you get into the small interactions.
2. Breaking the experience down into a sequence of small steps allows us to understand all the dependencies between different stages before it’s too late. There’s nothing worse on a project than having to re-design whole chunks because you missed a key dependency.
3. Looking at the customer experience as a series of small steps allows us to get the sequence of events in the correct order so the experience flows seamlessly from end to end.
4. It helps us identify and prioritise the problems within the experience, so we can concentrate our efforts in a way that gives real-world improvements quickly. It can also help us to keep projects to a manageable size.

Identifying stages

At this point in the process we know what the experience must be like on a brand level, and we have used Stanislavski’s mental reconnaissance techniques to identify the customer’s objectives. Starting with the super-objectives and objectives of the most important customer profile, start writing down the stages of the customer experience that are required to satisfy them.

The question you are trying to answer is ‘What does the customer need to do to satisfy their objectives?’ Referring back to our airport example, the question becomes ‘What does James need to do to get to the funeral in San Francisco on time?’

A mental re-run of my last flight gives me the following stages as a starting point:

• Plan the trip
• Find the best fare
• Buy a ticket
• Pack
• Get to the airport
• Check in
• Clear security
• Wait in departures
• Go to the gate
• Board the plane
• Fly to destination
• Disembark
• Pass immigration
• Collect luggage
• Meet friends at arrivals
• Leave airport
• Arrive at destination

We can see even from this example that there can be far more parts to the experience than we consider. My experience of air travel is that investment in the customer experience is limited to those stages in bold – fewer than half of them!

There are two approaches you can try to get started with identifying stages if you aren’t sure what they might be. One way is to think of the customer’s journey on the highest level first then work down. For example, you might start with four simple stages that apply to most businesses:

1. Discovering the brand
2. Shopping
3. Using the product or service (commonly called ‘in-life service’)
4. After-sales support

You could then break these down into smaller stages. If we were evaluating the experience for a supermarket we might take the stage shopping and split it down further. By the way, it’s always useful to give stages and steps a unique reference number in case you end up sharing them with colleagues.

SHOPPING STAGES

2.1 Orientating yourself in the shop

2.2 Browsing the products available

2.3 Asking staff for help with finding a product

2.4 Taking products to the check-out

2.5 Checking out

2.6 Taking the products to the car

Scenarios are helpful

A popular technique to help identify the stages of a customer journey is to write out a quick scenario to help guide your thinking, using the different profiles from the previous chapter as a starting point.

An example might be, ‘Mrs Miggins is doing her weekly food shop, as she does every Saturday morning.’ Or, ‘Clive is popping in to the supermarket on his way home from a long day at work to get something quick and easy for dinner that evening.’ You can do as many scenarios as you like, and add as little or as much detail as you like to help get your brain going. Scenarios are like scaffolding – they are there to help you, but they aren’t what you are designing. Don’t worry about formally covering every possible eventuality. This will all come out in the wash as you get into more detail.

Define the success criteria

It is essential to give each stage a set of success criteria to judge it by. This should be straightforward to do, since you can use the objectives you identified in the previous chapter as a starting point. I would include the business objectives for each stage in the success criteria too, so you can start to identify any conflicts between the two that may need to be resolved. The remaining chapters of the book will give you much more to work with when defining these criteria, but for now it is enough to note down what the customer’s objectives are so that they remain in the front of your mind at all times.

Success criteria not only help to keep the experience on track, but they also help to build consensus within the project team about what you are really trying to achieve. If you can’t relate a stage directly back to clearly defined objectives you need to think these through before you go any further. Success criteria are also vital for testing – if we don’t know what success looks like, we won’t know if we’ve succeeded. Here are some possible success criteria for the stage ‘Taking the products to the car’.

TAKING THE PRODUCTS TO THE CAR – CUSTOMER SUCCESS CRITERIA

• I can find my way out of the store easily
• I can find my car easily in the car-park
• I can move my goods to the car without damaging them
• I can load my shopping into the car without difficulty
• I can leave my trolley somewhere convenient for me

TAKING THE PRODUCTS TO THE CAR – BUSINESS OBJECTIVES

• Upsell more goods on the way out – coffee, newspapers, magazines, petrol
• Minimise theft or damage to property – trolleys, etc.
• Restrict car-park usage to our customers only

Even at this seemingly unimportant stage we can expose issues. Here are a few from my own experience:

• My local supermarket is in the middle of a town, and operates their car-park on a ‘pay and display’ basis to try to restrict its usage to customers. Although they refund the price of the parking at the check-out it is annoying to have to keep loose change with you whenever you want to go there. Some shoppers find they return to their car to a parking fine because they weren’t aware it was a paying car-park. This is especially bad if you have just paid the business in question for your food shopping.
• In an attempt to reduce trolley theft and incentivise customers from leaving them randomly around the car-park some shops force you to put a pound coin into the trolley to release it. This is extremely annoying.
• One supermarket near me uses giant revolving doors on the exit. This is terrible to try and navigate with a trolley, especially for families who also have a push-chair.
• It is also common to have to walk a long way to the trolley-park because the shop is more interested in maximising the number of parking spaces. The result is that trolleys just get left anywhere which can do damage to customers’ cars and requires someone from the shop to find and collect them.
• I’ve had bags split when carrying them to the car because they were too flimsy and were torn by the corner of some packaging.
• Walking a long way to the car can mean you and your shopping get soaked if it starts to rain.
• Elderly shoppers or those with small children may struggle to get their goods to the car, either because they are too heavy or they have too many other things to carry.

Each of these issues is an opportunity in disguise. Here are a few things we could consider, though I’m sure there are many more opportunities than I’ve covered:

• Customers could pay for parking using their loyalty card if they have one.
• The shop could provide tokens at the reception desk in case a customer doesn’t have change.
• Staff could be on duty at the doors with umbrellas to walk the customer to their car when it was raining. This wouldn’t have to cost anything – staff who were normally collecting trolleys could do it.
• Staff could be on hand to help take customers’ shopping to their cars. This might be further improved by asking the customer if they would like this service when they arrive at the check-out rather than afterwards to avoid any waiting around. If they were asked when they arrived at the supermarket it might encourage them to buy more if they knew they did not have to worry about carrying it back to the car.
• A large sign at the entrance of the car-park and the supermarket reminding you to bring your own bags from the car would help minimise the use of plastic bags. We often have bags, we just forget to bring them from the car.

Start at the start and end at the end

When looking at the stages of the customer experience, try to trace the experience all the way back to the beginning then right to the end. We touched on this earlier when looking at the stages involved in air travel. It’s always been my experience that businesses focus on too narrow an area and so miss opportunities.

Sticking with the supermarket theme, an excellent example of the value of broadening your horizons is online grocery shopping. Supermarkets focus on making items easy to find on the website, remembering items you shopped for previously and providing a good delivery service, but the experience actually starts way before the customer visits the website, and it ends way after the goods have been delivered. What triggers a supermarket shop? You could open the cupboard to find you’ve run out of something; you could be cooking a meal for friends at the weekend; you could have a regular routine that you go to the same supermarket every Saturday morning and buy broadly the same things every time; you could want to try something new to eat, or even be starting a new diet. Either way, whenever you buy food there is some sort of planning involved, especially if you are going to do the shop online.

This planning stage is woefully neglected. Often, to find a recipe you might want to try, you need to know upfront what that recipe is – they don’t make helpful suggestions for things to try – and integration with the rest of the shopping experience is weak. If they know you are ordering all the ingredients for a recipe from the website, why don’t they include a print out of it with the shopping when it arrives? Some online retailers do not even allow you to add the ingredients for a recipe straight to the basket.

If I know I am at home for breakfast five days this week, for dinner three days, and want to take a healthy lunch to work for four, why can’t a website present me with options for doing exactly that rather than making me trawl through all the products myself? Why not have an option to start the shop with a calendar view rather than a search box? All they are doing is trying to transpose the in-store experience onto a new medium, rather than taking advantage of the technology available on the web.

Break the stages down into steps

Once you have all the stages, you can break them down into the specific steps that are involved. For the stage checking out at a supermarket this may consist of:

STEPS WITHIN THE STAGE – CHECKING OUT

1. Find the check-out with the shortest queue
2. Load the items from the basket onto the check-out stand
3. Scan the items
4. Pack them into bags
5. Pay for the items

They may seem small, but each step is an opportunity to make improvements. The best thing about the steps and stages approach is that you can keep breaking both of them down as the project progresses. You can start big and then keep zooming in until no stone is unturned. You will never discover every step straight away, and it’s a messy, time-consuming process, but it’s worth doing and re-doing. Don’t be too concerned with working in a linear way. I often find it helps to just start anywhere then work outwards.

User profiling #3 – significant factors

When we look at the smallest steps involved in a customer journey, we find that the number of things that affect a customer’s decision making can grow. Consider our supermarket – these might include: whether the customer has a loyalty card, whether they are paying by cash or credit or debit card, whether they are using a basket or a trolley. Each of these is a significant factor in the design of the customer journey. A customer who is using a wheelchair may need help getting items from high shelves or might need a specially designed trolley, and yet these factors will not usually be picked up at the higher levels of profiling – the identity or objective level – even though they are very important. What we need to do is generate a complete list of significant factors that apply to the stage, then group them into combinations.

Start by generating the largest list possible of factors – Are they an existing customer or a new customer? Have they registered for an online service or not? Is the customer logged in or not? Are they doing these steps online or in a shop? Are they in a particular location – at home, in the office, on a train platform? This list of possible factors can grow very quickly, and you may find even more emerge as you read the rest of the chapters. Factors such as competence – is the customer a beginner, intermediate or expert? – can be of huge importance when designing an interaction. The point is to keep adding to the list and accept that you won’t know everything at the start. I worked with a client recently who charged ahead and designed a service only to find it alienated a large percentage of their customers because they missed a factor – don’t repeat their mistake.

Once you have the list of significant factors, the key is to prioritise them by looking at the most important combinations of factors. Suppose we identify three significant factors for the registration stage of an online supermarket: whether they have a loyalty card, whether they live in the delivery area, and how they are registering. These variables result in the possible combinations shown in Figure 6.1.

For example:

• Combination 1 – has loyalty card, lives within delivery area, is registering online – is an obvious one to work on since it allows the best experience for the customer. We might be able to pre-populate a large number of the registration form fields for the customer by using the data attached to their loyalty card, which saves them time. If we know their address we might be able to auto-check whether we can deliver to them, to save them having to go through this process manually too.

Figure 6.1 Possible combinations of significant factors for the registration stage of an online supermarket

• Combination 4 – no loyalty card, lives outside the delivery area, is registering online – is another good one to work on, since it represents the worst-case scenario. The customer must enter all the information manually, only to find out that we do not deliver to them. Combinations 2 and 3 re-use elements from 1 and 4, so we don’t need to worry about them so much because they are not unique.

For the stage you are working on, start with one or two combinations, and then document the steps required to meet the objectives successfully.

Specifically enter waiting times

If there is a significant wait time between one step of a journey and the next, add it to the list as its own step. If you know about these wait times there is a chance you can minimise them. If you don’t note them down, they won’t be improved.

Look at the sequencing and dependencies

Every customer journey has some dependencies between steps. If they are not actively addressed, they impact negatively on the experience. When I first started looking into online grocery shopping for a project I was working on, I found that some retailers would allow me to get all the way to checking out with a full basket of goods, to then tell me that they didn’t deliver to my area. Some took the opposite approach, forcing me to create an account before I could even have a look at what was available. Neither are great solutions.

If you are working on one specific stage of the experience, it is critical to model all the dependencies that come before that stage. If you don’t, you run the risk of designing something that doesn’t work, or having a scope that creeps further and further up-stream to incorporate the dependencies into the project.

To move the dependencies out into the open so that they can be addressed, whenever you discover one add it to a list of pre-conditions for each stage. In the example above, the dependency is that the supermarket can deliver to the address, so a pre-condition for the stage shopping is that the address can be delivered to. A far better solution for the supermarkets would be to have a call out box on the homepage that says something like, ‘First time shopping here? Enter your post code to check we deliver to you first.’ Of course, not every step of every customer experience is performed in the same sequence every time, but where there are sequential processes they need to be considered end-to-end.

Map out all the touchpoints

One of the biggest challenges for creating a great customer experience can be joining up all the touchpoints. We might use the website, a mobile app, a kiosk, a shop, or phone a call centre on our customer journey. There may be more than one touchpoint we can use to complete a given task. To address this problem, for every step you note down, you also need to write down what touchpoint the customer is using, or which ones it may be useful for them to use. I recommend documenting your as is customer experience, as well as your to be one. Your as is model will expose all of the broken joints and inconsistencies between different touchpoints which you can improve in the future.

I produce two diagrams to help with this. The first is a simple grid with the touchpoints on one axis, e.g. shop, call centre, e-mail, social media, website, mobile app, then the stages of the experience along the other axis, e.g. browse products, add product to basket, check out, specify delivery time. Then within each square note down what offerings there are, e.g. does our mobile app allow us to check out? Could we use social media to raise awareness of new products? Often the simple act of mapping these two variables together generates all sorts of new ideas that you had not considered. It also brings to the surface any obvious disconnects between the different touchpoints you have. See Figure 6.2 below for an example.

The second diagram that I find useful is a swimlane diagram that shows how the customer moves between touchpoints along their journey. You start by drawing a series of parallel horizontal lines, each of which represents a touchpoint. Then, giving each step of the journey a number, you add them onto the swimlanes in sequence. It will end up looking like a line graph, which shows how the customer moves between the different touchpoints along their journey. See Figures 6.3 and 6.4 for an example.

This kind of diagram is useful because it helps other members of the team to understand how the journey really looks. It also further helps to expose dependencies. When you start to look at allowing customers to perform tasks on various touchpoints the dependencies get much more complex.

To show these two kinds of diagram in action, I have mapped the first part of my train journey to London in Figure 6.2, which shows some of the stages I complete when I travel to London on the train and the possible touchpoints I can use. Here you can see that to pay for my parking at the car-park, I must use a telephone system where I enter my vehicle registration and credit card details, or pay with cash at a kiosk in the car-park. I then go on to use another kiosk or the ticket office to buy my train ticket, and can then use another screen to check which platform my train departs from.

The swimlane diagram in Figure 6.3 shows my typical journey in the morning and what touchpoints each interaction uses. I start by checking the train times the night before, using the mobile app. Before I leave the house I use the app again to see if there are any delays on the line. I then drive to the station, park in the car-park and pay for my ticket using their telephone parking system. I then use the kiosk at the station to buy my ticket, then check the departure screen for which platform I need.

Looking at this I can see two improvements straight away. First of all, I am using two separate touchpoints both of which require a credit/debit card transaction for two different tasks: paying for parking, and buying my rail ticket. The first improvement I would make here would be to allow the customer to pay for parking when buying the rail ticket using the ticket kiosk. This would reduce the number of transactions the customer needs to complete, and make them less likely to forget to pay for parking using the phone system if they are in a rush. The second improvement I would make is that while the customer is waiting for the ticket to print at the kiosk, it could show on the screen the next trains that matched the ticket they had bought and what platform they were on. This would further simplify the customer journey, saving the customer time and effort. The resulting customer journey is shown below in Figure 6.4.

Identity, objectives, stages and steps

The first three chapters of this text should have helped you to create a complete two-dimensional breakdown of the customer experience: looking at the experience from top to bottom we should now know how the brand experience must relate to the customer’s identity first and foremost, how the product or service experience must relate to their super-objectives, and on an interaction level, what tasks the customer must undertake as part of the experience, broken down first into stages, then into steps.

Looking at the experience chronologically, we now also know where the experience really begins and where it ends. This may be cyclical or linear, but we should now know every event, and the sequence in which they take place. Now we’ve got the flesh and bones of the experience, the rest of the book shows you how to bring it all to life by making each of these interactions as good as it can be. Before we move on though, here are four beautiful sets of details to inspire you.

Summary

• We care about details because they show that the business cares about us.
• If every detail is right, the overall experience will be right – lots of small gains create one large gain.
• Breaking the experience down into steps and stages reveals the details.
• Every stage of the experience needs to have documented success criteria.
• Tracing the experience right back to the start, then right down to the end reveals opportunities to innovate.
• Identifying the dependencies between stages as early as possible reduces risk on the project.
• Breaking the big stages down into small steps reveals opportunities for improvements.
• The steps that a customer takes depends on the combination of significant factors that applies to them.
• Modelling how customers move between touchpoints is the key to creating seamless multi-channel experiences.