(Apologies to those who aren’t Bob Dylan fans. Just to try something different, I’ve used the titles of Dylan songs as section/topic headings in this post. If you are a Dylan fan, feel free to click on the headings and that will take you to the song; and if you’re not a Dylan fan, please just humour me on this).
I’ve been a fan of Richard Bartle for quite some time now; I think I was introduced to his work by erstwhile colleague Kevin Marks a few decades ago. Bartle’s Hearts, Clubs, Diamonds, Spades classification of MUD players fascinated me, and I’m sure it was Kevin who introduced me to that paper.
Something else Bartle said really struck me when I was immersed in representing analogue things digitally; I referred to it in a post here nearly 16 years ago, and it bears repeating.
They don’t all use the same physics.
When we build digital experiences, we don’t have to replicate the physical equivalent. In some cases it may make sense to do so, but there isn’t an overarching compelling reason to do it. Doc Searls used to fantasise about having a single uber shopping trolley that he could use to visit whichever shops he wanted to go to, filling the trolley up with goods and services as he went along, criss-crossing markets and nations and timezones, then coming to the end of his shop and settling up once for the compound transaction. The “system” then deals with all the security, payment, completion and fulfilment issues, because it can. No reason why not.
Doc was saying this at least twenty years ago, probably even earlier. He’s still waiting, like Peter Thiel’s wait for his flying cars.
I’m waiting for something as well. I’m waiting for a time when it becomes easy for me to find products and services online. The arrival of search engines presaged this and built up my expectations. I was then disappointed to learn that some of the smartest people in the world preferred to make it easier for products to find me than for me to find them, that products have more privacy than people. (And those targeting tools, in the wrong hands or used for different purposes, can wreak havoc. As we’ve been finding out).
Why is this? Why is there so much complacency on the supply side, why is there an assumption that customers just won’t leave? Perhaps that complacency is based on strong historical evidence, that customer inertia is strong in many cases. So the game plan soon becomes “how can we make it harder and harder for the customer to leave?”. Walled gardens. Friction in the cancellation process. (One of my favourites is where you can sign up online but need to call or write to cancel, with the contact details and cancellation process buried deep in the gubbins of the system). Data- and history-driven lock-ins, as we’ve seen with mobile phones and with banking, the Hotel California route to customer migration.
That lock-in model will be harder and harder to sustain, as headwinds of regulation, increased competition and sheer customer power force change. The maxim of “You Ain’t Going Nowhere” has a limited lifespan. New forms of lock-in will emerge and evolve, but not with the potency of earlier variants.
Unless companies learn that the best customer retention tool is good service (at a price the customer can afford), and that the retention tool works even better when customers are “free to leave”, they’re going to face the cliff-edge model of customer attrition. Customers stay. Until they don’t have to. At which point they leave, and leave with angst, never to return.
Because they can use different “physics”, and they do, there are many routes to differentiation. Those who choose routes that attract customers through the quality of their service, make them feel comfortable by the quality of their experience will find that they stay despite being free to leave.
I don’t want to throw out phrases like “use different physics” as if it’s a solve-it-all mantra. I think it would be better if I gave some examples.
As you can see with my use of Bob Dylan songs in this post, I like music. I particularly like music made between 1962 and 1977: the edges blur and change every now and then, but that’s the core. I don’t just like listening to music, I like going to concerts. Have always done. (The less said about my playing music the better).
I like going to concerts. Which means that when services like Ticketmaster emerged, I was an early adopter. Some years later, as I logged in, I was offered the option of using a social login or my email address. And it made me think. What could Ticketmaster do if they had my “friend graph”? Well, they could tell me which of my friends were going to the same concert, and on which days. Perhaps Ticketmaster could even tell me where they were sitting. For it to work, my friends would all have to opt in to some form of information-sharing, on an informed consent basis. So all I had to do is to weigh up the pros and cons. I began using the social logins, but only where I thought I could get value that I could not get some other way, and where I thought that the value exceeded the cost.
When I have visitors at home, it has become a lot easier for me to let them use the wireless network, because there are services on my phone to recognise them as known to me.
If someone has stored my card details when I bought something (assuming I gave them permission and assuming they wanted to take the risk of storing such data) they could refund me more easily when things go wrong, a failure, cancellation or delay of some sort. Another stress buster.
I travel a lot by train. I’ve never driven a car, never had a licence to do so. Of late I’ve been using Trainline to make my bookings, and to save the tickets in the Apple Wallet. Now we’ve had a lot of train strikes recently, and one of the things that Trainline does is to check my journeys and to let me know if a train I’m booked on is delayed or cancelled. It gives me details about changes and transfers in useful ways, informing me about the platforms where I “land” and where I “take off” and the time I have to make the connection. (I’ve used the app to check which platform a train is scheduled to leave from even when I haven’t bought a ticket through the app, and the fact it lets me do that endears them to me).
Firms want us to visit them regularly, to give them our business. They want to be Cheers, where everybody knows your name. They want to be the traditional “local”, where the landlord asks “pint of the usual” as you walk in, and sometimes even starts pouring you your drink without asking. They want to be the much-loved restaurant, where the chef says things like “food’s off” or “here comes trouble” just seeing you come in.
We’re still some way from that place. The digital experience has to be different from the analogue, doing things that couldn’t be done in the analogue, removing physical constraints and replacing them with freedoms. The same information I guard jealously today I will hand over with pleasure …. when the value’s there.
In the late 1990s, as everyone strove to be the best there was in the emerging digitally transformed world, we were all learning-by-observing. Keep the interfaces simple, dispatch complexity to the back end. Make the onboarding experience a delight. Differentiate on trust. That kind of thing, observed, experimented with, iterated and then embedded.
One of the things we observed was how parcel companies led the way by exposing their operations cycle. Instead of having to staff call centres to respond to customer queries on parcel status, they could improve service by letting customers check on the status of their own parcels themselves. Customers were happy, they felt empowered and engaged; the novelty was itself of value; and, over time, the effort of building that service would reduce costs. So we did our equivalent, called it OpsTracker, and let customers query the status of their trades themselves.
As we played with those concepts more, one more trend emerged. Customers wanted to know what was happening. And if you kept them informed, they’d be happy. Even when problems arose (and problems always arise), being kept informed was key.
Twenty years later, we still haven’t solved that problem. And customer engagement suffers as a result. If you’re not informed you don’t feel part of it. If you don’t feel part of it you don’t allow much latitude for delay or failure. Kathy Sierra used to speak of that moment when the customer stops saying “your system doesn’t work” and instead says “my system doesn’t work”. I think she called it the Kool-Aid moment, it was a long time ago. But we learnt from her, helping ensure that customers felt engaged with the product and service enough to claim personal ownership.
We don’t know what the problem is, we’re investigating. We know what the problem is, but don’t have a fix as yet. We’re fixing it, but we haven’t finished. We will be finished by <some not-too-in-the-future time>.
Not hard. So why doesn’t it happen all the time? Sometimes the systems and processes aren’t in place, so it’s headless chicken syndrome. Customers have no faith that the company they’re trying to engage with is competent, and trust is lost. Hard to recover.
Sometimes the systems and processes are in place, but the information is hoarded rather than shared. Sometimes the information is shared but not with the customer. Sometimes the status is drip-fed, and changes for the worse over time. The list goes on. That leads to cheque-in-the-post syndrome. Customers have no faith in what is stated and then begin to doubt what they’re told. The replacement aircraft is on its way, and your flight will take off in four hours. The engineer is on his way, you’re on his job schedule today. The courier has been dispatched, you’ll get your parcel this afternoon.
The principle remains the same. Where’s My Stuff? Why Isn’t It Here? When Will You Fix It? That’s all the customer really wants to do “post-trade”, whatever the “trade” is.
I’ve reached a stage where, when I get told there’s a replacement plane, I find out the call sign and track the plane in. For plane read car read parcel read whatever you like. Tracking will become more important and act as a differentiator. When I can track service from A but not B, or when I can track services from both A as well as B but A is faster and more reliable in its status reporting, A will win my custom. And the purpose of business, as Drucker said, is to create a customer.
What’s happening? Where’s my stuff? When’s it being fixed?
No information no involvement. No involvement no trust. No trust no engagement. A variation of what Patrick Lencioni referred to in The Five Dysfunctions of a Team.
If you want your customers to be engaged, keep them informed. Accurately. And on a timely basis. Not broadcast but personalised.
Many years ago I remember finding this sketch, by Mitchell and Webb, really funny. Funny, in a sad sort of way. It’s still funny. Which is even sadder still.
Solving for authentication and authorisation/permissioning isn’t a trivial issue, and there’s no way I can do it justice here without making this a book-sized post. I just wanted to bookmark it here so that I come back to it in a later post.
Every human being is unique. Everything a person does is unique. Human beings are also capable of dealing with ambiguity; we may not be good at it, but we’re usually better at it than the machines we are faced with.
Which makes the whole topic of complaints very interesting. As we get better at building the ability to handle ambiguity into our systems, and we stop using phrases like “happy paths” and “edge cases” and outliers”, we’re going to start thinking differently about complaints. There will come a time when people will start paying bug bounties for the best complaints.
And yes, it’s that Gibson case of an unevenly distributed current future.
Even when a company does everything it can to attract customers and to retain them, this is not a forever phenomenon. Companies change. So do customers. There are seasons when the “business model” ( which is often a proxy for discovering values and ethics) chosen by the company suits classes of customers, and seasons when that congruence isn’t there any more. I think it was while reading Rene Mauborgne and Chan Kim’s Blue Ocean Strategy that I first understood that what made a customer”good” or “bad” was really the level of fit with the chosen business model.
When that happens, it’s important to make it a no-fault exit. A genteel parting of the ways. The company should do everything in its power to make that leaving easy and convenient. The business model may change, to make that customer attractive again. The customer may change, in terms of level of fit with the model. Whatever happens, that customer can be an advocate or a naysayer. (It’s probably the only context in which some form of NPS really makes sense to me).
Offboarding is as important as onboarding. When I’m at a restaurant for the first time, I make a point of quietly comparing how they welcome me and how they see me off. How long it takes for me to be shown to my table, for drinks orders to be taken, for food orders to be taken. If all that happens in good time, the restaurant builds a lot of leeway into my tolerance for delay when it comes to actually being served the meal. Then, when it’s time for me to leave, to pay the bill, I start a new counter. The time taken to present me with the bill and to collect payment. It has to be quick if they are to see me return.
Included in that offboarding process is the visual check that everyone enjoyed their meal; the offer of “doggie bags” or equivalent; the staffing of coat checks and lockers and whatever ; the ease with which transport is arranged or advice given. When it comes to off boarding, I Shall Be Released is more the order of the day. Otherwise It’s All Over is the refrain, and you’ve not just lost the customer, you’ve compounded the problem by ensuring they will never return.
A Hard Rain’s A-Gonna Fall.