I simply don't trust them

A customer support manager once asked me how much it would cost to open the office twenty-four hours a day.

Not metaphorically. She wanted the actual number.

Security. Insurance. Electricity. Heating. Cleaning. Presumably someone to make sure that the person working at 2.17am wasn't secretly watching Netflix. She wanted the whole spreadsheet, all the numbers and the liability cover we'd need.

This was, by the way, for a company selling cloud software for customer support – you know, the kind that works anywhere in the world, which made the question particularly interesting indeed.

The software was already available twenty-four hours a day. The customers did not care whether the person answering their question was sitting in an office or at a kitchen table in their underwear. The internet, for all its faults, had solved the geographical problem rather well, and we led the way with the technology.

So I asked why they needed to be in the building overnight. She gave me an answer that has lived a long time in my brain.

"Because I simply don't trust them. They need a manager watching them."

There are moments in corporate life when somebody says something so silly that it makes you wonder whether you're really working with adults at all.

This was one of them.

Not because she was a terrible person. She wasn't. She was, in fact, sometimes pleasant. She simply had a theory about people, and the theory had eventually made its way into the property insurance and overnight security.

If people cannot be trusted unless somebody is watching them, then you need to pay a little more for the building, and pay a little more for someone to watch them.

I, of course, responded with the obvious question: "If you don't trust them, why did you hire them?"

She stared at me blankly, turned to walk away and asked for those building costs.

What interested me was what happened next. Because distrust has a remarkable talent for becoming part of the culture.

The support team were not incompetent. Quite the opposite. They knew the product, knew the customers and were very capable of making sensible decisions all on their own. They were just not allowed to make very many of them.

Everything required approval. Exceptions travelled upwards. Decisions waited. Those tiny amounts of authority were carefully kept away from the people actually having the conversation with the customer – in other words, those who needed those tiny amounts of authority to do the right thing.

The organisation had, in effect, designed itself around the possibility of the worst employee doing the worst possible thing, most of the time.

This is one of the interesting, and stranger, habits we have in business. We hire the person we hope never turns up and use them to design the experience of everyone who did.

Someone might spend too much money, so nobody can spend anything without permission. Someone might make a bad decision, so everybody needs approval. Someone might misuse their freedom, so freedom gets removed. Someone might steal a cheap notebook, so we lock the stationery cabinet.

The hypothetical idiot becomes the most important person in the room, and the whole business is designed around them. Everyone else gets the consequences.

You can see how this happens. Each individual control sounds reasonable. Nobody wakes up one morning and thinks, Today I shall build an organisation in which intelligent adults require written permission to get a new pen.

It accumulates. One sensible precaution joins another sensible precaution until you have constructed something that stops sensible people from operating.

And then we wonder why everything is slow.

The customer gets the bill

There is a more interesting problem, though, because all this caution eventually has to come out somewhere, with costs to boot.

Imagine a customer with an unusual problem. In a good interaction, the person on the other end listens, understands what is happening and makes a judgement. They might bend a rule. They might find a workaround. Or they may use the agency they have to solve the problem.

They might spend ten minutes doing something that technically belongs to somebody else because the customer doesn't care which department owns the problem. They just want the problem solved.

This is the part of work that cannot be automated because it requires judgement. And if judgement requires permission, well, things start to go wrong.

If every decision has to be approved, people eventually stop making decisions. This isn't rebellion. It is economics. Why spend your afternoon thinking carefully about doing the right thing when you know someone else will change it anyway? Why take responsibility for a problem when the safest possible response is to pass it upwards, or along the line? Why exercise initiative when initiative is the very thing that attracts scrutiny?

Eventually the person answering the phone becomes very good at one thing: finding the policy and sticking with it.

And the customer hears it immediately. "I'm afraid that's our policy." Which, in other words, means I'm not actually going to help you, and you don't matter that much to us.

The reality is that the person saying it may be just as frustrated as the person hearing it. They can see the sensible answer. The customer can see the sensible answer. Everyone can see the sensible answer. But the sensible answer is not possible right now, even though your call is important to us – it's trapped somewhere upstairs in the manager's office.

So the customer gets the policy. The employee gets the blame. And somewhere else, somebody is updating a process document. And the customer is looking for somewhere else to take their business.

This is why I find customer service policies so endlessly fascinating. A policy is often a piece of distrust that has been turned into writing, laminated and stuck to the wall.

Somebody, somewhere, has had a thought along the lines of: I'm not sure everyone will use their judgement sensibly. And so we remove the need for judgement.

Sometimes this is exactly the right thing to do. Nobody wants the person operating the nuclear reactor to improvise because they had an interesting morning and too much coffee.

But most work isn't a nuclear reactor. A surprising amount of it is simply people trying to solve ordinary problems for other people. And yet we keep designing it as though the principal danger is that somebody might enjoy themselves too much with the available authority and give away 99% of the company to the first person who calls in.

The result is often a wicked inversion. The more capable the person, the more frustrating the system becomes. They can see what needs doing but cannot do it. They understand the customer's problem but cannot solve it. They know the exception makes sense but are not allowed to make exceptions.

You have taken a person with judgement and gradually converted them into an answering machine with excellent domain knowledge. This is not particularly efficient. It is also rather expensive.

The useful thing about trust is that it doesn't require quite so much machinery. Nor paper. You hire somebody who is capable. You explain what matters and you give them enough room to use their judgement.

Then you see what happens.

Usually, they do something sensible. Sometimes they don't. This is the bit that makes trust uncomfortable. It isn't a guarantee but nine times out of ten, it works. It's a bet and the odds are good.

The alternative is also a bet, mind.

Distrust says: I will accept the enormous cost of constraining everyone in order to reduce the possibility that someone might do something foolish.

Trust says: I will accept some possibility of error in exchange for allowing capable people to actually be capable.

Neither is risk-free. They simply put the risk in different places. And the second has a rather pleasant side effect. People tend to become more useful when you give them something useful to be responsible for.

Give someone a real problem and enough authority to solve it, and they will surprise you. They may solve it better than the person who wrote the procedure. They may discover a better way of doing something. They may spot a problem before it becomes expensive. They may even begin to care more deeply about the outcome because they have some influence over it.

This is why I think discretionary effort is such an interesting thing. It isn't a switch you can flick. It is something people offer.

And people are much more inclined to offer it when the climate around them suggests that it will be useful.

The manager who wanted the twenty-four-hour office thought she was solving a staffing problem. She was actually trying to solve a problem of trust with real estate and night managers.

This is one of my favourite examples of how an apparently small belief can become extraordinarily expensive.

I don't trust them.

Becomes:

They need watching.

Which becomes:

We need someone here overnight.

Which becomes:

How much will the building cost?

And eventually, if you follow the thread far enough, you find yourself discussing insurance premiums because somebody wasn't allowed to answer a customer's question, sat their kitchen table, without supervision.

The original belief was psychological. The final invoice was costly.

This happens everywhere. A lack of trust becomes a policy. A policy becomes a process. A process becomes a queue. A queue becomes a bad experience. A bad experience creates more policy to address the bad experience. And round we go.

The person who encounters the bad experience has absolutely no idea where it came from. They just know that nobody will let the person on the phone help them.

The remarkable thing is that the reverse journey is possible too. Trust a person and you can remove a little permission. Remove a permission and you can remove a little waiting. Remove the waiting and someone can solve the problem while it is still small, when the customer is still on the phone. Solve the problem while it is small and the customer has a much better day.

A rather large amount of organisational improvement works like this. Not by adding something impressive. By removing something that shouldn't have been there in the first place.

The customer always knows

We spend a lot of money trying to improve customer experience at the point where the customer experiences it.

Scripts. Training. Surveys. Mystery shoppers. Smiling. Perhaps a poster saying CUSTOMER CENTRIC in a nice font large enough to be seen from across the street in the Crown and Anchor pub.

All of that can help. But sometimes the better question is much further upstream than this.

What have we asked the person dealing with the customer to do? What are they allowed to decide? What happens when the sensible answer isn't in the manual? Who gets to say yes? How long does that take? And, underneath all of it, what do we believe about the person we've put in that position?

Because customers are brilliant at detecting the answer. They can hear the difference between someone who is trying to help and someone who is trying not to get into trouble. You don't need a survey to measure that. It takes about ten seconds. We've all felt it.

That customer support manager didn't need a twenty-four-hour office and night manager. She needed to make a choice. Either the people she'd hired were capable of doing the work, in which case they needed enough freedom to do it, or they weren't, in which case a night shift wasn't going to solve the problem in the long run.

The building is not the problem. It was simply where the problem had eventually arrived. And that, I think, is the useful thing to remember about distrust. It travels.

You can put it into a rule, a process, a policy, a permission or a building, but you can't make it disappear.

Eventually it reaches someone. Quite often, that someone is your customer. And they have already been waiting long enough already.