Cortex

Why automation fails: Q&A with Cortex’s Jonathan Hobday

Megan Wright bw
Posted: 10/06/2017

As decision makers face waves of competing information and mounting fears of robots, could the automation industry actually be headed in the wrong direction?

jonathan hobday hero

From the humble age of 11, when he started writing software, Jonathan Hobday’s enthusiasm for technology has been the driving force of his career. A self-described “part-time intellectual,” he has spent years perfecting his craft as entrepreneur, leader, enterprise tactician and cybernetics expert.

It was the realization on completion of his studies that “the real world was lagging far behind,” that initially drove Hobday down the entrepreneurial path. And he hasn’t looked back.

Today, in his role as sales director at Cortex, Hobday’s unique technical background and sales experience stand him in good stead as subject matter expert and automation innovator.

This month, as we welcome Hobday as a member of the Artificial Intelligence and Intelligent Automation Advisory Board, we sat down with him to discuss why automation fails, the biggest challenges facing the industry right now and how he plans to overcome them…

AiiA.net (AiiA):  While automation is still relatively new to most industries, you’ve been doing this long enough to have seen just about everything. What is the number one mistake you see people make time and again when it comes to automating?     

Jonathan Hobday (JH):  They choose the technology. Most organizations go to market and say, “We need the technology,” and that shows a fundamental misunderstanding of automation. The company should have first understood automation.

Automation is a state of operations; it’s not a thing you do.

If you think you can go and buy a group of consultants and say “automate,” then that’s a failure. It needs to be an effective part of your operation because it’s your operations that are being automated.

AiiA: What then should the decision making journey look like for an organization contemplating automation? What are the key things to consider?

JH: As an organization, you’ve either got an operation that is automated or it is not automated. It’s not the technology that makes it automated. Once you have that clear, you can then start to understand what that transition will mean for your organization—it’s what we call automation transformation.

When you move any kind of operation into an automated state, it will change the way that operation works. You need to understand how it will change the roles of people that operate that process, how it will change the way your company experiences that process, how it will change your capacity to run that process, and how it will change your relationship with your partners and your employees.

Read more: Automation is 10 years overdue according to Citi's Kelly Switt

Once you’ve understood all those changes, once you’ve defined your company objectives, and once you’ve determined if those changes align with your company objectives, then you automate.

Of course there are those who come to automation the other way around too—by understanding their objectives first and working backwards from there. For instance, are they looking to increase scale? Are they looking to reduce cost? Are they looking for more market traction? Are they looking for a combination of several things—and most people are.

AiiA: Once you’ve made the decision to automate, how can you ensure your organization is destined to succeed? Can you give an example? 

JH: We worked with an insurance company that outsources claims processing to third party providers. One of their service providers was providing a higher quality service than the other two, however they couldn’t meet their service level agreement (SLA).

When they looked in to why they saw they had too many people doing too many low value tasks in a chain of service and they were not able to shift that work. If they could, it would change the cost of delivery, increase capacity, increase velocity and reduce time to process. So they made the decision to automate.

Our first challenge was facing the team that was doing the processing. One of the key factors that leads to failure is not engaging the subject matter experts (SMEs). Ironically, the people that are usually presented as the SMEs are often the senior managers and unfortunately we have a rude truth to tell them—that they’re not the subject matter experts.

The SMEs are actually the people that have been doing the actual job, hands on, who have the greatest experience. It might seem fairly obvious but it’s not always a comfortable truth.

One sure road to failure is that senior executives will focus on the way things should be, not the way they actually are.

With the insurance company, we had to get the SMEs on board and the biggest challenge was that they thought automating was something that threatened their jobs. Our first job was to convince them automation was not a threat, and part of getting them on side was to make it absolutely clear to them what their role would look like in the future.

In this scenario it was simple: the division was loosing money, failing SLA, likely to lose their contract and, if it kept losing money, the company was likely to shut it down. But automation would triple the capacity of the team and enable them to take on new clients.

The company didn’t have a quality issue—they were just too slow and couldn’t handle the quantity. Net result was that we automated the environment following exactly what they do and that was how we maintained the quality.

teamwork hero
Photo by Štefan Štefančík on Unsplash

Interestingly, in most environments where you’re absolutely doing the right thing, you may not be following your standardized process. So it may become a diversion if you go and look at your documented standardized processes when it comes time to automate, because they will often be a sanitized, simplified version of a process that is actually happening.

On the other hand, when you do have a quality issue, you look at it the other way around—what should be happening.

It all comes down to the reality of how your business works, because automation needs to embrace that reality. Otherwise you’ll constantly have intervention which is the enemy of automation.

AiiA: Tell me about the relationship between automation and workplace culture.

JH: Now that’s a difficult one—and it’s not helped by people saying that automation is going to put people out of work.

We have no experience of our industry and of automation actually putting people out of work. Most people, as a result of the automated environment, have gone on to do far more productive and better paid working jobs.

For example, I recently spoke to a technical architect for one of our major clients—the operations he used to perform are now automated. What the original manager who had worked with us had said was, “I have too many highly paid people doing too many low level tasks”.

So this architect had started his job on the frontline doing 80 per cent operations and 20 per cent creative work. And now, thanks to automation, he can concentrate on the 20 per cent, which is actually more interesting.

Likewise, all the staff at this company are now doing jobs that are paying significantly higher salaries than before, that they enjoy. They no longer end up working at 11 o’clock on a Friday night trying to solve a problem on the frontline.

Read more: Chatbots are the next step in enterprise evolution

You get machines to work late. You get machines to do all the low value stuff and you move the people up to where they can deliver value. So the quality of life goes up with automation.

AiiA: How do you think enterprise stands to benefit from the adoption of automation technology?

JH: It’s going to change the world—just not in the way that the likes of Bill Gate and Elon Musk would have us think. Quite ironically Musk is not eradicating jobs; he has developed jobs for thousands of staff who are infinitely more valuable in what they do.

It’s not just enterprise though, I find it interesting  talking to taxi drivers because I think that is one of the industries that is going to go to automation. Most of them do it because they have to, most of them dread going out on Friday nights, picking up drunken people, getting them home. What we need to do is elevate these people.

And it’s not surprising, the number of taxi drivers I’ve met around the world—in Paris, in Utrecht, in Amsterdam, in Frankfurt, in Dusseldorf—who are actually studying degrees while they’re driving because they want to get a different job.

taxi hero

Photo by Peter Kasprzyk on Unsplash

We need to create a world where those opportunities are available to those people—and automation will do that by lowering costs, just like the industrial revolution lowered the cost of clothing to allow people to have more than two or three sets of clothing.

Today I’ve got a full wardrobe because I can afford the unit cost. Can you imagine how cheap a taxi would be if you didn’t have to pay for a driver? I would give up having my own car.

AiiA: Is the automation industry going in the right direction?

JH: I would argue no. It’s not going in the right direction because the right direction is hard.

The direction that the majority of the industry is going in is a direction that we call task-focussed automation. They think that the job of automation is to help people to do their current job.

I don’t think that the role of automation is to help people do their current job—it’s to transform their current job.          

I recently had a conversation with the global CIO of a very large financial services organization. To date he’s run four automation projects and, in his mind, all of them have failed. Today all of them are running, all of them do what they were designed to do, but according to him they all failed. In his mind, the objective was to reduce cost base but they haven’t done that. In fact, his cost base is still exactly the same. It’s a zero sum game, so he’s decided not to automate any more.

Someone reminded me when I told them this story, of the definition of insanity: It’s doing the same thing over and over again and expecting a different result. And I hear exactly the same story again and again.

AiiA: What do you think it will take to get the industry going in the right direction?

JH: It’s about joining it up. The manufacturing industry has been walking down this path for a long time and now we need people in other industries to step up.

We need to shift perceptions which limit people. The general public need to get their heads out of the idea that “I need to do manual labour,” “I’m a train driver, I’ve got to drive a train,” or “I’m a taxi driver so I’ve got to drive a taxi”. Imagine owning a fleet of 30 automated taxis and you have the ability to manage them all, not just drive one. 

But their mind is not thinking like that. And I think that’s the job of what we’re doing now. We need to get people thinking like that rather than being reluctant and fearful. The question of “What am I going to do?”—that’s not a scary or a bad question—that’s a good question.

Megan Wright bw
Posted: 10/06/2017

OUR BENEFACTORS