Chapter 8

(chapter 7, table of contents, chapter 9)

I got back to my desk feeling desperately tired, but quite pleased with myself. Norbert was right: I loved fighting fires. But when I looked down at my computer screen and saw a little yellow While-you-were-out sticky stuck to it, all my energy disappeared.

While I was out, Craig Lally had called. He wanted me to call him back.

The note said URGENT.

Another little nip from Norbert.

I screwed up the note, dropped it in the trash and turned my mind to more important things.

I started by looking up the meaning of the word petulant. It meant what I thought it meant. Then I sorted through my emails again, and then I spent a little time with our salary reviews. At 4:50 p.m., I dropped by the conference room (War Room) and glanced through the window. I noted plenty of activity and returned to my desk without interrupting the team. Then I ran out of excuses.

I fired up the internal phone directory and searched for Lally.

There was only one Lally, and his job title was "Flow Master." What the hell is a flow master? Was Craig one of the janitorial staff?

No. Judging by his desk location—which was a corner office above mine high enough to have a genuine sea view—he was way higher up the Wyx-Fin food chain that I was. I clicked through to his Wyxcomb intranet resume. It read like a buzzword bingo card: Lean, Six Sigma, TOC, and a whole bunch of other acronyms I didn't understand. No MBA, nor anything similar, but loads of manufacturing experience and—going way back—a half dozen patents to his name.

I noted that Craig was his middle name. His full name was Alistair Craig Lally. Made sense. Who'd take someone called Ali Lally seriously?

I pulled on my headset then reluctantly dialed his number, hoping *petulantly* that he had already left for the day.

An irritatingly chirpy male voice answered the phone. "Craig here."

He sounded older than I was; in his fifties, at least, but who could tell for sure over the phone.

I gritted my teeth, wondered if maybe I should not have made this call while I was chronically jet-lagged. Then I figured it was maybe just a few seconds too late to be asking such a question. I introduced myself.

"Steve. Thank you so much for calling me back. I'm very much looking forward to working with you."

He said Norbert had briefed him about my situation earlier in the day, and since his diary was atypically empty that week, he figured it was as good a time as any for us to catch up.

I should, he said with unrestrained enthusiasm, just name the time and he'd name the place.

Not so fast.

"Before we talk, I need to know what you did at Wyxcomb Health."

He replied with clearly diminished enthusiasm and a good bit less chirp. "I helped them figure out how to make more money, more happier customers, and happier staff. That's my job."

"You outsourced them," I said flatly.

The line went quiet for a moment.

"Is that why you didn't contact me earlier in the year, when Norbert suggested?" he asked quietly. "You think I want to outsource your department?"

I held my breath for a moment before answering. "Outsourcing is the last thing I want to happen to me or my team. The last thing."

"I understand. In that case, you should know that I did not outsource them, Steve. They chose, with my assistance, to replace their hand-cranked, custom-built software, which frankly was on its last legs, with a software package. That was their choice."

Oh come on. "That's the same thing, said with nicer words. They now pay someone else to build their software and a Wyx-Fin development team is without a job."

"True-ish. And they made their choice and they're very happy with it. I advised them, but I didn't force them to do anything other than think a lot and occasionally listen to some new ideas."

"Yeah, but ..."

"But nothing, Steve." He was getting snappy. "Norbert warned me you'd find any excuse not to talk with me, that you weren't the sort to ask for help. Would you like to hear the full story before you judge me?"

My jaw tightened. He didn't pull his punches and I couldn't disobey my boss's direct instruction. I rolled my chair back from my desk, the headset cable following me, and put my feet up on the corner of my desk.

"Sure, sorry, go for it." Though truthfully, I wasn't one bit sorry.

"Gladly. Wyxcomb Health asked me to help them figure out how to deliver projects on time. They said that their biggest problem was that they were spending a huge amount of time and money finding and fixing defects at the end of each project. They said this was common across your industry. Sound familiar?

"Uh-huh. Our standard development methodology finishes with a testing phase."

"And you do more than just test during that phase?"

"Well, obviously we fix any defects we find when we are testing."

"That's what the health team said too. They said their *test phase* should actually be called the *rework phase*. They said it was an incredibly inefficient and unpredictable way to build software, but they had no idea how to work any differently. Sound familiar?"

I kept my voice neutral, unhappy that he was correct. "Yes."

"Now, Steve, I've worked most my life in manufacturing and I've never written a line of software code in my life, but when I heard about the long rework phase I suspected I could help them. I've solved that particular problem many, many times before, though not in software development.

"You thought they could get rid of their testing phase?"

"Get rid of it, or reduce its duration enormously. Have you ever worked in a factory, Steve?"

"No."

"Well, imagine for a moment you're standing in an old-fashioned car factory, somewhere in the industrial midlands, in say the 1970s or 80s, before the Japanese quality movement hit the Western world. I'm sure you've seen pictures on TV. The factory builds cars in two distinct phases. In the first phase they build the car. Then, in the second phase, they find and fix the multitude of defects created during the first phase. If it helps you could picture a bunch of serious-looking men wearing white coats and carrying rubber mallets which they use to beat dents out of the newly-built cars."

I could picture it. I'd driven one of those cars, second-hand, as a teenager. I was never any good mechanically, but my friends were. I couldn't have afforded to own the car otherwise.

Craig went on. "We call this way of working *late-inspection*, though that name is a little deceptive because late-inspection is fine provided nothing needs fixing. Japanese manufacturers figured out how to build quality into their products during phase one, skip phase two, and ship their products directly to their retails. Much faster. Much cheaper. Much better product, and with far fewer warranty problems. Most Western manufacturers learned from the Japanese and work that way nowadays.

"I had hoped to help Wyx-Health's software development team figure out how to build quality in, but we soon realized they had a problem far bigger than their development methods. They called it their Camel Problem."

The question was begging to be asked, so I asked it, "Did it have something to do with straw?"

"Indeed it did. Their software and hardware were like an old, overloaded camel nervously watching as day after day its owners loaded it with more and more straw. It ran on an outdated technology platform that hadn't been formally supported by its vendor for over six years, and the code was buggy and difficult to change without making things worse. Things were so bad, they'd stopped selling their products to new customers. They were terrified the entire system would collapse due to increased volumes."

I said I understood. Wyxcomb Health weren't the first to be crippled by an antiquated, brittle software system. The first project I'd ever managed was at Wyx-Fin Bank, where I replaced a similarly crippled check-processing system. We'd done that in house, though; we hadn't needed to outsource it.

"So they shot the camel and bought a newer, better camel instead?"

He chuckled, and I thought I heard him slap his hand on his desk. "Exactly."

I said, "Okay, Wyxcomb Health had to replace their software—I understand that. But why'd they buy new software? Why not build their own? Surely it would have been cheaper?"

"That's a fair question with a simple answer. We ran the numbers and it was indisputably more profitable for them to buy rather than build."

I shook my head and started waving my hands in the air as I talked into the headset.

"No. No way. That doesn't sound right. Short term, maybe, yes, it might look cheaper, but not in the long run."

He listened as I elaborated that, in my experience, the upfront costs of buying a software package might sometimes appear cheaper, but that Wyxcomb Health people surely couldn't have taken into account the longer term costs of ownership: the annual fees, the extortionate costs for custom work, the overhead of having to upgrade whenever the vendor issued a new release. And so on.

"Oh, it wasn't the cheaper option, Steve. It was the more profitable option. I'm not paid to save money, though that's often a nice side-effect. I'm paid to help the Wyxcomb Group make more money."

Huh? "What's the difference?"

"The difference, in this case, was millions of pounds. It would have taken twelve to twenty-four months to rewrite their existing system. It took less than three months to install the first tranche of the new package, and a month after that, Wyx-Fin retail branches started selling new Wyxcomb Health branded products. With all the extra revenue bought in, the project paid for itself within three months of launching.

He asked me if I understood. I said I did. But that didn't mean I liked it.

He said, "The Wyxcomb Health business couldn't afford to wait for the internal software development team to rewrite their system; they would have lost millions in sales. Like I said, we ran the numbers. Millions. Commercially and technically speaking, this was a good decision.

I heard what he was saying, but it still wasn't the full story. I gave it one more shot. "But buying a healthcare package limited them to selling more-or-less the same product as their competitors."

He said, "True, but in this case Health's products are commodities. No matter how they, or their competitors, dress themselves up in fancy marketing clothing, the only substantial difference between the products in this market came down to branding and scale. Wyx-Fin Health's new products sell well because we have a respected brand and a branch network spread across Europe, which has aggressively cross-sold the product to existing customers.

"And if I may pre-empt your next two objections. Wyxcomb Health pay their vendor significantly more than the internal IT team used to cost, but their fees are directly proportionate to the number of our customers using their system. We negotiated it that way. What's good for us is good for them. What's good for them is good for us. You'd be surprised how motivated they are to keep enhancing their product. The more we sell, the more money they make; the more money they make, the more they can afford to enhance their product. It's what we call a virtuous circle."

"Win-win?" I said, trotting out the old cliché.

"Yes, indeed," he said. "Plus, you'll be pleased to know that Wyx-Fin Health's internal IT team kept their jobs. They're busy learning the new package and figuring out how to migrate their existing customers to it. They like their jobs now, much more than they did a year ago. They're looking forward to ridding their lives of the camel. So, really, it's win-win-win."

He stopped talking and I found myself in the awkward position of not being able to argue with him. "That's interesting, but I still have no interest in outsourcing my team."

"I should imagine you wouldn't. Is your software also like the broken old camel? Norbert didn't seem to think it was."

I said, "No, of course not." Which wasn't quite true. Some of it was, but most wasn't.

"Good," he said. "Where and when, then, shall we meet?"

"Give me a minute." I figured he'd assume I was checking my diary, but really, I was checking my pulse.

The fight—and the petulance—had all gone out of me. Craig had dangled the bait in front of my nose. I'd gobbled it up and I quite liked the taste. I wasn't hooked, but I did want to meet him in person, to perhaps hear a little more. But, equally, I didn't want to be part of anyone's experiment. And that, without a doubt, was why Craig wanted to "help" me. He needed a guinea pig.

But that wasn't all. I just hated that Norbert had forced me to make this call. I hated being nagged. It was some stupid teenage thing, I guess, that had never left me. Whenever I was nagged, my inclination, as foolish as it was, was to do the exact opposite.

After a little negotiation, we agreed to meet for a long brunch at ten on Tuesday morning the following week. At least I'd get some bacon out of this.