Revenue: $1.06 billion
CIO: Matthew J. Dattilo
Business Colleague: Jeffrey B. Killian, Ph.D., director of global research and development operations
Business/IT Challenge: Optimizing R&D operations post-acquisition.
Upshot: Resource allocation software has brought visibility to project workflow and availability among 450 R&D scientists and engineers.
The human genome was first mapped in 2000, but already that's so "yesterday." So says Jeffrey Killian, the director of global research and development operations at PerkinElmer Inc.'s Life and Analytical Sciences Division. Killian oversees 450 scientists and engineers and a pipeline of as many as 150 new technologies that will be used in products for drug discovery, genetic screening, and environmental and chemical analysis.
The $1.06 billion division brings in the bulk of PerkinElmer's $1.7 billion in annual revenue through some 15,000 actively updated products. Some of those are new to the PerkinElmer fold, having come from acquisitions of companies such as New England Nuclear (where Killian worked) and Packard Biosciences. The acquisitions have filled out PerkinElmer's product lines; for example, the company now has offerings in every piece of the drug discovery process, from the instruments used to run experiments to the chemicals (called "reagents") used to identify chemical reactions and the software that analyzes the results.
The company has spent the past couple of years integrating those businesses and some far-flung offices into its Wellesley, Mass.-based operations. For R&D, that meant finding a way to centralize information about who was doing what across a half-dozen sites, to resolve issues of uneven workflow and division-wide inefficiencies. Killian sought out division CIO Matt Dattilo and his 90-person IT group for help, and today his group uses resource allocation software from Primavera Systems Inc. Primavera's application tracks R&D projects and the amount of time people work on them. It went live a little more than two years ago.
Killian, who has moved his desk into the IT group for easy application support, also has engineers investigating a new IT project: a product lifecycle management (PLM) system. Here is more on how he and Dattilo work together to bring IT to bear on the organization.
Talk a little bit about PerkinElmer's space.
Matt Dattilo: The technology changes very rapidly. If the products don't get refreshed, they don't have a very long shelf life. I came from [22 years at] General Electric, where a transformer design is from the 1940s, and they've improved it 1% since then, and [it still] makes a lot of money. Here the products are being upgraded very consistently (and even then, 10 years old is a lot).
Jeffrey Killian: Who would have thought 10 years ago we'd be saying, 'Hey, they sequenced the human genome. Oh yeah, well, five years later you can go and download that in like three seconds from the Internet.' Now we want to look at much more complex characteristics in the cell by analyzing proteins and what is a cell expressing, what is it doing. That is a very big area to work on, and the tools that we're delivering are pretty cool.
The resource allocation software project started with a PowerPoint presentation. What was it all about?
Dattilo: PerkinElmer had grown through a number of acquisitions. So all these businesses were crammed together. And if you sat in Turku, [Finland], and you were an R&D person in Turku, every priority looked like a Turku project. And if you sat in Downer's Grove, Ill., every priority looked like a Downer's Grove priority. So as a corporation, we were suboptimizing because the most important project for the business may not have been in Downer's Grove; but if you sat there, you weren't going to work on anything else.
Killian: It was really a matter, as Matt says, of trying to pull all these different groups together -- first of all, to understand who they were, and then to streamline the project portfolio so that it was a continuous flow and worked across all the different pieces that we're trying to deliver to the customer.
This sounds frustrating.
Dattilo: The [R&D] guys didn't really know who was working on what, who was needed, what skills were available in the organization. It always looked like the pot was full, when in fact it was inefficient allocation. So a lot of [the effort] was identifying skills and putting some boundaries and expectations around projects. Most projects didn't even have expected budgets; it was really time-based.
I can give you an example. Shortly after I came, we discovered there were three guys in Turku who were software engineers. The software [project they were working on] was a little slow, so they were developing their own webcasting tool. [Laughs.] So you ask, why on earth are we working on this? And it was essentially because we can. This is a way to [kill] the time. And meanwhile, these guys are facing issues in Downer's Grove where they need additional software engineers, and there's no way to make that connection.
Now that you've worked together a while, how would you characterize each other's work styles?
Killian: Going back to that PowerPoint presentation, Matt just sat there and listened. It became really easy to speak with him and work with him because he was very receptive to the needs that we had. He's very patient with someone who doesn't know IT and [doesn't] know the language; he tries to understand what we're doing, he tries to accommodate us, and to look after our function and to deliver to the best of our needs. For me, it's been outstanding.
Dattilo: From our side, Jeff is a pretty analytical guy, so he's not coming in with a reaction. He's got some very clear objectives on what he wants to accomplish to meet a business need, so that makes our lives much, much easier. I'd say he was very cooperative and very collaborative for the entire project, and he continues to be.
How does your staff become knowledgeable about the business?
Dattilo: That's one of the real challenges of this business, understanding the products, because they are highly scientific, highly technical. A lot of this [the IT work] is just common process, though. On the ERP [enterprise resource planning] side, a lot of the folks have come out of the business, so that really is helpful.
Killian: For me, on the other side of the equation, it's 'How do I learn how to use the IT systems?' So I actually moved my office right across from a bunch of Matt's staff who worked with the data systems. I can walk across the hall and say, 'How do I access this system?' and 'Can you set me up in this so I can pull the data?' I don't know all the terminology and the intricacies of it all, but I know enough to ask for what I need.
So what kind of executive support does IT have?
Dattilo: I'd say we have very good linkage. When I came, I would describe it as an uneasy alliance. It was very cost-based. There was a perception that IT costs were out of line, that the linkages with the business weren't as strong as they needed to be. And really, guys like Jeff and other people in the functions [helped] senior executives see that there was value created.
Our annual budgeting process is very engaged. Every IT project, with the exception of some infrastructure work, has a functional sponsor. So the debates we have over funding are not me going to the executive team; the executive team members argue amongst themselves over who's going to get the resources for the coming year. So there's much better discussion.
What else have you been working on?
Killian: PDM [product data management] is something else that we're interested in, to take in the whole manufacturing process, from the basic architectural design of a product to the procurement of the parts. It gets really complicated, because we're not just building one certain type of instrument; we're mixing instruments with the reagents and all the software and all the types of different markets that we cater to.
Dattilo: PDM is actually expanded to product lifecycle management. It takes it down to really every piece of information associated with a product and trying to have those linkages. For example, if there's an engineering change, does that then tie into the marketing literature -- does it change the specs at all? All those things get tied together through PLM. And that's the beauty and the danger of the whole project; it gets so big very quickly that it starts to become a little bit unmanageable.
Where are you with PLM?
Dattilo: It's in the planning stages right now. We've really got to get our processes down before we start to apply tools to that.
So you have to build in processes before you put a system in place.
Dattilo: Very often what we see is that business knows it has a problem, but the initial step is to react to one of the manifestations of that problem. So you go after the symptoms. One of the things that we've done very successfully as a team is we stop and say, 'OK, let's understand the root cause of this issue so that when we implement a tool, we know we're after the cause and not the symptom.' That's really what we're doing with the PLM project right now; we're getting at what's the core issue here [which is productivity through workflow]. There's a lot of PLMs out there. We could implement any one of them and the software would run, but would it solve the underlying issue? Would it improve productivity? And that's really where our value is.
Anne McCrory is editorial director of CIO Decisions and the CIO Decisions conference. Write to her at [email protected].