Young, upwardly mobile enterprise seeks internal IT partners for long term, harmonious relationship.If you’ve got the speed, we’ve got the need!
Strolling through the Toronto offices of Lavalife Corp., you could easily mistake it for one of the countless dot-com start-ups of a few years back that expired when the Internet bubble burst. Casualness is the order of the day, with T-shirts and ponytails much in evidence.
But there is an essential difference between Lavalife and yesterday’s dot-coms. Whereas the latter survived on a wing and a venture capitalist’s prayer, Lavalife plies its trade in one of the hottest paid categories on the Internet: online dating. In 2003, US consumers spent $313 million on online dating, and that is expected to grow to $398 million this year and $642 million (all figures US$) by 2008.1
Needless to say, times are good, and cash flow is not a problem. In fact if a problem exists at all, it is in keeping up with this hot growth market. And that problem must largely be dealt with by the IT department.
“The competition in the personal space is intense,” said CIO Loren Hicks. “Lavalife is one of the top five players globally, and is number one in Canada, and on any given day several new start-ups are trying to get into this space because there’s money to be made. So we are in a constant competitive battle, and time-to-market for us is everything.”
It is because of these time-to-market issues that Lavalife has embarked on a highly successful Rapid Application Development (RAD) process, which it developed internally.
A year and a half ago the IT department was drowning in BAU [business as usual] changes. “These little changes are easier to spec, there’s lots of them and they would essentially swamp everything else,” noted CTO George Howitt. “We’d find that we were having trouble getting the important tactical or strategic projects done. All we were doing was lots of little stuff that wasn’t going to move the needle on the business.”
Added Hicks, “Everyone in the business is pretty bright, and we were awash with good ideas, but there were too many to deliver on. We were not getting consistent aim, load and fire. There was a lot of aim, aim, aim, but not a lot of load and fire.”
Lavalife’s application development process at the time was one that many companies will be familiar with, characterized by the business kicking projects over the fence to IT and IT kicking them back with its own alterations – and so on and so forth, resulting in many projects carrying on much longer than anticipated. You all know the drill.
overhauling the development process
With time-to-market being so important in the online dating business, something had to be done to enable Lavalife to get products out the door faster. So about a year and a half ago, Hicks and Howitt entered into discussions with senior management around how to speed up the process.
Through these discussions it became pretty clear where the bottlenecks were: the design/develop/test process itself and the availability of leadership resources. In an effort to address these problems, Lavalife made some structural changes, shifting responsibility for different types of projects to different groups. The smaller, “business-as-usual” types of projects would reside with the business units, while the larger tactical and strategic projects would reside with a special product development team.
“We’ve moved the bigger projects to a separate group with dedicated leadership, dedicated designers, and dedicated project managers,” explained Howitt. “The designers work with the developers on a daily basis. Instead of someone going off and spending weeks building a business-requirements document and kicking it over the fence, we now have two people sitting down and working together, making sure that both the business and the technical requirements are met. It lets us get a lot of products out the door much quicker.”
So far the results have been impressive. The lapsed time between the start and finish of a project is now about half what it was in the past. Said Hicks, “Now we’ve got a gun, we know how to aim, we know how to use it, and it’s working very well.”
the right level of involvement
A key aspect of the overhauled process is dedicated leadership. Co-chairing the design and development group are Howitt and Chief Product Officer Ed Lum, one of the company’s cofounders.
“Having dedicated leadership gives us somebody who is there all the time, focusing on the product changes, rather than spending part of their time on the budget and operational problems, or dealing with HR issues,” said Howitt. “The other big thing is having dedicated project managers, who manage all the task lists and project lists, and schedule resources and launches. Having that administrative help moves the process along quite a bit.”
In addition to dedicated project managers and product designers, the right individuals for a specific project are cherry-picked out of the most appropriate areas of the tech department (e.g. the Web tech team, or the voice technology development team).
The process is quite a bit different than in other organizations using rapid prototyping. In many such organizations, one or two analysts from the business side are put in a room with a few people from the tech side in order to get the job done. Not so at Lavalife.
“Here the process is much more open,” said Hicks. “The media people are involved, the finance people are involved – it’s a real open forum. A lot of projects are built with a cast of almost dozens before the product gets to market. Everyone is invited to participate.”
He added that in normal circumstances these meetings might be considered far too large. But in Lavalife’s case, where everything is customer-facing, they are not, because there is a need for many people’s input into the development process.
first test for the RAD process
Having created the new RAD process, the next step was to put it to the test. That opportunity came with the challenge of adding video to the Lavalife Web site, enabling visitors with Web cams to record and send video instant messages to other users, in addition to the standard text instant messages.
This was a feature that the business had looked at for some time, musing over what customer acceptance would be and how the company could make money on such a service. But little progress was being made in turning ideas into reality.
Seeing an opportunity here, Howitt and Hicks decided to go out and make the opportunity happen. Howitt started looking at video technologies and assembled a list of things that would work, which he passed along to the R&D group within tech. Then the business side got involved on specing out the new product and fitting it into the site. Within four months the capability was delivered in prototype form to the firm’s Manline