S h o r t S t o r i e s

// Tales from software development

How to lose £20 million by saving £1 million

leave a comment »

My involvement with the project that I’ve worked on for the past 21 months has come to an end. It’s been a great experience with a mix of highs and lows but one thing that really stands out it just how stupid customers can be when it comes to saving money.

The client, let’s call them ‘A’, contracted with the provider, a company that we’ll call ‘B’, to pay £3.5m for the delivery of a solution that would provide a complete video on demand service with a client application that runs on a set top box (STB) and a backend server platform integrated with B’s existing systems such as network management and billing.

The project took about a year to deliver the initial release of the software with a team that at its height comprised around 60 people – architects, developers, testers, infrastructure specialists, build specialists, project and delivery managers. After the initial release the project team switched into continuing development mode to deliver service packs every few months. With the delivery of the sixth service pack, 10 months after the initial delivery, the contracted functionality was complete and it was time for ‘A’ to pay ‘B’ the balance of the agreed amount.

Around this time, ‘A’ made it clear that they considered the cost of using ‘B’ to continue developing the software to be too expensive. It made the decision to outsource future development to a leading offshore services company, ‘C’. One day, on a whim, I made a quick calculation of what the project team costs were and came up with a figure of £1.2m per month for staff alone. By coincidence, the subject came up in an engineering meeting a couple of weeks later and although a cost wasn’t given, it was clear that the team management had calculated what it was. A private conversation with one of management provided an off the record figure of £2m per month. This amount included the staff costs along with overheads such as office facilities and hardware resources. As well as high-end desktop workstations for all the developers, testers, and build team, the project also used about half a dozen testing environments made up of a total of around 150 rack servers.

The cost of running the team was subsidised by ‘B’ because it wanted to establish a presence in the video on demand marketplace. There were rumours to the effect that ‘B’ had a global budget of $200m to do this. The net cost of running the project up to the delivery of the sixth service pack was probably around £30m but the cost to ‘A’ was only £3.5m. So, ‘B’ had spent £26.5m of its own money on the project.

And now, incredibly, ‘A’ was going to transfer the development to another company to save money! ‘B’ was not unhappy with this as it took the view that it had pushed the project as far as it could in furthering its market position. Needless to say, it didn’t take long for ‘A’ to find out that ‘C’ simply could not deliver the sustaining engineering service packs at a significantly lower cost than ‘B’ had. How could they ? ‘A’ had been paying little more than 10% of the actual cost of the project.


Written by Sea Monkey

August 30, 2008 at 9:00 pm

Posted in Comment

Tagged with

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: