When a contract has a fixed price and a fixed date, who should absorb the cost of delivering late or delivering less than the original scope on the fixed date? This is one discussion we’ve been having on the mailing list.
Jens Ostergaard: 80% of the functionality of high quality is guaranteed at the fixed price. The 80% of high quality means that you deliver according to your definition of DONE (which is an agreement between customer and supplier on a feature level).
Jeff Sutherland: Vendor assumes the risk of late delivery of the agreed upon scope.
Jens: As to sharing the risk, if you have a confident team who have done this several times you could do what Jeff says. I would however be cautious starting this with a new company, and have some kind of risk sharing. Remember we are in this together and do the best we can. Customer knows what they want built, supplier knows how to get there, and both parties need to recognize that we have to work together.
If you work in an Agile company, how do you share risk with your customers?
Great discussion.
Jeff…it the we (the vendor) is supposed to take the risk of late delivery/scope, then you miss the hole point in a good agile contract. How are we supposed to get the customer involved, if they don’t risk anything (money vise)?
Customers (and the people representing the customers) are very different. Even with a very clear definition of done, it is hard to synchronize on what “High quality” is for that particular customer.
The problem is that quality is many things: bug free code, maintainable code, tested code, good usability, nice design, good performance. All thinks that are subjective and which are very hard to define in a contract (agile or not).
By getting the customer involved in both the project and the risk, we ensure that they try to get the most value for the money, while we still get paid for the time we use.
When starting a new project with a new customer/project, the only sound way to form a contract, is to share the risk. This way the customer is forced to actively participate in defining “good enough”, based on business value/cost/time to market etc.
Comment by Thomas Jespersen on October 25, 2008 at 10:37 am
Where we currently work we are trying to move our suppliers to a hybrid model. In say, a 9 sprint or iternation project lasting 18 weeks, we would work on a time and materials basis for the first 3 sprints and then have fixed price for the remainder.
The first 3 sprints allow the supplier to get their hooks into the highest priority stories and to assess their velocity accurately. For this period, risk is effectively shared 50/50. This highly incentivises the customer to be involved at the stage of the project where customer involvement can make or break the project. The supplier is always incentivised as each sprint has a break clause, and if the customer perceives there to be not enough value being delivered, they can break. After 3 sprints, risk to supplier is reduced as they have a much better understanding of their team velocity, while risk to customer is reduced by fixing the price for a set number of story points to be delivered.
Comment by Asitha Rodrigo on October 27, 2008 at 12:02 pm
I think I like the Hybrid model. We had a T&M model with one of our customers. Despite the team delivering high quality work, as planned, the team had to take the blame of ‘Lower productivity’. Measuring the productivity in itself is a big/complicated task, anyway.
Hybrid Model gives both parties an opportunity to get the project going, while limiting the risk to the early sprints for the customer and at the same time giving the vendor an opportunity to figure out and understand what his velocity is and better estimate the fixed price costs later on.
However, one question still remains as to how much flexibility should be built into the contract process to account for change in requirements and in what manner. I would be very interested to see further progress on this project as this will help me move to fixed price contracts from the current T&M model.
Comment by Venugopal Mallarapu on March 4, 2009 at 12:21 am