An incident in a large software development organization:
Here is a part of a conversation between a Sig Sigma Expert (SSE) and the Delivery Head (DH) in a software development/ maintenance organization where most projects were run on a T & M or Headcount based billing for their customers.
SSE: “Initial analysis shows that with minor changes in the processes and the use of some spreadsheet macros, we can eliminate some non-value add steps. This can reduce the effort required by 25% for the current mix and volume of work.”
DH: “But that will reduce my head-count and billing by 25%! My target this year for improving efficiency/ productivity is only 5%. Maybe we can implement the changes a bit every year, and not all at once. If I implement all these change right now, I will miss my revenue and headcount targets – these have the highest weight in my performance objectives.”
Another incident in a different organization:
A Project Manager’s project end bonus was slashed because she delivered her project at a much lower cost than what was estimated (the estimate was done by someone else). She was informed that her lower bonus was because the project total billing was much lower than the project estimate.
Both these incidents occured in situations where the projects were being run in a T & M (Time and Material) mode by a software service vendor organization.
The T & M mode of engagement basically shifts the cost related risks and benefits (cost overruns, cost efficiencies) to the customer, while the vendor organization has a steady return, and cannot make large profits or losses. The T & M mode is suitable in many situations – e.g., when requirements are unclear and likely to change, when the customer wants to work closely with the vendor team, when the customer wants more micro control (sometimes interference), or when the customer-vendor organizations are in the initial phase of establishing a relationship. A variation of this is Committed Head Count, where the customer and vendor agree on a fixed number of staff assigned to the customer’s work over a period, independent of the actual quantum of work. Another variation is the dedicated ODC (Offshore Development Center).
As against this, there is the Fixed Price (FP) mode, where the billing amounts and billing timelines are fixed based on an agreed value and agreed deliverables. The FP contract may have penalties and incentives built in (for delivery dates and/or quality). Effort overruns are the problem of the vendor, and effort savings are additional profits made by the vendor. Variations of the FP model include billing by volume, quality and timeliness of work done. In such cases the vendor is usually free to utilize the staff in an optimal way (maybe on multiple projects).
Many engagements between customers and vendor organizations start off as T & M, for good reasons. However, they continue in the T & M mode, even when the FP mode would serve everyone better. This could be because of inertia, because no one wants to rock the boat, or because no one has examined the issue for that engagement.
Structurally, the T & M model does not create incentives for the vendor to initiate and pursue improvements that will reduce the effort and headcount. The software industry has got addicted to T & M model to such an extent that head count growth, and billable person-days have become stated performance objectives for senior executives in many software services organizations.
Maybe the title of this post should have been “T&M model kills process improvement”, like the changed statutory warnings on tobacco products. Or is that overstating the case?
Please feel free to share your views, experiences or queries, using the “comments” feature available at the top of this article/ post.
Nothing Official About It! – The views presented above are in no manner reflective of the official views of any organization, community, group, institute, or association.