Last Tuesday I gave a presentation at Flow Con France. Alain Buzzacaro (@abuzzacaro) took this great photo.
This section of the presentation started with a short introduction to Cynefin (At the Snowdonia Retreat in October, I became an Authorised Cynefin Trainer). A number of classes of investment were categorised using the Cynefin framework. (Thank you again to Alain).
“Regulatory” investments are obvious. If you want to continue a business, you need to comply with the regulations. You either comply with new regulations or divest the building. It is possible to calculate the value of the investment.
“Cost Saving” investments normally involve defined cost savings. Analysis will lead to a defined value. They are complicated.
“User Needs” investments involve identifying the often unexpressed needs of customers and forming a hypothesis about the need. It is then necessary to test those needs. Forming and Testing hypotheses is by definition in the Complex domain. Testing hypotheses will lead to learning. The value of the learning outcome is by definition difficult to value. Disproving a hypothesis has different value to not disproving.
“Technical Debt” investments reduce lead time, the uncertainty surrounding lead time and the the probability of a bug in production. Accurately estimating the benefits in these terms is uncertain, expressing the benefit in economic terms is a double level of uncertainty. As such, these forms of investment require either hypotheses or action.
“Innovation” investments require action. By their very nature, there is no data to form a hypothesis.
Whereas “Regulatory” and “Cost Saving” investments can be expressed in economic terms, “User Needs”, “Technical Debt” and “Innovation” investments are more commonly expressed in non economic terms. They are expressed in value metrics such as number of customers or “engagement”. An example of comparing “User Needs” investments with “Cost Saving” investments might be “Fifteen million new customers in Vietnam” versus “Two million dollars cost saving in the United States”. There is no exchange rate to perform this conversion. In fact, the exchange rate is a pure expression of strategy. What is more important to the organisation, short term cash flow or longer term growth.
In order to perform an economic comparison, the investments must all be expressed in the same economic units. This means that “User Needs”, “Technical Debt” and “Innovation” investments are at a serious disadvantage in any discussion focused on economic comparison.
Cost of Delay is fundamentally based on the concept of economic comparison. Therefore, if your portfolio contains a mixture of “Regulatory” / “Cost Saving” investments and “User Needs” / “Technical Debt” / “Innovation” investments, extreme care must be taken when using Cost Of Delay.
OR
- GIVEN the portfolio contains a mixture of “Regulatory” / “Cost Saving” investments and “User Needs” / “Technical Debt” / “Innovation” investments
- WHEN using Cost of Delay
- THEN Extreme Care must be taken.
A shorter Cynefin explanation is that Cost of Delay is a governing constraint that is appropriate in Complicated situations, but is not appropriate in the Complex or Chaos domain. In order to prioritise a portfolio containing investments in all five domains, you need an enabling constraint… More in the next blog.
December 8th, 2018 at 1:54 pm
“Technical Debt” investments impact Throughput first ….. reduce lead time is not the first choice .
December 9th, 2018 at 8:48 am
[…] of approaches to portfolio prioritisation approaches led to an understanding that Cost of Delay is a governing constraint. It is effective providing the portfolio only contains items whose value is in the obvious and […]
December 9th, 2018 at 9:29 am
[…] of approaches to portfolio prioritisation approaches led to an understanding that Cost of Delay is a governing constraint. It is effective providing the portfolio only contains items whose value is in the obvious and […]
December 9th, 2018 at 10:09 am
You could argue that regulatory investments and cost savings are also exploratory in nature at a more granular level.
In other words, at a macro level, yes, it’s “obvious” that the organisation needs to be compliant, but *how* it achieves that goal at a tactical/ operational level has some degree of uncertainty (until the techniques can be codified and ultimately automated).
December 11th, 2018 at 6:50 am
[…] Extreme Care when using Cost of Delay Written by: Chris Matts […]
December 17th, 2018 at 5:00 am
[…] Chris Matts notes that cost of delay is only a valid element of some investment cost models. 2 minutes to read. […]