Expand your services to include predictability and predictability. This will increase customer satisfaction and strengthen customer loyalty. That's how it works
Author: Alexander Kern
If you show up at a popular restaurant at rush hour without a reservation, you will hardly be able to get a free table right away. If you then ask about the expected waiting time, you will receive the following information: «From experience, somewhere between a few minutes and several hours». No matter how charming the service is, this answer is not satisfactory.
In IT projects, effort is traditionally taken into account in detail - regardless of whether it is measured in person days, story points or T-shirt sizes. But this says nothing about the «lead time» of individual work packages, because effort estimates ignore dependencies that are neutral in terms of effort, such as waiting times. Effort does not provide any indication of the time required from the start of the implementation work to the actual delivery to the customer.
The exact completion date of individual functions is treated similarly half-heartedly: In Agile, you simply specify which stories are to be implemented in the next two to four weeks. According to the «waterfall model», there is a detailed plan for the entire duration of the project - but this plan is constantly being revised due to the unpredictability of complex requirements and dependencies. If you are asked about the exact completion date of a new feature, a serious answer is difficult in both cases.
Companies are required to deliver innovations to their customers quickly. The «time-to-market» is becoming a critical success factor. Approaches such as DevOps, Lean Enterprise, Bimodal and Agile are all taking the same line: new developments must not be completed and delivered to the customer in large release cycles, but at the shortest possible intervals. This minimizes the lead time.
For the business, however, the crucial question remains: When will the new feature be ready? By when can I promise my customers? What deadline may marketing communicate? Bandwidths and averages are usually based on great variance and are therefore very imprecise. When it comes to the completion of features with entrepreneurial significance, this is not enough. If predictability is not increased, business and customers will look elsewhere for solutions. Nobody waits for a table in a restaurant if experience shows that it takes between a few minutes and several hours.
The road to predictability is rocky. A rethink is necessary: Away from optimal capacity utilisation, systematic leaving of blocked work packages and high parallelism. Because all these approaches prevent the most important enabler of predictability: flow. Flow. flow promotes a low variance in the lead time of individual work packages.* In simple terms: the lower the variance, the higher the predictability.
Flow can be promoted. A small extract of the possible measures:
Once the flow is optimized, variances between individual work packages decrease. Based on the recorded empirical values, an exact forecast for the business can be made at any time. In contrast to estimates, this is done automatically and without effort. «If we include this work package in the development process today, there is an 85th percentile probability that it will be completed in five days» - such a statement can be made for each new feature based on the current project situation.
«With an 85th percentile chance your table will be ready in 10 minutes.» When it comes to a seat at your favorite restaurant, that accuracy takes on strange touches. Percentiles are not exactly charming - a time in minutes is perfectly adequate. But when it comes to the completion of differentiating functionality and a competitive advantage, business is grateful.
Restaurants and IT are all about the same thing: An extension of the service to include predictability and predictability. The goal: to increase customer satisfaction and strengthen customer loyalty. In the restaurant business, the focus on customer satisfaction is an everyday occurrence. In IT it can - still - be a differentiating factor.
* In addition to low variance, flow can bring other advantages, such as less waste through inventory, higher quality, etc.