You are a program manager with an IT Service Provider handling a project for a U.S. based client. The client uses three different telecom service providers and pays large bills every month – for a few hundred thousand calls that cost them millions of dollars. Your client negotiates customized tariff plans with each of the telecom service providers (vendors) and keeps reviewing them periodically.
Along with the summary of the bill, your client receives detailed call data records in the form of electronic data files from the telecom service providers. Client suspects that the telecom service providers are over-charging. The project is about reconciling the call data with the actual bill amount to identify anomalies in the billing.
Call data files from the telecom service providers would be imported into your application every month for a billing reconciliation run. Current tariff plan structure for each telecom service provider will also be stored in your application. The application would use the call data and the tariff plan to calculate the expected bill amount and compare with the bill from the telecom service provider. The application could potentially save the client a few hundred thousand dollars each month. Armed with the reconciliation analysis, your client will be able to take their telecom vendors to task and claim refund with penalties.
The six-month project follows agile approach and is in its first sprint. Customer had provided a sample soft copy of the call data files and tariff plans during the proposal stage. Customer has also committed to provide access to their test server pre-release – this test server has close to production volume data.
Exercise:
What are some of the uncertainties? What are some actions that you would take to tackle them? What kind of resistance do you anticipate in addressing these uncertainties?
Solution:
Let us look at some of the sources of uncertainties and see if we are able to find any in this case:
1. External interfaces
One complexity would be in the changes in the structure of the tariff plan over the course of time from multiple vendors. Now the question is: when will the complexity unravel itself? It is likely that it will happen when you test on the client’s test server. You will then be faced with all kinds of historic data and plans.
Team would like to bring this forward to the design stage during Sprint 0 or before coding the reconciliation logic in early sprints. They can push for access to the test server as early as possible. Alternatively, they could ask for past tariff plans and past data from multiple vendors and build a server with right kind of test data. Or even better ask for both.
What kind of resistance can you anticipate? Client has committed to providing access to test server pre-release only for certain specific reason – you may need to understand the reason and overcome that. For example, it may be client’s IT security policy that limits access. Client business user may be too busy to provide you with past tariff plans from multiple vendors. As a manager, you need to push hard directly and indirectly to get what you need – directly with your IT SPOC, through Account Management teams, through your senior management team to get you whatever you need to resolve the uncertainty asap.
2. Assumptions
Most important assumption is the basis of the project: Client suspects anomalies of significant monetary value in order to justify the investment in this development.
When will this assumption be validated? At the end of the project, right? – when you have analyzed all the past data from multiple vendors.
Would you like to validate the assumption sooner? The answer is ‘yes’ – from your perspective and your client’s perspective. There may be ways of analyzing a statistical sample to get an idea of what to expect with the whole population.
What kind of resistance to expect? For starters, your own management may resist the idea – loss of revenue. Your client may not want to spend money on the sampling exercise depending on the strength of their suspicion. However, with the mindset to do the right thing, you may be able to convince them.
In conclusion, this CHOW is not about how smart you are in identifying uncertainties. It is an attempt to illustrate the mindset and ownership you need to push through the resistance to resolve uncertainties sooner.