How to reduce waste in IT from over processing

The next major waste we will look at in this series is over processing. In manufacturing, over processing means doing more than the customer requires. For example: painting areas customers don’t see or producing products with unnecessarily tight tolerances. Over processing increases labor, material, time and costs without delivering extra value to customers.

In software development over processing is caused by designing and building more complex applications than users require.
For example:
  • Rebuilding an application on a new platform when the old platform could have been upgraded for a much lower cost,
  • Using a complex middleware product that causes a lot of extra work,
  • Building an ideal server platform that can cater for far more users than you expect to use the platform.

Over processing is common in traditional software development because design is done upfront by the software architecture team, where requirements are often overstated, the solution is still uncertain and the costs are generally unknown. As a result traditional software architects design large, complex solutions using the latest technology with little consideration of the cost. Once again losing value for the customer.

In IT Operations over processing occurs when you give customers more than they need sooner than they need it.

For example;
  • Setting up a new employee’s computer weeks before they start,
  • Getting new servers into production when most projects won’t use them for another two weeks.

In lean manufacturing you minimise over processing by developing acceptance criteria that customers value and standard operating procedures to meet the criteria with the minimum time and effort possible. In software development we can minimise over engineering by moving from big design up front to continuous just in time design that evolves as we learn what we really need to build. We do this by making design a role within a cross functional team that develops software from beginning to end in small batches, with a short cycle time and a lot of feedback from users. This aligns well with the Agile principle that “Simplicity–the art of maximising the amount of work not done–is essential”.

In IT Operations we can minimise over processing by asking users what they need and then developing standard services with standard processes and end to measures of customer value to meet those needs. Anything that doesn’t add value to users can be taken out of the process.

Unfortunately many traditional IT departments apply Lean Principles in a way that cuts IT costs locally at the expense of customer value. For example it’s common for Contract Managers to outsource IT services on a long term contract to the IT provider that promises to meet service level agreements (not user needs) for IT transactions (not end to end services) at the lowest cost (not highest value). To minimise their costs commodity service providers typically apply strict operating procedures that push as much responsibility for the work as possible onto users and other teams. Then they reduce labor costs as much as possible by reducing the team size and skill and sending the work offshore until they just barely meet the minimum transaction requirements. In these arrangements it’s common for new employees to find that they can’t get a PC or login until weeks after they’ve started, it can take weeks to get a firewall rule changed, and months to get new servers in production. This approach is not recommended because it substantially reduces the value of IT to the organisation.

The next article will discuss how to reduce inventory waste in IT.