More Handy Tools, Same Amount of Work. Why?

The short answer

A tool usually speeds up one step in a job. If entering, copying, checking and waiting still happen throughout the process, the overall effort barely changes. Start by writing the job out from beginning to end and look for where the same information is handled more than once.

One step versus the whole job

For example, suppose you replace paper applications with an online form. Receiving them gets easier. But if someone still retypes each submission into the customer register, a colleague then checks the register, and nothing can move on until that check is done, the job as a whole is not much lighter.

Typing information in by hand is called data entry. Copying information from one place to another is transcription, and entering the same information in two or more places is double entry. These, along with time spent waiting for a check, are where effort tends to linger.

Write out the flow

For one job, write down the following in order.

1. Who receives what
2. Where they enter it
3. Who they pass it to next

The written-out sequence of a job is called a business workflow. There is no need for a diagram; a bulleted list is enough. You can copy the memo below as it is.

Task name:
Places the same information is entered:
Places it is copied to:
Who checks it:
Where waiting time builds up:
Work that could be reduced:

Once it is written, mark any entry nobody needs, double entry, copying, and waiting for a check. It helps to have the people involved in the job look at it too, since they may see steps and delays that are invisible from your seat.

Before automating, ask whether you can cut

The marked steps are tempting to automate straight away. First ask whether they can simply stop. Entering data into a field nobody reads, or two people checking the same thing, can sometimes be dropped.

If you connect tools so information passes between them automatically, decide how to handle the cases where it goes wrong.

• When wrong information gets in, who corrects it, and where?
• How are applications that do not fit the usual pattern handled?
• Who notices if the connection stops working?

What to compare

When comparing before and after, do not look only at the time spent entering data. Also look at the effort spent fixing mistakes, how many checks there are and how long they wait, and how many cases get finished in a day. If entry is faster but corrections or checks have increased, the job as a whole may not be any lighter.

In summary