Defining business measures

There are so many methods, strategies, books and theories about business measures but I never found an answer for a very basic question:

What shall I exactly define to have an usable, implementable and meaningful measure?

So I decided to digest a summary from many practical working situations, illustrated by an actual example and considering IT implementation details (this is the side I usually seat!)

Using a schematic definition for each single measure produces an homogeneous, cross-referenced result that is easier to prepare, update and understand and agree compared to classic narrative.

Here is the result:

Information

Description

Rationale

Name

A unique short name for the measure.

Avoid confusion: use always the same name for a single measure.

Description

Brief description of usage and importance of the measure.

If you can't define its importance forget defining anything else: the measure is useless!

Interpretation

How value shall be interpreted:

  • optimum and limiting values,

  • expected (statistic) dispersion,

  • how evolution shall be interpreted.

Interpretation shall be as much consensual as possible to avoid endless discussions once the measurement system is up and running.

Classification

Identify dimensions:

  • potential grouping (organization, product, geography ...)

  • update frequency.

Key information for your IT staff planning Data Warehouse granularity.

Interrelationship

Indicate other measures that:

  • are antagonist to the measure,

  • can be used to improve interpretation,

  • are someway related to the measure.

In business, like in medicine, most factors are related and shall be considered together to identify causes and plan improvements.

Audience

Recipients and potential recipients of the measure.

Another key IT information to plan access and visualization of measured values.

Computation

How are values obtained?

  • formula and algorithms used,

  • data source.

Shall be “crystal clear†not only to accrue data but also to allow traceability whenever needed.

Examples

Supply examples including units and format.

Nothing is more elucidating that some examples!


And the promised example:

 

Name

Close Rate

Description

The proportion of prospects concluded with success (sale made) over all concluded prospects. Even not measuring economic result, close rate is indicative of sales force success and directly reflects on its enthusiasm and motivation.

Interpretation

A low close rate can have several causes:

  1. poor lead generation,

  2. adverse conditions (price, competition),

  3. failure in sales process.

Case 1 can be usually improved by making more selective lead sources (see classification) with low close rate. Case 2 is normally indicated by unexpected time variation of close rate, sometimes in specific geographic regions. Shall be analyzed on case-by-case prospect as soon as detected to create immediate response strategy. Case 3 is usually restricted to specific sales reps, possibly due to limited expertise and can be usually addressed by sales supervisor.


Classification

By representative (and surrounding hierarchy) and lead source.

Collected monthly.

Interrelationship

Close rate shall be compared with markdown measure to prevent cannibalizing success (aggressive discount and conditions to close sales anyway).

Audience

Sales managers at all levels.

Computation

Close rate shall consider prospects lost with high aging (elapsed time since prospect opening). Such adjustment is needed to avoid measure distortion caused by reps tendency to leave opportunities with limited chance. Thus, when extracting data, unclosed prospects with aging >= 90 days shall be considered lost.

Close rate = S / (S + L + A)

where:

S is the number of prospects closed with success during period,

L is the number of prospects concluded as lost during period,

A is the number of prospects that passes aging limit during period.

All information is retrieved from Prospects database


Examples

Prospects closed with success in June: 12.

Prospects concluded as “lost†in June: 46.

Open prospects that passed aging limit in June: 31

 

Close rate = 12 / ( 12 + 46 + 31) = 13.48%


Am I forgetting something? Let me know!

 

Updates

After publishing this post and receiving positive appreciation I decided to make freely available a spreadsheet to collect this data, and finally KPIStudio: a free application to validate the spreadsheet and produce cross-referenced documentation.


Last modified on 2011-05-23 by Administrator