Improving operational reports with infographic style
 Regular readers will know I'm on something of a mission to improve the way reports are presented. A lot of the reporting I've had to contend with in the past has been numerically perfect but awkward to interpret and thus difficult to act effectively on. The fact is that tools such as Excel have created a perception that "reporting is easy" - leading to the mess of management information, charts, tables and reports that we have to contend with each and every day.
My thinking was how to improve a monthly operational report - the file performance of an outbound call centre. Each period a call centre receives a file of records, which have to be screened to remove any contacts who may have registered with the Telephone Preference Service, that is then fed into the dialler. Calls are placed, contacts made and somewhere along the line someone buys a product. File Performance is, in effect, a count of what happened to the records. Usually presented as a table with various flags and - if you're lucky - charts - it is supposed to help management understand what is going on with the file, which in turn indicates possible problems with the nature of data coming through, why contacts are not made, sales fail to materialise and so on. It is, therefore, a prime candidate for being improved.
The thought process I went through was to understand what sort of decisions are being made about the file by operations. This was an interesting exercise as sometimes the operational needs can be overwhelmed by requirements from elsewhere. My concern is making sure the operation is there to support the sales process, not to drive the sales process itself. The report I came up with is this one….
 Full size version is on Flickr. Click the link to see it, refer to it and download it.
On the top of the report is the headline performance, showing me how many records passed through each step in the journey to a successful sale. I can see: what I started with;
what was left after the TPS filters had been applied;
how many records we actually dialled;
how many were valid contacts;
how many were eligible to buy the product; and
how many were sold. Each is mapped against its target - green means we beat target, red is not so good. The presentation of the bar chart gives me a quick visual cue that tells me right away something went a bit wrong when we started looking for people who were eligible for our product.
The next layer gives me the "what" for that question. Each step in the process is broken down and presented as a pie chart, so right away I can see where the big issues are. For example, in my "Non Sales" I can see straight away that a third of non-sales had "Reason not stated" as the outcome. In other words, we don't know why these people did not buy.
Which takes me to the layer beneath it - the commentary. Numbers are numbers, but unless I document my assumptions and my reasoning when this report is picked up in 3 years time no one is going to know what I did with it. Within each major block is documented some basic, short form thoughts on what is of concern and what action needs to be taken. Following up on the "Reason not stated" issue, I have asked Learning & Development to review this to determine if agents are simply not completing the record completely, or we need to think about inserting some new outcome codes.
This then drops into my final layer - what actions are being taken, what have to be completed. You'll note, for example, that Account Management were supposed to resolve a data quality issue, which has yet to be completed.
What I've done with this one report is fairly basic and simple. There could be further refinements and improvements made, details put in place, version control, summaries and so on. What is important, however, is this has been built ontop of Apple Numbers in a way that is entirely repeatable. Next month I can receive the "easy to produce" tables of stats and pump it straight into my template and produce another report.
It is worth noting that this is just a snapshot in time and there is no trending. I haven't forgotten it, I just believe that is a different picture that needs to be drawn, and mixing the two can create confusion.
Putting this together took me a couple of hours as a one-time hit. Maintaining it each month might take another hour. Yet I think the impact of this report is much greater and its usefulness increased exponentially by its simplification, its clarity and its very specific purpose - to support me in managing that file.
Note: All data on the report is purely simulated and isn't taken from real data. Basically it is fluff produced to give the report some meat.
Previously on this blog...
the global leader in Contact Center Consolidation 2.0
2.0 has become a meaningless addition to already poor tag lines.
A dozen beautiful images of Saturn
Wired presents a dozen of the best images from the Cassini mission
Setting up shop in a new country: beyond the website
Building a website for multiple languages is not just about translation. It is a critical business decision that has to be taken carefully.
Why call centre staff deserve your respect
If call centre staff set the first impression for your business, why do we treat them so badly?
Becoming a Specialist? A hard decision to make ...
Specialising requires hard strategic decisions to be made about your business.
When good people move on
Losing a member of staff to another company is not necessarily a bad thing
The quest for quality in Agile Software Development
Why quality assurance remains a central part of project management, regardless of the use of Agile methods
© 2010 Ross Hall. All Rights Reserved.
If you wish to use any of the content from this site please contact me.
All contents provided for information purposes only.
|
|
About Ross Hall