Identifying Trends!

[missmcp]

Hi,

I am trying to concentrate on the Proactive side of Problem Management, eg identifying trends etc from the calls logged by our service desk however I am really finding it difficult to identify the type of stats I should be running to do this.

At present I have a list of top ten calls raised per week but thats it!!

I've not really used trend analysis etc before and was wondering if someone could provide some advise on what type of searches I should be doing to provide results that my managers would find useful.

Any help would be greatfully appriciated

Regards
Miss Mcp

[_GrenDeL_]

Have you tried to present the data historicaly on line graphs?

For example this 10 top categories weekly for last 3 months. You can check in loger period fluctuation of calls. Necessary condition is that the data have to be categorised in proper way.
Sometimes some interesting trend can be found. Use the Excel (especialy pivot table) because it is excellent tool for this work.
On this analysis you can indicate managers the area for improvement.

[javierarcal]

Hi

I have been done a trend analysis based in systems (Clusters or Servers) also this can be done using categories, that were causing most of incidents and calls recorded by Service Desk, we prepared some bars graphics like Pareto Diagrams

We have found Pareto rule 80/20 (80 percent of incidents were caused by 20 percent causes)

Hope it helps
Javier
ITIL Consultant
Madrid-Spain
javier.arcal@gmail.com

[shaster]

Hi, bit of a n00b here, but I basically qote some success from using a real tool we use at our callcentre for call stats capture. We use an in house developed tool called a "call driver", where the agents on our helpdesk click on a pre defined button for known incidents. This data is captured with the type/incident code, and the captured data can then be exported into an Excel spreadsheet for further problem management diagnosis via control charting & Pareto charts, etc.
The program is essentially an Excel workbook with VBA integrated to allow for the functionality of the captured data.
I'm not sure exactly how it was created, but suffice to say that the captured data is invaluable for immediate trend analysis!!