Freeze those cubes now!

There is no real reason for it to be unpoular because actually it takes the 'boring' grunt work away from the IT departments and puts it firmly in the hands of the user. So there will be no more statements such as '.. I know I said but I actually meant ..'

In reality this division between IT & the business in somewhat in the past - except of course in 'flame mail' articles such as this, whose job it is to make an impact

Of course the flip side is that users will have to look after themselves and if they choose to do their own thing cannot expect the IT department to bail them out if it all goes wrong - basically you break it, you fix it

The only reason anything is about being in control is if one is accountable in the case of something going wrong - who in their right mind is going to accept responsibility when they have not control over their charge?

Would you sign off an audit on figures prepared by someone else without knowing what they contained?

Sensitive doesn't mean that anything goes or that best practice should be ignored in the headlong rush to trash the IT departments & get your own way - on this basis it is surprising that there is any PI insurance available to firms that go down this route

No accountability, no risk assessment, no documentation, no training & no understanding of any potential wider impact

In other words the more powerful the tool the bigger the scope for getting it wrong - Sounds just about right for the Excel evangelists who on the whole do not even attempt to adopt best practice or any standards whatsoever.

Or to bring it home - when was the last time that DC actually provided - specfication(at the outset), documentation etc.. for a spreadsheet? Post this information and we can all use it as a guidance and a discussion template

Unfortunately we seem to adopt the attitude that we have a new tool and it is our right to use it however we choose; irrespective or understanding or consequences

Well, the contributors so far have pitched us right into the fiery furnace of management reporting's longest-running debate - the role and control of end-user computing. It's already a little heated, but can Anonymous remember that David's original purpose isn't to post an incendiary article, but to get a constructive conversation going about Excel reporting - you're raising some very salient points, but within a semi-private discussion group. (Please note that I may well draw on things said here in articles for the wider AccountingWEB readership).

Although it was founded on the principle of helping people to be more effective with their spreadsheets, ExcelZone has long promoted the idea that Excel is not the be-all and end-all of management reporting - there are a lot of tasks for which it is not appropriate. Ever since Enron and the Sarbanes-Oxley Act came in in 2002, Big Four auditors like PricewaterhouseCoopers have cracked down on uncontrolled spreadsheet use in corporate reporting, and have laid down strict guidelines about end-user computing.

Yet Excel is embedded as a feature of business culture, people are using it and getting a lot of benefit out of its flexibility, adaptabilty and presentation features. While I can see both sides of the argument so far, I would think that one of the strengths of PowerPivots is that they rely on the central SQL Server database - although the PowerPivot users will be able to download and distort the numbers to their hearts' content, the central IT team should still have adequate control to ensure that inaccurate figures are not written back. Surely that would bring some improvement to less than satisfying reporting environments?

We've tagged relevant content on PowerPivots so that you can see all the relevant articles on AccountingWEB in once place, and as part of the SQL Server 2008 launch, Microsoft opened a very useful online resource centre. You can also follow the PowerPivot team on Twitter, but they've gone rather quiet since the launch last month.

As I commented on David's original PowerPivot post in this group, you need to have all the requisite Microsoft facilities in place and then it will still take some time to get the tools to do what you want. Can any other members of this group offer there experiences of the pros and cons of the PowerPivot approach?

@David Carter

Just to jump in

In fact a number of Cloud applications currently have options to deliver reports in Excel, PDF etc. so surely this is not a new concept except in so far as it now becomes more of a BI tool

Once they adopt this approach it should provide even better analysis/reporting facilities from the Cloud/SaaS

But in a Cloud environment IT departments have very little input other than to keep the hardware running or write the initial export, so the impact on them could well be minimal. But the initial export still has to be developped in the first place

One of the advantages I see to products like PowerPivot that 'empower' a user to produce reports or perform data extracts/summaries is the increase in timeliness of the information to the end user.

Thats a weakness of most existing BI products - the time it takes to get the info out. It is the speed of extracting the required info. that decides whether it is "very useful info." or "thats interesting, but its now to late to use ...".

The accuracy may not be 100%, but any increase in the timeliness of info. goes along way to providing pertient info. on which managers can act before its too late.

 A lot of the problems I have when performing analysis at work is the sheer inflexibility of corporate solutions whereby I cannot directly access the underlying company data unless I make use of their own "reports".  I think this also frequently encourages the end-user to end up storing the data that is available in their own end-user files such as Access/Excel/csv.

The availability of PowerPivot will not necessarily improve this state of affairs unless access (read-only, naturally) is granted to the underlying data.  This is not to say I am not excited by the prospect: I am, I can just see its potential being somewhat constrained.

Of course, should my concerns prove unfounded then I will gratefully embrace the technology and analyse to my heart's content - without any need to submit a change request to IT Dept A (which needs to be signed off by IT Depts B, C and D) before they will consider altering a field name in their output of Report Z...

@RichardSchollar

Agree with you about steps to make PowerPivot more useful with access to underlying data - however

• this may mean a knowledge of SQL and the underyling table stuctures
• it is all too easy to use an incorrect JOIN and only extract some of the required data - at which point the user doesn't realise the data is complete and bases future judgements on incorrect information; possibly the most dangerous scenario?
• alternatively data dictionaries are required

Nevertheless unless everything is 'wrapped' in some form of user friendly data selection module that takes care of these issues the users will not have the implied intrinsic flexibility to make full use of this tool.

At that point does it only become a bigger version of Excel?  So without simple access to user 'defined/selected' data it really looses its edge?