register login shopping cart shopping cart 0

function libraries, tools and services for Microsoft SQL Server



Business Intelligence


Reliable, accurate, consistent, repeatable & timely reporting

Business Intelligence is the art of extracting important and actionable facts from your business data. Much of the Business Intelligence industry - at least the software end of it - focuses on the reporting tools, with the promise of end-users and stakeholders being able to create their own reports and analyses without the need for technical assistance. 

This is a worthy and important goal - it's usually hard to predict how useful a particular report or analysis might be until you've seen the result based on the actual data - and you often have to play around with it and tweak it until something interesting emerges. Such a process does not lend itself well to the traditional 'waterfall' approach of specifying requirements up-front and then waiting for the finished result to be delivered, and in every organisation we've ever worked in it's very hard for most users to get any time or attention out of the limited pool of people in the IT department capable of helping with such a task on an ad-hoc basis. If you know what answers you're looking for in the data (or at least if you will know once you've found them!), then the ideal situation is for you to be able to do the analysis yourself, quickly and efficiently, and still be confident in the accuracy of the information you've uncovered and your interpretation of it. 

However, it's never just a case of finding a clever reporting tool and pointing it at the raw data: at best that will waste a lot of your time, at worst you risk drawing false conclusions and taking the wrong actions as a result - to the detriment of your business.

Five common mistakes in business reporting 

  1. Focusing too much on presentation, not on the numbers.
    This is commonly seen when reports are created by report developers who have wizard-like skills in using the reporting tools but no real understanding of or interest in the numbers or business fundamentals that they are reporting on. They will spend a disproportionate amount of time making the report look really slick and professional, and be surprised and disappointed when the report doesn't land them praise and adulation simply because the numbers were out by 20%.
  2. Building reports that the person managing the data doesn't know about.
    This is particularly common in larger businesses where one part of the organisation has built some reports based on a database managed by another part - without putting in place a mechanism to ensure that any changes or known problems in the structure or processing of the data are picked up and dealt with by those managing the reports. When this happens a simple change in the database or processing hiccup can break dozens of reports: in the worst cases this happens without anyone realising, resulting in false reporting. Accountability for data management and accurate reporting should rest in the same place.
  3. Building management reports on live operations data
    This is very common, given that most SME businesses do not have any dedicated data warehouse for reporting against. It's not always wrong - if you have limited resources it may be the only practical option - but it carries with it certain dangers that at the very least you need to be aware of: 
    • Transactional databases are not usually designed with reporting in mind, and not usually documented sufficiently for someone who doesn't know them inside out to built reliable reports against, so constructing the reports may be difficult and the results may be totally misleading.
    • See point 2 above: the need to keep the live operational systems alive and up to date will usually override any reporting considerations - particularly for 'unofficial' reports - so don't be surprised if changes are made that break the reports.
    • Well-designed reports shouldn't impact the performance of a database but naively-designed ones often do: many's the time a sales or order processing system has been brought to a grinding halt on the last sales day of the month by eager managers running queries and reports to see how they're doing! 
  4. Using stale or inconsistent data
     The flipside of point 3 above is that if you are using a dedicated reporting database or data warehouse you do need to be 100% confident that it is up to date and accurately reflects the real operational situation. Sounds obvious, but it's easy to assume that because it reconciled once it will reconcile for all time. All systems and business practices evolve and change, and processes sometimes fail silently, so it is critical to build automated reconciliation into the system at every stage (e.g. using checksums & date/time comparisons) and to make sure that if the reconciliation itself fails it fails noisily. There's no surer way to look stupid than to make a high-level presentation or recommendation based on completely false data.
  5. Being unable to consistently reproduce an earlier report
     The progress of time is often one of the least well-handled aspects of corporate reporting solutions - especially when reports are based on live data. This is because most databases - especially operational databases - are designed to keep a transactional history of the business but not a detailed history of master data (e.g. product lists, price lists etc.) or of their own state (which at any one time may well be inconsistent with the business reality). The result of this is that it may be impossible to accurately track down and analyse apparent anomalies or surprising results highlighted by a report because the data has since changed in ways that cannot easily be unwound or reconstructed. A well-designed reporting database or data warehouse should take full account of this (known as 'temporal data') so that you can at any exactly reproduce an earlier report - even if that report was based on wrong data that has since been corrected - and hence can always get to the bottom of what's going on.

How can we help ?

For a complete 'green field' implementation our preferred and recommended approach is as follows:

  1. Understand your business and reporting goals
    We learn from you the fundamentals of how your business operates and what business needs you are trying to address through reporting and business intelligence: this enables us to put forwards intelligent and thoughtful designs and recommendations for a solution that is perfectly tailored to your needs. 
  2. Find all data sources and reliable reference points
    We work with you to identify all the data sources and systems that you already have in place and all known good reference points against which reporting data can be reconciled on a continuous basis (e.g. audited accounts). 
  3. Design and build a robust reporting database
    We design and build a data warehouse (reporting database) that efficiently and reliably captures, summarises, marries together and reconciles data from all your transactional systems using temporal records so that reporting can be rolled back to any earlier point in time from both a business and system point of view.
    The system will be delivered fully documented and with operational management interfaces and dashboards so you can support it yourselves (should you wish to do so - we can also provide ongoing support). This clean and reconciled view of the data can then be the basis for all reporting. 
  4. Choose simple and accessible reporting tools
    We work with you to identify the best tools for delivering the reports and analysing the data. In most cases we recommend familiar tools you already have access to and licenses for - such as Microsoft Excel or the Microsoft SQL Server reporting and analysis tools (SQL Server Analysis Services, SQL Server Reporting Services, PowerPivot). 
  5. Feed you and then teach you to fish
    We can then build any number of tailored static or interactive reports for you, and/or work alongside you to train you in building your own standard or ad-hoc reports and analyses.  Reports can be hosted online on intranet web pages or Sharepoint, or viewed locally using auto-updated Excel workbooks.

However if you have already made investments in or inroads into addressing your reporting requirements we are very happy to review, tweak, adjust and add to existing systems.

Call or write to us today to discuss your requirements

  • Europe (UK): +44 (1865) 58 9009

  • North America: +1 (909) 366-5333

...or send us an email enquiry or request a callback at a time that is convenient to you using the buttons above. 



Please give your telephone number and preferred call-back times.

Number to call:   (please include country code + area code)
When to call:


Thank you

Your callback request has been registered. Someone will contact you within the timeframe you have specified.