That optimistic executive in the large corner office with a view may be more anxious than we think.

More than likely, she or he would be making game changing decisions without all the data evidence that can be made available.

Many live in the comfort that their decisions are not being monitored, measured, and reported.

These “anxious optimists” could do with an enterprise data warehouse (EDW) to support the decisions they make.

Those living in comfort-zone-closets would do well to consider the philosophy: “It is not what you don’t know that is the problem; the problem is what you do know that isn’t quite so.”

And that goes way beyond the oft spoken words of comfort “you don’t know what you don’t know,” or “what you don’t know cannot hurt”, or the ultimate cop out “it is what it is”.

Frequent refrain: ‘but we already use data!’

Do you really believe that?

Get out of your comfort zones, take a closer look at how information (data) is stored and shared across your enterprise.

What percentage of your daily business decisions are you making powered by detailed cross-functional data?

If you are seeing inconsistencies in data and reporting, difficulty sharing data, and seeing multiple data sources that are not integrated, these are signs of a huge wastage of time and money.

You’re likely risking deep down customer dissatisfaction.

Looking at the very basics of data

This article’s purpose is to position data and its use in effective decision making for breakthrough initiatives and life-changing impact.

We will examine the very basics of data, data warehousing.

We will leave aside digital transformation and Artificial Intelligence because evidence-based analysis of any situation is imperative before making decisions about what to transform and determining which algorithms make sense to develop on which data.

AI does not happen without data.

Exposing data warehousing processes through a data warehouse technology stack is the recommendation.

Take note that data warehousing is an ongoing process which is designed for some future state.

It is best delivered incrementally, initiative by initiative within that overarching design and architecture to ensure integration of future Initiatives and departments.

What is a data warehouse?

A data warehouse (DW) is also referred to as an enterprise data warehouse (EDW).

It is a system used for providing reports and analysing data.

EDWs are generally central repositories of data integrated from one or more sources.

The sources are usually applications carrying out operational functions such as accounting, purchasing, sales and marketing.

They are systems that generate data from daily operations.

Basic business use of data

An EDW stores current and historical data and is used for creating trending reports making comparisons on things such as annual and quarterly business performances: this year versus last, current quarter versus previous, and current month versus the same month last year.

Often an organisation will want a comparison going back three, five, or seven or more years depending on the purpose for comparison particularly when strategic policy decisions need to be made.

These performance comparisons can also be prepared by sourcing transaction data from the various applications and consolidating it for reporting purposes on tools such as excel, and a host of other tools typically called business intelligence or visualisation tools.

Real life case study

You may face a problem or two going down the tools approach.

Here’s a real-life case where a tools approach made life very difficult for some senior business analysts when preparing end of month reporting for operational reviews: The client had a world renown ERP system which had recently been upgraded to the latest version available at the time.

While the financial accounting modules — accounts receivables, accounts payables, general ledger, purchasing, sales, and inventory were being used across eight business units, consolidated reporting was cumbersome and often pushed several days past month end.

Four business analysts were stretched, working extremely long hours every day of the month to get the numbers ready for end-of-month operational reviews.

There were extremely difficult challenges ranging from spreadsheets not coping with the amount of data and freezing in the middle of calculations.

Further complications arose when data sets from the business units required consolidation to provide a group-level view of the numbers, further stretching the capacity of spreadsheet tools.

And to their credit the analyst team managed to get the job done in the end.

But not without burnout, their home lives being adversely affected, and frustrations setting in at the office thus straining relationships.

Processing time down from 30 days to four

An inordinate amount of time was spent getting the data right, manipulating and ensuring the necessary data was being considered versus performing their actual job roles of analysing the data and making recommendations for executive action.

Getting the data right mostly meant splitting up large excel spreadsheets into smaller, manageable chunks, sometimes summarising the data thus losing the detail and ability to quickly do a deep dive to understand root cause because the detail was no longer readily available.

The ratio was easily 80 per cent data manipulation to 20 per cent analysis.

A Minimum Viable Product (MVP) solution was built to prove the value of data warehousing and present the concept to executive.

What was needed was all the data from across all business processes stored in one location.

Harmonised data.

That meant product numbers, vendor numbers, purchase orders, and payment approvals and payments were in sync by time, date, location, and business unit.

The MVP solution reversed the 80-20 ratio allowing business analysts a lot more time to look at the business strategically and engage strategically with the business.

Data harmonisation

The ERP software did its job as well as it was configured, allowing goods to be sold, invoices to be issued, goods to be delivered, payments to be collected.

Likewise for the order-to-payment process.

But the data was not in harmony.

Sometimes and inadvertently, this data disharmony caused late payment to vendors impacting the organisation’s reputation.

Competitors and some vendors made hay of the doubt, claiming the company was not doing so well and couldn’t pay the bills.

And that was the slightest of the problems faced.

Operational systems v EDW

Also known as On-Line Transaction Processing (OLTP), operational systems are designed to enable business functions individually.

They’d have a fair bit of harmonisation of data, but the fact is that’s pretty much all it did, with functional reporting within the departmental silo.

Operational systems are designed to “run the business”.

An enterprise data warehouse (EDW) is not designed to run the business, it stores the data generated from the operational systems to help understand what’s happening in the business.

An EDW is designed to tell you “about the business”.

A fundamental difference on how you see the business to enable high impact business decisions.

An EDW helps source the data from one or several operational applications, harmonise the data, and structure it to reflect how executives think about the business.

Data warehousing process improvements can transform enterprise culture, strategy, and performance trajectory by sharing information across
functional units, provide vital business answers quickly, and provide a single, trusted version of truth.

Do you need an EDW?

What are the signs that you do?

Ask yourself three fundamental questions:

• Is my data stored in various source systems?

Gathering data that is structurally different from operational system and legacy systems can be challenging for many organisations.

How can you integrate data from disparate systems with different structures?

How long does it take to collect data to produce regular reports and how long does it take to answer new business questions you hadn’t needed to before?

Am I not getting the full picture of the health of my organisation due to the lack of data integration across multiple business units?

What is it worth to my organisation to access such data from a central location, to allow us to make better business decisions faster, saving time and money that would’ve been wasted trying to retrieve and blend data from multiple operational systems?

• Are we experiencing system performance issues?

Do my operational systems have volatile data that changes frequently?

Do we run reports directly against such operational systems with real-time data that causes performance problems, affecting insights gathered resulting in inconsistent information?

What’s it worth to implement an EDW that solves this problem?

• Are we all working off a single, trusted source of the truth?

Reporting on data that is stored and formatted differently across silos results in inconsistency across departments.

Well built data warehouses improve data quality by cleaning up and harmonising data as an ongoing process as it is imported (not a one-time event).

Does my organisation need more accurate data?

Meaning do we need one version of the truth across all departments, providing consistency and assurance that each department is
using the same harmonised data.

What is it worth to have consistent results, enabling everyone to work off the same page?

 

• NALEEN NAGESHWAR is a practicing data and digital transformation adviser and implementation consultant. The opinions in tis article are his and do not necessarily reflect the views of The Fiji Times. For feedback and questions, contact: naleen@ data4digital.com