Business intelligence for the SME

By Mark Canes
Illustration: Jason Schneider

As smaller firms face competition and grow, it’s imperative they make good decisions based on even better information

When asked what business intelligence (BI) tools are used to measure their organizational performance, the common response by entrepreneurs might include Excel spreadsheets, report writers and canned reports. BI can be defined as the ability to extract actionable insight from data available to the organization, both internal and external, for the purposes of supporting decision-making and improving corporate performance.

Small and medium-sized enterprises (SMEs) are mostly owner-managed, entrepreneurial companies. And for many entrepreneurs, decision support tools tend to be a combination of static historical reports, analysis spreadsheets and gut feel, which is fine while an organization is small and easily managed on this basis.

However, as smaller companies grow or face stiffer competition, the need to make good decisions based on meaningful information quickly becomes an imperative. Succession planning adds to the importance of being a well-managed company, and being able to prove it. This is where detailed reports, summaries that are not drillable and spreadsheets can fall short. Spreadsheets in particular are potentially dangerous tools.

Alex Resnick, CA, president of management consulting and software firm The Catalytics Group Inc., points out the dangers of relying exclusively on information contained in traditional reports. Reports often contain information relating to a particular transaction type, such as sales revenue and related costs. These reports typically ignore other factors that might affect the interpretation of the reported data, such as number of customers acquired and lost during the period under analysis.

Typically the barriers to adoption of BI are cost and complexity. Costs include software tools and services, as well as people and time costs. The SME faces additional hurdles, such as tighter budgets, less sophistication and organizational knowledge, technology hurdles and fewer people, meaning less time to spend on planning and analysis.

The potential benefits include insight from data, improved decision-making and risk mitigation.
The factors involved in the implementation of a BI solution can be broken into four areas:

information (the data available to the organization),
technology,
intelligence and
implementation.

Information/data
The SME faces many of the same BI challenges as larger organizations, and some additional ones to boot. The smaller you are, the less data you are likely to have access to. In terms of internal data, a basic entry-level accounting program stores substantially less information than a higher-end enterprise resource planning (ERP) system. And externally, in the context of industry-specific or competitor data, there’s not much out there on small businesses.

Most SMEs do manage supplementary key business data, frequently in the form of spreadsheets, contact manager databases, payroll systems and other home-grown databases. These can potentially be harnessed to provide the BI backbone, provided the data is reliable and accurate. In many cases, data outside the accounting or ERP system is not subject to the same controls and may provide more red herrings than pearls of wisdom.

Technology
BI software tools include scorecards, dashboards, analytics, data mining and reports with drill downs. The available software solutions delivering these toolsets used to be beyond the financial reach of the typical SME.

In recent years, however, there have been a number of alter-natives priced more aggressively, bringing the technology for a comprehensive BI implementation into the reach of the SME. Some of these tools are available online, using the software as a service model. Others use the traditional licensing model.

Michael Burns, a CA and CAmagazine columnist, writes frequently about BI. He states that some mid-range ERP systems are adding dashboards that are configurable, with drill downs, frequently at little or no additional cost. Burns says he is seeing more affordable on-line query and browser-based tools emerging.

But is this being effectively communicated? If the perception among those in the IT industry is that the cost of technology is no longer a barrier to BI for the SME, that view is not necessarily shared by end users. Beth Crawford, CMA, controller for Toppits Foods Ltd. in Vaughan, Ont., has been working on a project to implement a comprehensive BI solution for some time. “It’s hard to find a cost-effective platform that is affordable now, but that you can grow with,” she says.

And then, of course, there is the ubiquitous spreadsheet. Packages such as Excel are undoubtedly valuable and powerful tools, and many companies have mined their ERP and other data by live-linking spreadsheets to data sources, using pivot tables, conditional formatting, graphs and charts, and other built-in analysis tools. But it is important to understand the limitations and the dangers. The flexibility of a spreadsheet is at the expense of controls, thus you can enter (or edit) any formula anywhere. We have all seen a complex spreadsheet with one incorrect (but unnoticed) formula or reference that materially alters the entire picture. Because of the absence of controls and structure, though, how would you know? And these spreadsheets are usually built and maintained by one person. If that person leaves the company, someone else will have to unravel the logic in order to maintain and update it. For key business analysis, a more structured software tool is simply more reliable and maintainable — and nowadays, affordable.

Intelligence
The single most important aspect of any BI project is the determination of what needs to be measured, and how to measure. Burns says a BI project has to begin with an organization accurately defining its critical success factors (CSFs), which he defines as “what the organization must do to be successful.” For each CSF, the trick is to determine the metrics that are appropriate to measure how the organization is doing and alert the appropriate level of management and employees to exceptions or problem areas. Those metrics will then form the underpinning of the architecture of the BI solution: the elements that are measured and displayed on dashboards and analytics.
One might think the definition of CSFs and the consequent metrics are easier for the SME because:

the entrepreneur knows the business inside out and therefore knows what’s important and
there are fewer issues to worry about in a smaller business, so the SME can use a simpler set of metrics.

But dig a little deeper and it’s soon apparent that the opposite is true. Most entrepreneurs are experts in one aspect of the business, not all. The owner with a sales bent may carry pertinent sales and margin analysis data in his or her head, but would not necessarily be able to factor in cash-flow considerations. Conversely, the engineer who started a successful business will understand what’s critical in the product design and production areas, but will he or she know what salespeople have to do really well to succeed?

And in some cases the SME faces the same problem as larger companies: too much data. Crawford says the amount of data available from the ERP system in the form of reports and data extracts sometimes makes it hard to avoid being overwhelmed with data, making it difficult to focus on what’s truly important. This highlights the importance of taking a strategic view in defining CSFs. Few entrepreneurs have been trained to think and plan in a truly strategic fashion, so this is a stretch for the average SME.

SMEs often have no one trained in financial matters, but they need someone with financial experience to ask the right questions. Entrepreneurs are not used to managing (or being managed) by metrics, unlike managers at larger companies.

Consequently, the typical SME needs outside help defining CSFs and the relevant metrics. Where do they turn? Frequently the vendor of their BI solution (or ERP solution) is the first resource. The good news about this approach is that the vendor will know how its software works, will have experience doing this exercise with other customers and will likely know how to interact with the available data. This will only be helpful, however, where the vendor is experienced in the same industry as the SME, and even then only for CSFs and metrics that conform to best practices for the industry.

A safer option may be an experienced BI consultant who can draw relevant information out of the entrepreneur, employees, competitors, vendors and customers, and then document the requirements in an actionable format. Ideally this would be completed before any decision is made on the actual software tools, to avoid any bias in the intelligence phase based on limitations or features of already selected software. Of course, engaging an outside expert will add a layer of cost to the project, but this cost differential should be evaluated against the project’s success.

I know this from experience. As a CA•IT running a software company, I assumed I knew what to measure and how we were doing on these measures. A few years ago, as the company grew, we appointed an external advisory board. With the board, we quickly identified a couple of CSFs not previously considered, and the benefits of measuring just these two items and being able to take action based on the measures have paid back the cost of implementing the advisory board many times over.

As Burns says, “You would want to reduce the risks by starting with an understanding of the CSFs of the organization and the metrics that measure whether or not CSFs are being achieved.”

Implementation
Implementing a BI solution should ultimately be the simplest part if the upfront work on intelligence and tool selection has been successfully carried out. Creating and or mapping the data store, dashboard design, report building, drill-down definition and software configuration are obvious steps in implementing a BI solution. The key step sometimes overlooked or sidestepped is the design and dissemination of a communications strategy.

“An element frequently missing is the communication piece, explaining to the users what the analytics mean, how to interpret them, and what actions to take, for example, when an indicator is red, yellow or green,” says Resnick. For the SME this becomes even more important, specifically in terms of documentation, because employees frequently wear multiple hats, and when someone leaves, their replacement needs to know those same things.

Over the next several years we will see an increasing number of SMEs adopt BI technology solutions with mixed results. Those that focus on the intelligence in BI will derive positive and measurable benefits. Those that don’t run the risk of making poor decisions based on accurate but inappropriate data.

Mark Canes, CA•IT, is president of Blue Link Associates Ltd. (www.bluelink.ca), developer of ERP solutions for SMEs. He can be reached at mark@bluelink.ca

Technical Editor: Yves Godbout, CA•IT, CA•CISA, Office of the Auditor General of Canada