Business Analytics Made Easy

“Is a supermarket a business that sells food, or is it a business that exploits knowledge about customer preferences, geographical biases, supply chain logistics, product life cycle, many kinds of sales information to optimize its operations delivery, inventory, pricing, product placement, promotion to grow and increase margin? The answer to that question may determine your company’s long-term viability in the Information Age.”

– Source: Business Intelligence, by David Loshin, page 11.

The growth of quantitative analysis has been the second-biggest revolution in management in the past two decades. Of course, the biggest revolution in marketing has been introduction of the internet.

As companies grow, the skills required of senior managers change. At some point, senior managers can no longer afford to develop first-hand knowledge of details throughout the business. Two new organizational skills become crucial to success.

• Delegation. Senior managers must gather information through peers and subordinates, and they must execute through peers and subordinates. A senior manager can no longer succeed by being a bigger version of a first-line manager.
• Use of indirect information: Quantitative analysis and inputs from peers and subordinates replace much of the first-hand knowledge of segment details that senior managers rely on to run smaller companies.

Designing and using quantitative information wisely is an art in itself.

• Managers who use quantitative information must balance it with conceptual inputs and subjective opinions. For example, in evaluating a proposed investment project, a manager must balance strategic concepts like “market attractiveness” and “competitive strength” (the focus of our first example in this paper) with product profitability metrics (the focus of our second example), and subjective opinions of trusted peers and subordinates.
• Designers and users of quantitative information must know what approximations to accept and how they limit the credibility of the resulting metrics. For example, marketing program contribution margins (the focus of our third example) are computed using some pretty heroic assumptions – which marketing touches are connected to purchase decisions inside a customer organization, the relative impact of different kinds of marketing touches (e.g. live seminars, sales visits, web visits), the relative weight of marketing touches that generate interest and those that help close an order, the rate at which a marketing touch loses its effectiveness over time. These assumptions don’t have to be right in every case to yield very useful decision metrics, especially when quantitative information is combined with first-hand field knowledge.

The Role of Spreadsheets

Spreadsheets and word processors were the first major “WYSIWYG” (What You See is What You Get) applications of computers. They drove the desktop revolution in business (1980-2000) and, along with video games in the home, the broad adoption of early personal computers. Spreadsheets succeeded because many business analysts prefer to author models represented visually as tables and sheet layouts, instead of as pages of program code. Spreadsheets thereby empowered many people with limited programming background to build their own models. Spreadsheets empowered decentralized innovation in business analysis.

The key design innovations of spreadsheets are (1) to represent the logic of quantitative models to authors with WYSYWIG sheet layouts, and (2) to express virtually all relationships with formulas that are subordinate to cell layout[1] and that are expressed in terms of cell addresses.

Innovation continued. Lotus Improv pioneered pivot tables, but in a product separate from its spreadsheet cousin, Lotus 1-2-3. In 1993, Microsoft Excel 5.0 introduced multiple worksheets and pivot tables. Since that time, many substantial improvements were made in Excel: performance, macros, algorithm libraries, model size limits, optional add-ons and other features. However, the basic modeling paradigm of spreadsheets has not changed since the late 1970s – to capture model logic with sheet layout, and to express relationships at the cell level.

The “desktop revolution” greatly reduced the dependence of business analysts on professional programmers. At the same time, the cost of computer power was plunging, and the amounts of data available inside companies was, and still is, increasing exponentially. These developments freed business domains experts to innovate as they never could before.

Spreadsheets have some serious areas of weakness.  The field of business analytics needs a solution that preserves what is best about spreadsheets while preserving the independence and innovation that they catalyze. One of the major drawbacks of business intelligence systems is that they tend to undo the desktop revolution that freed more people to innovate. ModelSheet aims to do just that: to address the weaknesses of spreadsheets while preservering their strengths and extending the freedom and innovation of the desktop revolution.

You can read more about ModelSheet’s response to this problem and opportunity in two of our white papers.

• Customized Templates – The Promise of Spreadsheet Templates Fulfilled. This paper explains how ModelSheet technology  improves the spreadsheet modeling experience for people who never need learn anyting about ModelSheet technology. See http://www.modelsheetsoft.com/doc/whitepaper_CusTemplates.pdf.
• The ModelSheet® Authoring Environment for Spreadsheet Models.  This paper expands on the role of spreadsheets, their strengths and areas of weakness, how ModelSheet technology works, and how it improves the situation for spreadsheet users. See http://www.modelsheetsoft.com/doc/whitepaper_Intro.pdf.