The spreadsheet was the killer application that drove adoption of the personal computer in business. With Visicalc, Dan Bricklin created the archetype of the spreadsheet as we know it today, and it is this model that was adopted by Lotus for 1-2-3 and then by Microsoft for Excel. Most of us use spreadsheets everyday for everything from note taking (the grid pattern helps us organize lists), to simple calculations, to complex business models. We share these spreadsheets through e-mail, post them to corporate portals, and in some companies we even use on-line versions like the one in Google Docs. Together with presentation tools like PowerPoint and document editors like Microsoft Word, spreadsheets defined the average business user's experience of IT before the World Wide Web (in fact IT departments were much more concerned with databases and their management, a disconnect that still haunts us today).
.
Because they are ubiquitous and easy to get started with, spreadsheets are often used to try to embed complex business frameworks and new ways of thinking into organizations. A marketing group or its consultants will develop a complex spreadsheet with a lot of macros (simple programs that can be executed inside a spreadsheet) and restricted fields and then push it out to potential users throuh e-mail and role-based portals. Sometimes they will provide a seminar or coaching on the new tool, and when there is a budget there may be some eLearning or on-line help built in. At LeveragePoint we often see complex value models built by pricing experts in spreadsheets, or applications that are meant to help marketers do everything from marketing mix optimization to conjoint analysis of consumer interests. How successful are these spreadsheets in activating, embedding and evolving new capabilities?
In most cases, the answer is "not very well." The next time you see one of these complex tools, ask the sponsor if they know four things:
- How many people are using the tool?
- How are they using the tool?
- Where do they struggle?
- How have the people who are using the tool changed and improved it?
In most cases, nobody knows the answers to these questions. But if we cannot answer such basic questions, it is hard to know if adoption is actually happening and if it is having the intended business impact. It is also hard to know what actions to take to remedy any issues.
.
Then think about the speadsheets you have created yourself. How easy are these for another user to understand and to modify? In most cases it is the person who created the spreadsheet model who can best understand and modify it. We struggle to understand other people's spreadsheets, and sometimes we even go to the trouble of rebuilding our own version! Why is this? The natural evolution of spreadsheet-based applications is sketched in Figure 1.
Spreadsheets are general-purpose tools that were designed to simplify the process of creating and iterating on business calculations. They do this job very well. They were not designed to guide people through new frameworks for thinking about business problems. Nor were they designed to help people collaborate or to facilitate conversations. But we are using them for all of these tasks. It is not surprising that we struggle so much when we are using the wrong tool. Working harder with the wrong tool often causes more damage than it does good.
.
In many cases we ask business analysts to design and program sophisticated spreadsheets for us. We even use the spreadsheet as a database or an interface to a database (usually Microsoft Access). But spreadsheets were not designed to be programming environments and these monster spreadsheets are generally brittle, expensive to program and impossible to evolve and maintain. They cannot leverage the many advances in programming that have been made over the past decade, from object oriented design to web services and RESTful architectures. And business analysts are seldom capable programmers or user interface designers. The result is spreadsheet applications that are overly complex and satisfy only the business analyst.
.
The World Wide Web has been with us for more than a decade now, and is broadly used to deliver all sorts of business applications. In the past decade we have moved beyond the broadcast model of the first generation of web applications to the social web, or as Tim O'Reilly calls it, Web 2.0. Social media thinker Raul Pacheco, who tweets as @hummingbird604, has identified four aspects of web-based information that need to be considered in creating tools that foster thinking and communication (and in business the two need to come together). These can be applied to the design of applications for activating and embedding new capabilities.
- Reciprocity - applications need to support communication
- Durability - applications need to make sure that information is saved (and can be found)
- Traceability (and Accountability) - applications need to be able to trace where information comes from and what decisions are being made on the basis of that information
- Scalability - applications need to support distribution of information to and conversations between many different people, with different needs and points of view
Business tools built on spreadsheets fail to support any of these requirements. We struggle to use them to communicate with other people at all, and they are generally not a good way to facilitate communications (reciprocity). It is easy to modify spreadsheets (one of their strengths when used by one or two people to model business issues) but they are not a safe place to store data (durability and traceability). Spreadsheets are personal productivity tools, designed to be used by one person; we should not try to make them into something they are not by using them as an information distribution or collaboration system (scalability).
.
Alternatives to spreadsheets are just coming into view. For some uses the combination of a spreadsheet and online collaboration can work. Google Docs is providing some good functionality here (though it is not yet a fully satisfactory spreadsheet). Another approach is to blend spreadsheets and wikis, something that Dan Bricklin is trying to do with wikiCalc. These are potential solutions when only a small number of users are engaged. For larger and more scalable solutions we will need something more like eWorking solutions. These solutions blend step-by-step guidance through a framework combined with collaboration, learning and data. Well-designed, eWorking solutions leave critical dimensions open to be modified by users and allow these modifications to flow back into the general framework, driving a process of continuous improvement. Adoption requires more powerful applications than spreadsheets can naturally support. Let's expand our toolbox and leverage the four aspects of information that the web opens up to us.
.
Steven Forth
CEO
LeveragePoint
