While trawling through the old files on my organization’s server recently, I stumbled across a 2008 report in which Forrester Wave for Enterprise Business Intelligence Platforms assessed 12 BI vendors/tools. While the report was incredibly interesting, I realized that it did not include Excel as one of the 12 solutions—which got me thinking: Can Excel be considered a complete BI solution in its own right?

The Forrester report, available for free here, specifies that in order to qualify for the survey, the vendor (and solution) must provide for the following:

• At least three out of the four major functional BI components: production/operational reporting, ad hoc querying, OLAP, and dashboards.
• At least three out of the four major functional BI components: production/operational reporting, ad hoc querying, OLAP, and dashboards.
• The ability to query databases using SQL and MDX.
• A self-contained, complete, fully functioning BI environment, i.e. the tool must be a self-contained, complete BI environment or platform that does not have to be necessarily embedded in other applications.
• Sufficient market presence and interest from Forrester clients, defined as at least 100 in-production customers present in more than one major geographical region, with more than 10 percent enterprise-grade, cross-line-of-business installations with more than 100 users.
• Significant BI revenues: at least $40 million in BI revenues.

Interestingly, it’s not too much of a stretch to make Excel fit into all of the above criteria, although it’s important to consider how outdated these criteria may be:

• Excel can be (and is often) used for operational reporting, by connecting to databases, and also for analysis and dashboards using Pivot Tables and Charts (plus the power of VBA — Visual Basic for Applications).
• Excel does not necessarily have to be embedded in other applications.
• Excel certainly has sufficient market presence, well exceeding the stipulation above.
• Excel comfortably crosses $40 million in revenues, even when constrained for reporting and analysis usage (Microsoft Office accounted for some $10 billion annual revenue back in 2008, give or take a couple.)

Which leads us to the question: Should Excel be considered a BI tool, worthy of inclusion in surveys such as the Forrester Wave? To draw a rough analogy, can a Swiss Army knife — tiny knife, clippers, corkscrew and all — be considered a tool for hunting?

This thought was ran by the author of the report, Forrester analyst Boris Evelson who, incidentally, is also the author of an interesting report the year after the report in question on the use of Excel as a BI tool (Ouch! Get Ready — Spreadsheets Are Here To Stay For Business Intelligence).

In his response, Evelson explained that if he included Excel as a standalone BI solution and not just a BI presentation tool, it would have scored rather poorly. This is because it still relies on the underlying BI/DBMS infrastructure for components that are critical for large, complex enterprise BI, such as real-time BI, OLAP beyond Pivot Tables, report formatting and scheduling, unstructured content, etc.

“I vote for Excel as a BI UI, or Excel as a lightweight, departmental or SMB BI solution,” he wrote. “But standalone Excel cannot be the one-and-only BI tool to fulfill complex and broad BI requirements in mid-to-large enterprises.”

Excel continues to pose an intriguing challenge to BI solution vendors and practitioners alike, the central question being: When implementing a BI solution, to what extent do we intentionally architect a solution that caters to users preferring to stay with Excel, and/or design the solution to draw users away from Excel?