Paul Trotter, Author-it

When people first started creating software to automate their business processes, it was natural enough to think of it as a kind of simulator to imitate what was already happening on paper. So in accounting, for example, you got these massive spreadsheets that basically just simulated the ledgers accountants were working with at the time. In its way it was marvellous because it meant you could use formulas to add stuff up and automate a lot of the work that accountants previously had to do manually. Even so, it didn’t take long to figure out that working with your data in its final form really restricted the way that you could manipulate and control the data itself.

From there you got the database-driven systems. These offered best-of-breed solutions in very specific areas, so you had your human resources programs and your accounts receivable programs and your inventory programs. But while they were really good for what they did, you had to integrate them all if you wanted them to talk to one another, and eventually you ended up with a strip-mall of products from all sorts of different places. Not only did businesses have to spend a lot of money on integration consultants but they also tended to get stuck with the basic versions of the platforms because they’d put so much work into integrating them that upgrading just wasn’t worth it.

To read the rest of the article, please visit <>