On Switching CAD Products

CADinfo.NET has recently received reader correspondence lamenting "what the articles don’t say" about the design-efficiency claims associated with moving to a different CAD program. A common thread in the correspondence is that a change is probably a complete waste of money unless you also invest training on the new product as well. Perhaps in many cases all that is really necessary to achieve performance gains is better training in the system you already have. Geoff Harrod ventures an opinion.

There is no doubt that good training pays dividends. But it is too often seen by managers as unnecessary expense or even as expensively subsidized ‘time off’. But CAD training is often provided at the wrong time in the trainees’ work experience. It is often shoehorned-in when there is a lull in productive workflow -- rather than provided at the outset of a person’s introduction to the software. Immediate training is also vital to extract full benefit from major product upgrades.

If training is not provided at the outset, people have to fumble around and work out their own ways of operating, wasting time while doing so, and then possibly developing a method that is not as effective as could be. If training is provided later on, after they have developed some way of working with the software, they are likely to be rather resistant to any different way. They prefer to keep doing things the way they have always have rather than changing to perhaps more efficient work habits. In such cases, the training time and cost is wasted.

Machine manufacturers don’t expect their customers to "just figure it out" when they ship a new model, so why are the design office staff left to "just figure it out" when new software arrives?

Architects don’t expect their buildings to be thrown together and managed by untrained people. So why are designers and documenters in architectural offices left to "just figure out how it works" when it comes to using core design software applications.

Too much work – no time. No work – no money

There is never a convenient time to do training. When the work is coming in fast, there’s no time for training. When there is a slump in work, there’s no money for training, even though the people might be sitting around with nothing to do. Employing contractors solves the problem of paying people when there is no work. You just ask them not to come in. Then it becomes the contractor’s responsibility to give themselves training, and still the same two obstacles apply.

Training needs to be provided before new people get involved in the main workflow, and as a fundamental part of the process and budgeted cost of implementing software upgrades. Upgrades are always a pain in the butt. Just when you get the last upgrade working and sometimes even before that, along comes another "must-have" upgrade. Even assuming they don’t bring unforeseen bugs and down-time as side-effects of their great new features (and you should not assume that), there is inevitably a lot of disruption and cost involved in converting the software installations. Plus the desirable addition of training the users to take advantage of the new features or learn how to use a different interface.

One needs to be very sure that the upgrade will in fact deliver work improvements through its new features, and then to make sure that the users will know how to derive those improvements, which means providing training. In too many cases, neither of these criteria prevails.

Often, CAD software upgrades are bought and implemented on a ‘no-think’ basis, causing disruption without benefit. In many situations this is done mainly to ensure compatibility with the latest version of the system’s data file format, for interchanging data with outside associates. That has, I think, been a major ploy adopted many big software companies to force upgrades.

In the past eagerness to upgrade was also driven by a desire to obtain improvements to obvious deficiencies in the software. But now, most of the major software products, not just in CAD, have incrementally improved to the point where there is little scope for worthwhile new features. As for or bug fixes, we should not have to wait for and be required to buy the next upgrade to have these rectified. Eagerness to upgrade has waned, and there is much more careful consideration being given to what real benefits the proposed upgrade promises. Upgrade aversion is a clearly evident and is giving software companies a bad time. It also coincides with an apparent market saturation evidenced by low new unit sales. The result is a sudden and significant drop in software company revenues.

Developers have responded in various ways: Enforce an upgrade revenue stream by requiring subscription or rental; promote hyped-up new features that are really not much use to anyone; or try sneaky tricks to force upgrades. The improvement potential of CAD software generally has not yet plateaued to the extent that  general office software has, although in the area of 2D drafting, it probably has. There is a big push to persuade users to move to radically different design methods, such as 3D modeling and simulation. Fortunately for the software makers, there are real benefits to be gained by the users from shifting the design paradigm, but it also presents considerable challenges in implementation and re-training.

The Unique Challenges of AutoCAD

I spend a lot of time training AutoCAD users. AutoCAD presents some unique challenges because of its long development history from humble beginnings. Its user interface and features have undergone drastic changes over the years, but generally, Autodesk has nrarely removed any of the features or controls that have been replaced by new and better ones. The result is that there is often a plethora of possible ways of achieving any one particular result. When teaching upgrade classes, I have found that many ‘old hands’ just don’t want to know about new methods. It is common to find people who have paid for every upgrade on offer, but never use any command or technique that wasn’t present in the version they started with! Presumably they only bought the upgrades to keep compatible with the latest DWG file format. That’s expensive compatibility!

It took a very long time to convince people to use the PaperSpace system, although admittedly it was excessively obtuse until recently. With AutoCAD 2000, I find that very few existing users want anything to do with the new Lineweight facility or the plotting options. Those who are completely new to CAD are more likely to make use of the Lineweight property, and those converting from other CAD products wonder how AutoCAD users ever managed without it. (Lineweight has been a long-standing feature of every CAD system except AutoCAD).

AutoCAD 2000 also introduces a right-button context menu system that is very helpful and productive. But existing users often want to turn off the right button menu system because it interferes with their accustomed use of the right button to do the same thing as the enter or key or space bar. A similar attitude applies to Grip editing. With this sort of resistance to new features, the value of upgrades, and even of training in them, becomes very dubious.

Switching for Productivity

Switching to a different CAD product, especially one embodying an entirely new design approach, obviously involves much more re-training and disruption than a version upgrade. But in the case of shifting from a 2D ‘electronic drawing board’ method to a 3D modeling method, the benefits of the change of design methodology are very likely to far outweigh the dislocation involved, at least in the long run. It is a similar situation to the shift from pens-and-paper to CAD, only more so, because then, the new CAD method (2D) emulated the paper method.

I think the claims of productivity and design quality benefits being made for switching to various CAD products are all about switching design methods from 2D o 3D, rather than from one CAD product to another. I’m sure there could be no significant benefit in switching from 2D drafting in AutoCAD to doing the same in MicroStation, or vice-versa. In fact there could only be period of depressed productivity while the people re-learned the system and re-gained speed of operation. There are always cases where a product change may be necessary or even enforced (for say, client compatibility). Furthermore, switching to another CAD system is not difficult for an experienced CAD user, but the switch alone can rarely be said to confer any productivity improvement.