If you aren’t sure how a certain feature works, and nobody else in your company knows, and there’s no way to find out one way or the other, here’s a radical idea: just leave that feature out of the documentation. The same goes for product specs that nobody can certify as accurate. Simply put, it’s more damaging to give your customers the wrong information than no information at all. In the former case, you’re basically lying to them, but in the latter, you’re inviting them to ask for clarification and giving yourself a chance to get it right when you respond.

For the past six months, my local Target has dedicated an endcap display to foldable utility carts. This sort of thing can come in really handy in the city: you take the folded-up cart with you to the store, unfold it during checkout, put all your bags in there, and wheel it home. And since few people have one, promoting them in the store should generate quite a few impulse purchases. But there’s one problem: the folding carts are sold unassembled, making it very difficult to buy the product and use it during the same shopping trip.

I’ve actually thought about buying a cart the last two times I visited the store. The first time, I saw that the wheels aren’t attached in the package, and figured that it would be hard to assemble. The second time, I actually looked at the instructions on the back, which confirmed my fears. As soon as I saw you needed a pliers and glanced at the illustration where you’re installing some sort of spring mechanism, I knew this was more than I’d be willing to fuss with on the way home. I’m sure other potential buyers have felt the same way.

Considering that the fully-assembled cart folds nearly flat, the space required to store ready-to-use models would be about the same as the partially-assembled ones they sell today. Thus, I’d love to see Target experiment with offering fully-assembled carts for say $5 extra. Plus, this opens up a whole new world of marketing opportunities for the product. I can envision today’s boring endcap display transformed to show a before and after scenario. On one panel, you would see a customer struggling to carry all their bags home, and on the other panel, you’d see the same customer wheeling all those bags home with their shiny new cart. If buying the cart meant that level of instant gratification for the customer — rather than just promising some nebulous future benefit once you put it together — I bet sales would skyrocket.

I’ve been thinking about the time-saving features that I take for granted in everyday products. Here’s a few that come to mind:

– The way my BlackBerry remembers which email account I used to send a message, and selects it automatically next time.

– The “Cancel” button on laser printers that instantly deletes the job, without having to mess with the print queue on my computer.

– The “Close Door” button in elevators, even if it’s usually too slow to stop the guy with too much cologne or the woman with a rabid child from sliding in at the last minute.

I wonder how long it took each of these features to appear, and whether it was due to customer requests or just clever innovation on the designer’s part.

While reading restaurant reviews online, I noticed something about how I evaluate other people’s opinions of a business. Specifically, I would much rather go to a restaurant or other establishment that consistently rates four out of five stars, rather than one that gets equal numbers of threes and fives. In other words, when comparing places that score the same on the average, I always prefer the one that has the least variation in the individual scores.

I think this behavioral trend is widespread. For me, the desire to have a predictably good experience means that I don’t try new places that frequently fall victim to off nights. After all, if you only visit a particular restaurant once in a while, or you’re going there for a special occasion, do you really want to end up with a two star or three star experience at the usual four or five star price? Unless there’s some way to compensate you for the experience falling short, the risk averse part of us says to pick the place that’s more likely to come through consistently.

All else equal, I believe that customers prefer a predictable level of quality. In other words, you should set realistic expectations for your product or service, and then meet them on a regular basis. Ironically, if you sometimes exceed these expectations by a wide margin, it may contribute to such glowing reviews that future customers are more likely to be disappointed. Then, unless you adopt a policy of compensating customers who show up during an off night, you’ll end up with more disappointed reviews than if you had simply delivered a consistent product in the first place. In short, it’s best to deliver a solid experience as often as you can — while minimizing the daily ups and downs that frustrate your customers.

One of my coworkers asked me to look over the contract from a coffee service that we’re considering. We’ve actually had this sort of thing for years, where they lease you the machine and you pay for the coffee that gets delivered every few weeks. But our old provider nearly tripled their delivery fees in January and refused to negotiate, so it was time for a new vendor.

Looking at the contract from the new service, I was pretty shocked. Among other things, they reserve the right to forcibly enter your office to reclaim the machine if you don’t pay your bill. All this for a piece of hardware that costs no more than $250. Needless to say, we told them their contract is absurd and way out of line for something as simple as coffee delivery. So who did we buy from? Amazon.com sells the same coffee for almost 1/3 less, enabling us to buy the machine (and even purchase spares) with the money we save. More importantly, there’s no onerous contract to deal with.

When I think about it, the coffee vendor violated a fundamental rule of sales: they’re not making it easy for customers to do business with their company. For what should be a very simple transaction using a brief order form with some basic terms and conditions, they dump a lengthy contract on the buyer. You might argue that they’ve gotten burned before, so they’re just trying to protect their interests. Ok, but at what cost? If your worst-case scenario is losing a few hundred dollars in hardware, and it only happens 1% of the time, do you really want to put up roadblocks that scare off say 20% of your potential buyers? For most companies, I’m pretty sure that’s a losing proposition.

I almost made a fool of myself while joining a conference call yesterday. Things started off normally: the automated prompt asked me to enter the number assigned to my call, and then it asked me to say my name after the tone. But instead of the usual “beep” or “ding”, it played a bizarre, almost circus-like noise. Trust me, it was weird. I actually began laughing out loud, but managed to regain my composure before I had to say my name for the system.

So, what’s the big deal? Well, the name you record on these things is either played back to the call moderator, the other participants on the call, or both. I really don’t want to be the guy whose name is announced as a jumble of nervous laughter. It doesn’t really set the stage for a professional first impression. I’d much rather have other people hear my name and company instead.

Frankly, I don’t know whether the crazy noise I heard is due to a malfunction in the conference calling software, human error, or something else entirely. Some people might argue that the system should let you review your recorded name before it’s saved, regardless of the type of beep you hear first. I think this would add too much complexity to what should otherwise be a simple process. The right solution is for the folks who maintain the conferencing system to test it out more frequently, so they can see what callers are experiencing. Ideally, they would compare each test to a recorded copy that shows how things should sound, so there’s no mistaking when the sounds or other parts of the process have changed.