First, a caution. You’ve heard me mention Amory Lovins and his assertion that people want cold beer and warm showers. That still holds. Energy savings may perk up interest, but don’t except harping on energy savings to drive endless business your way. Delivering on your promise, though, it always a good thing.
For some background, read how Perry Bigelow has been doing this in Chicago for more the 20 years. As you’re reading note the mechanics. But don’t skip these paragraphs, either:
Bigelow insists that energy efficiency is not what sells homes. He does not emphasize the construction techniques or the energy features to his sales staff; he doesn’t want them dwelling on this angle, because saying where you put what type of caulk or insulation isn’t what generally motivates a buyer to buy. What the heating-bill guarantee does do is generate traffic.
The guaranteed bill is the value that sets us apart, says Bigelow. Home shoppers who notice the guarantee in our ads say `I can’t afford to not go look at what that guy is doing.’ If they come, and they like what they see, they buy. If they like what they see, our guarantee and the benefits from our energy package give them rational reasons to make their emotional buying decision.
Steve Andrews in Home Energy Magazine, 1994
Now to the simplified mechanics. If you’re going to offer a guarantee, you have to be able to predict the savings the savings at some level. Modeling nerds want to spend hours tweaking models to do this. High performing modelers can get their models to within +/- 10% of the actual result. I’ll take it at face value that those claims are backed with sufficient comparison to actual results, and that they mean what they say and thus they never have outliers in the +/- 25% range or beyond. That’s great! Even there, I wouldn’t offer a guarantee though, where I expected 50% of the people to fall below my prediction and then point them to fine print saying “I meant within 10%.” If you’re modeling at that level, I’d recommend guaranteeing at 90% of what you predicted back at the office. Now, since you never deliver less than 10% of your prediction, you will always hit your guarantee. You can back that with not just works, but cash! Even better, almost all of your clients will be pleasantly surprised that they saved more, and that’s a good thing.
You don’t have the actual utility bill data to prove that you’re estimates are +/- 10%? Well then you can still make the guarantee, but you have to be much more conservative until you gain the experience and gain the data.
Perry Bigelow did this in the early 1990s using models that were not as “good” as the ones we have now (they’re still are not very good at the individual house level–read more on this from Michael Blasnik). Bigelow did this by being very conservative and promising less than he thought he could deliver. He even got it wrong at the beginning and had to adjust. You should be prepared to do the same.
I’m using Perry as an example, because he’s the guy that first caught my attention with this. But there are others on the new construction side, and even on the existing homes side, who’ve taken very similar approaches.
A couple of further observations. Modeling isn’t free. And the better you want the crappy model to be, the more time you have to put into it. Unless you’re living off a trust fund and just enjoy doing modeling, time is indeed money. Either you or your customer pays for this. Your customer might like a new window instead of your model. Or a half dozen CREE CR6 fixtures. Or beer. OK, some actually might want the model and want to pay for it, too. But I can assure you, not everyone does.
Rather than using expensive modeling, you can use simplified models, with wider error bars and simply be more conservative with your promise. If you promise a bit low and deliver at a greater savings rate every time, then you still have at least a 100% realization rate.
“Why would I want promise a lower amount? That’s insane.” Well, no, it isn’t. You only want to promise what you can deliver. Reread the mechanics above. As you gain experience and data you can refine your predictions, and you can dial in your guarantee better. BTW, you need that same experience and data to make any claims about your intensive modeling, too. (Some folks have assured me their models are within 10% everytime–but they have no data to show it. Well then, how do they know??) In the meantime, the question is what is the marginal value of the few additional BTUs that you can squeeze out to offer a guarantee at a higher level. And accurate precision comes with a cost. If you think the value is there, go for it. If you think you can offer a better guarantee than someone else, go for it. That’s called differentiation, and it may offer you a competitive advantage (and it may not–you’ll have to market test it). You can decide is the cost is worth it. But claims that guaranteeing a lower amount is insanity is nothing but bluster.
You can guarantee savings. Be conservative and deliver on the guarantee.