Perhaps this explanation might make it a little more intuitively obvious:

If something has a 1% chance of happening when you do a thing, you have to do that thing 100 times (on average, of course) to make it happen once.

If it has a 2% chance, you only have to do the thing 50 times.

If it has a 3% chance, you have to do the thing 33-ish times, which is better than 50, but not as much better than 50 as 50 is better than 100, and this trend of 1% more being less of an improvement continues as the base rate increases.

That is an excellent point. My concern lies with taking the mantra blindly without knowing where you should focus your attention. So moving beyond this point, when we know that 1-2% gains are significant, where should we focus on making them?

Should it matter that we even have 1-2% gains, or that we're putting our focus on what is more important? I would say that if you're pursuing a greater benefit for your customers or your product, the only thing measuring the gain does is reaffirm that what you chose to do is working.

To your examples, if your goal is better patient health, faster arrival times, or whatnot, do you look to the percentages as goals or as reaffirmations that what you're doing is working?

Actually, Steven Covey covers this exact point. He talks about "watering down the chowder"... I won't tell the whole story, but Google it if you're curious. It's pretty good. Long story short - I agree with you, I think cuts to product quality tend to undermine businesses in the long run, despite the short term gains.

But 1%/2% isn't just about cutting expenses. A physician with a slightly better sanitation/preparation checklist saves more lives, airplanes with 1% more efficiency burn way less gasoline (less cost and lower pollution), and so on, and so on. So yeah, you've brought up a good point to be aware of, but small numbers apply broadly to many things - not just cutting costs.

Good comment, Shanna, thanks. Agree on people especially.

> I'm curious as to why you think you're not a good 'splainer.

I re-read the post myself, and I'm not sure that someone who didn't know the concept beforehand would understand it on one read. Any reader here would probably understand it in a few reads, but that's because they did the work to understand it. Ideally I could have explained it more clearly to guarantee reader success, but this post was the best I could have come up with.

I get it, but it's also one of the same things that grates on me as an excuse to make terrible business decisions. You can make great ones with it, sure, but my brain tends to think the people using this strategy are doing really stupid things with it. It's the reason that Ben and Jerry's Ice Cream will eventually cease to have buttermilk as an ingredient, why fewer and fewer people will qualify for heatlhcare, and why big business cease looking at innovation as a means to further change.

For example, let's say I'm in a meeting with a bunch of executives and say "I can save us $$$$$ over X years by decreasing our buttermilk content 1% from 6% to 5%, replacing it with cheap pseudo-chemical food product Y. That's a 20% savings on buttermilk!" Hooray, +1 for you, your own corner office, etc. You stay a couple years in a posh corner office, leave, and nobody's the wiser in proposing a couple years later "why don't we decrease the buttermilk content?" like you never did it before...Never mind the decreasing quality of a product when your kids eat ice cream and think "You used to love this ice cream, Dad?" "Back in my day, it was better..."

Let's now again imagine I'm an actuary working at a Healthcare company. I find that families with X kids tend to have a 20-30% increase in use of insurance (not true, but play along). I propose that we (the company) offer all customers a tracking badge that lets them deduct 5% off their premium payments just for wearing it. Now, using the badge's tracking data (GPS, health stats, etc.), I can now pick apart any subset of customers that I find might save us 1% here, 2.2% there. Change the contract for them (again) and they never notice the little addition stating the right to deny coverage for people with your unique(not really) situation, and eventually whittle down to just people that can afford our premiums, never get sick, and don't cost us anything. Hooray, another corner office for Mr. Winner over here.

To point #3. US Patents have seemingly gotten a bit out of hand at the moment, with companies patenting anything they can think up, so you get things like Amazon's One-Click patent and the recent Lodsys patent case, where SMB's are literally scared out of business for fear of even having to deal with patent litigation based on silly patents like ”A Method and System for Communication, Advertising, Searching, Sharing and Dynamically Providing a Journal Feed”, such as RSS, Atom, or feed reading technology in general.

So while, yes, it is mind-blowing that going from 1%-2% isn't about the difference in percentage but the velocity of change, I would hate to live my life thinking like that as a primary driver. The forest and the trees and all that. Blindly following this as your mantra or gestalt( :) ) would end up with the same mentality that results in companies and the people working for them to nickel and dime people until they break, all the while thinking "I didn't do anything wrong, honest".

I'm curious as to why you think you're not a good 'splainer. (Especially since you've evidently reiterated this lesson a lot.) People like big, splashy changes. Which is fine, as long as they don't discount small changes. Maybe you need to change their philosophy first (introduce kaizen). Or is it the math they have problems with? Because I'm really surprised at the number of people of people who don't understand what 100% more means.

I would also add that small number mean A LOT when those numbers represent people. People sell drugs that are only slightly better than placebo, and this is ethical because slightly better means that more people are given relief from their illness. People use alternative therapies that have not been proven because simply not working better than placebo is meaningless if *you're the one getting relief*. Never discount the people behind the numbers.