There is a number that gets quoted in almost every marketing meeting, and almost nobody in the room has ever checked it: 80/20. Eighty percent of your sales come from twenty percent of your customers. The loyal core. The superfans. The people who really get what you are building. Budgets get built on it, whole retention teams get hired on it, and it feels too tidy to be wrong.
The people who did check it work at the Ehrenberg-Bass Institute, the marketing science group at the University of South Australia, which has been working through real buying data - hundreds of product categories, a number of countries, decades of panels - since long before anyone called marketing "data-driven." In December 2019, Byron Sharp, Jenni Romaniuk and Charles Graham published a short paper whose title is itself the correction: "Marketing's 60/20 Pareto Law." The abstract opens by killing the folk version outright: "It's wrong to talk about an 80/20 law in marketing." The heaviest fifth of a brand's buyers, they report, "generally contribute not much more than half of a brand's sales."
And the other side of that split is where this whole article lives. "It's reasonable to expect that almost half of your brand's sales will always come from your very lightest 80% of buyers" - the people who buy you once or twice in a year, who would walk past your stand at a trade fair, who do not follow you and are not about to. Almost half the money arrives from the shallow end, from people who barely know you exist.
Loyalty follows size
Ehrenberg-Bass is famous for a pattern called double jeopardy, and it is the backbone of the book that collected these laws, How Brands Grow, published by Oxford University Press in 2010. Ehrenberg, Goodhardt and Barwise described the pattern in the Journal of Marketing in 1990: "a small brand typically has far fewer buyers than a larger brand. In addition, its buyers tend to buy it less often." Small brands get hit twice. And the pattern repeats across categories, countries and decades steadily enough that the institute calls it a law rather than a finding.
Read it as a marketer and it lands somewhere uncomfortable. Loyalty is not a cause you can work on and then watch sales follow. It is a shadow of market share. When critics accuse the school of claiming loyalty does not exist, Sharp's answer is the book's own page 92: "Brand loyalty - a natural part of buying behaviour. Brand loyalty is part of every market." Loyalty is real, it is everywhere, and precisely because it is everywhere it varies only slightly between rival brands. Big brands do not have wonderful customers while small brands have defective ones. They have nearly the same loyalty numbers, at different sizes.
The top fifth will not stay put
The second pillar sounds like a joke until you sit with it. Track one brand's buyers across two years and the roster of heavy buyers churns underneath you, even when nothing changes and nobody defects. The 2019 paper states it plainly: "even for stable brands half of last year's heavy buyers will then not even qualify to be in the top 20%, while the people who were light or non-brand buyers last year will contribute more to sales this year than they did last year." Regression to the mean, wearing a lanyard. The superfans you are courting are a rotating cast playing the same roles.
So where does growth come from? Headcount. Comparing brands that rose with brands that slid, Sharp's group reports that "brands that grew or lost share mainly experienced large change in their penetration," and a 2014 paper in the Journal of Business Research found sales and profit movements tracked "over or under performance in customer acquisition, not performance in retaining customers." Or as Sharp compresses the whole program into one line: "Growth comes from nudging everyone's propensity up just a little bit."
The shallow end has a listed price
This is why I think paid position is the most underrated line in small-team marketing, and why this site's whole product is a serious answer to that finding. The front page of Bid Spotlight is a single ranked table: bid more, sit higher, every placement labeled. The people scrolling it are browsing, and on Ehrenberg-Bass's arithmetic browsers are mostly non-buyers and light category buyers of whoever they end up finding - the exact population the data says growth comes from. You cannot buy loyalty at any price. You can buy reach into strangers at a listed one.
It is reach you can count, too. Every outbound buyer link on the board ships with rel="sponsored noopener" - a combination that does not suppress the referrer - under a site-wide strict-origin-when-cross-origin policy, so a click off the board lands in the buyer's analytics labeled with the bidspotlight.com origin. No attribution modeling required: the day's arrivals show up as a line that says where they came from.
One honest limit, because double jeopardy does not stop at the checkout. Position recruits; it does not retain. Getting bigger is what lifts loyalty metrics, not the other way around, and nothing about buying a rank improves your product. What a slot buys is the introduction. What happens after the click, on your page, is the half of the deal no leaderboard can sell you - but the introduction is the half your growth depends on.
The growth is in the shallow end - introduce yourself from the top of the board at bidspotlight.com/submit.