So in the 1970s the Billboard Hot 100 reflected wholesale numbers, not retail sales? But...but...Disco Duck and The Night Chicago Died each went to number one and each went gold. They had to have been bought by at least a million "paying customers"---they just had to be. Come to think of it, Rick Dees used to joke that only seven people bought his Disco Duck album. Maybe he's right!
The single "Disco Duck" was #1 on the Hot 100 for one week. The album, "The Original Disco Duck", peaked at #157 on the Billboard Top 200 LPs list.
Seven people is comic exaggeration, but maybe not by much. I posted a lengthy explanation of this on the Classic Hits board nearly two years ago. I'll try to distill it now, but there's quite a bit to it, and if you want to know, you have to read. Given the passion level on this board, that shouldn't be asking too much.
Billboard was not originally intended to be a mass-market magazine. It was a trade paper, aimed first at vaudeville and other live theater, and eventually branching out into coin-operated amusement devices, radio, television and records.
The idea behind the record charts was two-fold: To give retailers an idea of what was hot so they could order accordingly (only a handful of very large stores could simply stock everything)...and to give record labels the incentive to advertise in the pages of Billboard to those retailers who would drive up those chart numbers by ordering and re-ordering more copies of those very records.
But how to determine those charts? Calling every store that sold records in America would be prohibitive both in terms of time and money (we paid for long-distance back in those days). Calling a "representative sample of stores" would still be time-consuming and expensive, just less so. But having a direct connection with the labels (to whom you were trying to sell advertising every single week) would tell you raw numbers...how many copies were getting shipped out every week. So throughout the 1960s and until the fall of 1981, the Hot 100 Singles chart was based on wholesale sales, as was the Billboard Top 200 LPs.
The trouble is, it couldn't tell you how many of those copies wound up in the hands of paying customers and how many went unsold. Labels generally had return policies...90 days after delivery, a store could send back unsold 45s for credit. Six months for albums. But there was no chart in Billboard for what was sent back. They did report returns, in general numbers as an industry total, in small news stories buried in the middle of the magazine, but those didn't discuss specific titles. There certainly was no re-calculation of the chart (and no point to that, since the chart was all about guiding the retailer at the moment and selling advertising), and nobody was taking back Gold and Platinum records.
The ultimate example of that was the soundtrack of the 1978 movie "Sergeant Pepper's Lonely Hearts Club Band". RSO shipped out one million copies of that in the first week. It debuted at number one and earned a Platinum record in its first week. But at retail, it stiffed. The returns were legendary. The industry joke was that it was the first album to ship platinum and return double platinum. RSO was never going to divulge the real numbers (no label ever did), but the prevailing wisdom was that fewer than 100,000 copies were sold at retail.
That's the most notorious case, but a smaller version of that was playing out on the charts every week for 20+ years before. Once labels figured out the system, all they had to do was over-press and over-ship at wholesale. That produced a higher chart number, giving retailers confidence to order and get the inventory out of distributors' warehouses and into retailers' stock. If that (and promotion) resulted in radio play and people liked the record, then it might even sell in the stores. Maybe. If not, the label and the wholesalers have the retailers' money, and returns are paid off with credit toward future orders, not cash, so the label still comes out ahead.
So what did those chart numbers mean?
A record's first week on the Hot 100 wasn't an indicator of how many people bought the record, but a snapshot of how many wholesale copies had been shipped to distributors. So debut numbers were virtually useless. The next couple of weeks worth of chart action were stores ordering first-time stock.
If a record peaked at #50 or below, that wasn't a "Hot 100 hit", it was a record that most record stores didn't think was worth stocking (or re-stocking beyond their initial 5 or 10 copy order). #40 wasn't much better, nor #30. Those basically indicate that more (but by no means all) stores placed an order, but doesn't suggest that demand was high enough for them to ever have to re-stock.
Above that, it's important to look at where a record peaks with this in mind: Let's say a record moved 25-20 last week (again, we're traveling back in time to the 60s-80s...pre-Soundscan). That suggests that stores were optimistic about the record and bought copies. But what if it slides to 24 this week? That means the optimism on the part of the record stores came after the record had already peaked with the record buyers. They were ordering stock when it was at 25 expecting it to go Top 15 or higher. But it went the other way. That "Top 20 hit" was actually a #25 wholesale record that became overstocked compared to buyer demand that week.
Was the Hot 100 in any way reflecting the true popularity of records?
Well, any record that made the Top 10 by climbing over a period of several weeks probably was legit. Those chart jumps and bullets ("Stars", as Billboard called them) were indicative of more orders for a record being made this week than the week before. Sometimes that could be accomplished simply by more stores placing first-time orders...but the longer the timespan, the more likely it was that the stores were replacing stock that they had sold to paying customers.
Even records that debuted big fast (a new Beatles, for example) were probably legit if they stayed high on the charts for a certain period of time. But a rapid fall-off from a high chart number suggests that big peak was record stores anticipating demand that wasn't really there.
Which is why you've seen me say that anything peaking under #15 really isn't a hit....and a lot of things that peaked between #11 and #15 weren't, either (Royal Scots Dragoon Guards "Amazing Grace", anyone?). Because of what the Hot 100 measured (store owner optimism), a record that cracks #20 is a record the store owners believed could go Top 10. Those that didn't under-performed expectations. They weren't "gotta haves".
There's one exception to that rule: There are some records that were strong performers in certain markets, but it just didn't translate nationally. Tower of Power's "You're Still A Young Man" was a huge record in Los Angeles and San Francisco....Top 5, in fact. But it peaked at #29 in Billboard. Not enough stores stocking or selling it elsewhere (TOP was a California band).
Again, none of this was ever supposed to matter to the public. But when American Top 40 needed a chart for its weekly countdown, it settled on Billboard (largely for convenience sake...Billboard and Watermark were a mile away from each other in L.A. and they could get the numbers as soon as they were tabulated).
Billboard was unknown to the general public until its weekly exposure in AT40, and it was slow to capitalize on that. Eventually (after 1980, IIRC), Joel Whitburn began publishing mass-market chart books you could buy at mainstream book stores ( his Record Research books had always been mail order), Fred Bronson followed suit, and the Billboard branded greatest hits CDs were released.
It was based largely on the public's misunderstanding of what those numbers really represented. Billboard never said they were accurately charting actual retail sales, but they never corrected the impression, either...and furthered it by marketing products based on their charts to the public at large.
If Billboard (and Casey Kasem) had gone out of their way to say "The Hot 100 chart represents wholesale shipments of 45 RPM records to distributors and record stores and is not indicative of actual retail sales. Such sales may not occur, in which case the records are returned to the label. No re-accounting of the chart to reflect the actual sales of the record, which may be substantially less than the number shipped wholesale, will be done and no adjustment made to chart rankings as a result", on every AT40 broadcast, Whitburn or Bronson book or CD compilation, I doubt the Billboard chart would be anything the public would have paid much attention to.