Columbia House closed because its model depended on customer inertia. It lured members with a free or penny introductory offer, then shipped full-price records unless they said no. That negative-option design angered customers, digital music broke its physical-media moat, and regulators now restrict the tactic. Quartz reported it would stop accepting orders on 15 September 2026, after 71 years.
Key takeaways
- Columbia House peaked at $1.4 billion in annual revenue (1996) on its 'penny for a dozen' pitch.
- It stopped accepting new orders on 15 September 2026, ending 71 years of mail-order club sales.
- The demand for music never went away. What died was a model that made money when customers forgot to cancel.
- Negative-option billing is now a legal risk in the US, UK and Australia, not a growth tactic.
Picture the ad. A dozen albums. One cent. Just tick the boxes and mail the coupon.
Millions of people did. Then the next record turned up, at full price, because they hadn't sent back the little card saying "no thanks."
That was Columbia House. For decades it was one of the most successful subscription businesses in the world. In September 2026, after 71 years, it stopped taking orders.
I used Columbia in Chapter 1 of The Subscription Playbook as a warning. Here's how the story ended, and why it matters to anyone selling a subscription in 2026.
The numbers: a $1.4 billion business that ran on "no"
| When | What happened | Source |
|---|---|---|
| First year | 128,000 members | The Subscription Playbook |
| 1963 | More than 7 million records shipped a year, about 10% of all US record sales | The Subscription Playbook |
| 1996 | Peak revenue of $1.4 billion a year | Quartz |
| 1990s | Clubs like Columbia made up about 15% of US record sales | The Subscription Playbook |
| 20 Aug 2026 | Quartz reports the club will close | Quartz |
| 15 Sep 2026 | Columbia House stops accepting new orders, after 71 years | Quartz, Fox61 |
How did Columbia House make its money?
The model had three moving parts.
- An irresistible intro offer. Twelve CDs (or records, or cassettes) for a penny. People love free, and this was close enough.
- A commitment. In return, members agreed to buy more albums at full price over time. That's the minimum commitment model, in consumer form.
- The negative option. Each month a featured album shipped automatically unless the member sent back a card declining it. Miss the card, get the album, get the bill.
Parts one and two are fine. Plenty of good businesses use a generous offer to win a customer and a commitment to keep them.
Part three is the problem. It made money in proportion to how little attention customers paid.
Why did Columbia House fail?
In the book I answered that in a word: greed.
That's a little unfair (but only a little). There were really two failures.
It traded trust for acquisition. Columbia focused on signing up new members, not on looking after the ones it had. People signed up for free records without realizing what they'd committed to, then got angry when full-price albums arrived. As I wrote in 2022, Columbia "actively made its subscription process less transparent than it should have been." Angry customers churn, and they tell their friends.
It didn't protect its moat. Columbia's advantage was convenience in a world of physical media: a huge catalog, delivered to your door, without a trip to the record store. MP3s, file sharing and then streaming took that away. Once music was one click away, the club's only remaining edge was inertia.
In the MOATS framework, Columbia had Marketing supremacy and a big Audience. It had no real Switching cost (nothing was lost by leaving) and its Toughness disappeared with the CD.
The inertia test
If your revenue would fall sharply the day every customer started paying full attention, you have a Columbia House problem. Regulators call it a subscription trap. Customers call it a rip-off. Either way, it has an expiry date.
What the book said in 2022
I called the Columbia story "From $1.4 Billion in Annual Profit to Complete Collapse." Two corrections.
First, the $1.4 billion was revenue, not profit. Quartz puts the 1996 peak at $1.4 billion a year in revenue.
Second, the "complete collapse" took longer than I implied. The club was still limping along when the book came out. It finally stopped taking orders on 15 September 2026.
The lesson I drew still stands: "the subscription model isn't infallible. Bad practices and an inability to adapt to a market's changing needs can (and will) result in a shrinking moat to the point where it isn't even there anymore."
What's happened since
Three things.
It closed. Quartz reported in August 2026 that Columbia House would stop accepting new orders on 15 September. A relaunch as a vinyl subscription club was floated, but according to local news coverage of the closure, it never happened.
The demand moved to streaming. People didn't stop loving music. They moved to services like Spotify, which now has more than 300 million paying subscribers on a model with no default shipments. See how Spotify turned its subscribers into a profit.
The tactic became a legal risk. Negative-option billing has been restricted for years, and in 2026 it is squarely in regulators' sights in the US, UK and Australia. In Australia, the Unfair Trading Practices Act 2026 brings in subscription-specific rules on disclosure, ongoing notices and easy cancellation from 1 July 2027. Read the Australian subscription trap laws guide for what that means in practice (general information only, as at October 2026).
And the irony: Columbia's ancestor, Book of the Month Club, celebrated its centenary in 2026 with a model where members choose their book instead of receiving one by default. Read how Book of the Month made it to 100.
Is your sign-up flow a Columbia House?
Check your sign-up, renewal and cancellation flow against Australia's new subscription rules with the free AU Subscription Compliance Checker.
What you can steal from Columbia House (the good parts)
Columbia wasn't all bad. The intro offer was brilliant marketing. Keep that and drop the trap.
- Audit your offer for inertia. List every way your revenue depends on customers forgetting: auto-renewals with no reminder, trials that convert silently, cancellation buried three screens deep. Redesign them before a regulator or a competitor does it for you.
- Keep the irresistible intro offer. A first box at cost, a first month for $1, a free first service. Pair it with terms a customer can read in ten seconds.
- Remind customers before you charge them. Yes, a reminder costs a little churn. It buys a lot of trust. Use the renewal reminder email templates as a starting point.
- Make leaving easy. A clear, one-step exit is now expected, and in Australia it will be required. My guide to cancellation flows that save customers legally shows how to offer a pause or downgrade without trapping anyone.
- Replace the negative option with a positive one. Instead of "we'll ship unless you say no," try "here are this month's picks, choose one." It's how Book of the Month survived.
Steal this: the penny offer, done honestly
Run a "first month for $1" offer with three lines under the button: what they'll pay after, when they'll be charged, and how to cancel in one click. Then email a reminder three days before the first full charge. You keep the pull of the offer and lose the backlash.
Your next step
Open your own sign-up page and read it as a tired customer at 11 p.m. would.
Would they know exactly what they're agreeing to? Could they leave as easily as they joined? If the honest answer is no, fix that first. Then compare your design with the eight subscription models to see whether a commitment, a box or a membership suits you better.
If your subscription only works when customers forget about it, it doesn't work. It just hasn't failed yet.
Frequently asked questions
How did the Columbia House 12 CDs for a penny deal work?
New members got a dozen albums for a penny (or free) and committed to buying more albums at full price later. Each month a featured selection shipped automatically unless the member sent back a card declining it. The intro offer was the bait; the commitment and the default shipments were where the money was made.
When did Columbia House close?
Quartz reported in August 2026 that Columbia House would stop accepting new orders on 15 September 2026, ending the mail-order club after 71 years. A plan to relaunch it as a vinyl subscription club had been floated but, according to local news coverage of the closure, never happened.
Was Columbia House a scam?
It wasn't illegal in its heyday, but many customers felt tricked. The terms were disclosed, yet the design relied on people not reading them and forgetting to return cards. That gap between what was technically disclosed and what customers understood is exactly what today's consumer regulators focus on.
Is negative option billing legal in Australia?
As at October 2026, Australian Consumer Law already bans unsolicited supplies and misleading conduct, and the Unfair Trading Practices Act 2026 adds subscription-specific rules on disclosure, notices and easy cancellation from 1 July 2027. Whether a particular design is lawful depends on the details, so get legal advice before relying on any default-shipment model.
What replaced Columbia House?
Streaming. Spotify and similar services offer unlimited access for a monthly fee, with no default shipments and no minimum purchase. The demand Columbia House served is larger than ever. The difference is that customers pay because they use the service, not because they forgot to cancel it.
Get both books free
The Subscription Playbook (ebook + audiobook) and Feed a Starving Crowd, instantly. In Australia? Claim the free hardcover.
Robert Coorey, MBA
Robert is the author of The Subscription Playbook and Feed a Starving Crowd and a Director at BPAG Accountants in Sydney. More about Robert.
Sources
- Quartz, Columbia House mail-order club shutting down in September 2026 (20 Aug 2026)
- Fox61/TEGNA, Columbia House is shutting down for good after 71 years (Aug 2026)
- Wikipedia, Columbia House (negative option billing background)
Facts and prices checked 27 October 2026. Spotted something out of date? Tell us.