Sold (exit)

Bingo Card Creator: a record $64,791 year, 48 weeks on autopilot

A web app that makes printable bingo cards for elementary schoolteachers. In 2012 it sold $64,791.81 net of refunds and cleared $38,598 in profit, its best year ever. The owner spent roughly 15 hours on it plus 20 minutes a week of support. He sold it in 2015 under an NDA on the price.

Bingo Card Creator: a record $64,791 year, 48 weeks on autopilot
Reader

Isn't a record year usually the year the owner worked hardest on the product?

This year-in-review says the opposite.

What you'll learn
  • Seven years of books for a one-time-purchase bingo card app
  • Why the record came in 2012, a year spent mostly untouched
  • The Stripe bug that charged one customer 36 times, and the confession about support
  • A sale whose price can't be printed, plus the market rate and the work he did disclose

Bingo Card Creator finished 2012 with $64,791.81 in sales net of refunds and $38,598 in profit. It was the best year in its history.

The obvious guess about a record year is that it was the year the owner worked hardest on the product.

Patrick McKenzie (patio11) wrote down the opposite.

Bingo Card Creator spent roughly 48 weeks of 2012 in maintenance mode. His own labor amounted to about 15 hours integrating a new design plus 20 minutes a week of support.

In the same post he put it plainly: “I’m almost totally superfluous to the day-to-day operation of the business.”

Seven years of the books

YearUnits soldSales net of refundsProfit
2008815$21,141.60$8,823.06
20091,049$31,156.18$18,525.71
20111,451$46,233.68$23,230
20122,254$64,791.81$38,598
20131,734$50,156.16about $23,000
About the table

The 2010 figures don't appear in the four year-in-review posts this article draws on. The 2012 expense and profit lines were later revised to $29,145.20 and $35,676 in the following year's post. The monthly profit in our frontmatter divides the revised $35,676 by twelve.

The revision was flagged in advance.

McKenzie had opened the 2012 post by warning readers it was a good thing he was CEO rather than bookkeeper, because a bookkeeper that sloppy would be fired. Doing the official accounts for tax purposes, he wrote, almost always turned up a few thousand dollars of additional expenses.

It did.

The product itself is plain.

It generates printable bingo cards for American elementary schoolteachers to use in class. Visitors try it for free, and the ones who like it pay $29.95 once.

The price had been raised from $24.95 in 2009, and the one-time model never changed.

It launched in 2006 as a Java desktop application and moved to a Rails web app in the summer of 2009.

The record came in the year he stopped touching it

Two things happened in 2012, by his own account.

What worked
  • The application and marketing site were reskinned
  • Stripe replaced a checkout that had been running through PayPal and Google Checkout, letting the business charge cards directly

Both went in as A/B tests, and he calls the pair a smashing success.

The result showed up in trial conversion. The 1.8% of 2011 became 2.4% in 2012.

Trial signups rose only from 83,000 to 87,000. Units sold went from 1,451 to 2,254, a 55% increase.

Roughly the same number of people arrived. A different share of them got through.

The design work itself was done by someone else. That left him the Rails integration and a few tweaks.

The record year was the year he stopped building, not the year he built.

To put our own view on it, the decisive design choices in this business were finished years before 2012.

Google organic search supplied 56% of visits, AdWords 12%, and Bing and Yahoo together 11%.

Annual visits ran to 1.08 million and page views to 3.4 million. That volume kept arriving without additional effort.

What he added in 2012 raised the machinery on the receiving end by one notch.

The part that’s easy to miss is that the whole structure sits on six years of accumulation.

He’d written his own Rails A/B testing library back in 2009. His notes describe the method as a percent here and two percent there, which adds up when you keep doing it for six years.

The 2.4% conversion rate of 2012 is a balance carried forward, not the yield of a single year’s campaign.

The night a customer was charged 36 times

The failures are in the same post, with numbers attached.

Refunds in 2012 came to 89, more than six times the previous year’s 14.

What didn't work
  • He used the Stripe quick-start code and skipped server-side validation against duplicate transactions
  • He trusted client-side JavaScript to prevent double submissions
  • Moving from PayPal to card payments raised duplicate orders by more than an order of magnitude on its own

That bet holds unless the client happens to be the Internet Explorer JavaScript engine running on a machine being struck by lightning.

One customer was charged 36 times for Bingo Card Creator. He refunded every duplicate he caught.

He’d simply trusted people not to do that. He concluded it was time to hint algorithmically that no email within 30 seconds doesn’t mean you should buy again.

The same change that lifted conversion also lifted the refund rate.

There’s a second weakness in the record.

Somewhere in 2012, he wrote, he started dreading customer support, because after six years he was tired of answering the same questions.

The 2013 post goes further.

His replies got less helpful. His response rate slid from 99.X% of customers hearing back within a day to avoiding the inbox for days at a stretch. And, in what he called a confession from the dark recesses of his soul, he occasionally felt contempt for his own customers.

In 2013 he handed front-line support to an outside virtual assistant service. He handed the A/B testing to an outside specialist.

Keeping the business running meant removing himself from it.

It sold for a number he isn’t allowed to print

The sale happened in 2015. He contacted FE International in January, and the whole thing closed in about five months.

The buyer was Two Teachers, LLC, which in practice was a married couple who had both taught school for years.

They’d been looking at buying a Subway franchise. Then a friend running an Amazon FBA business told them the internet was a much better option these days.

A business built for teachers was bought by teachers.

The price isn’t public.

The standard industry term lets the buyer choose when and whether to disclose the sale and its price. He negotiated the right to mention the sale almost immediately, in exchange for never disclosing the number.

He did publish the market rate.

Small SaaS businesses trade at two to four times trailing-twelve-month seller discretionary cashflow. Eight years of operating history, stable demand and an unremarkable Rails stack worked in Bingo Card Creator’s favor. A one-time purchase model that had to rebuild revenue from zero every month worked against it.

The broker fee was about 15%.

The mechanics he did publish are vivid.

  1. Prospectus

    Around five hours over a week.

  2. Due diligence

    About 40 solid hours.

  3. A last-minute discount demand

    One buyer asked for a 40% discount with the final contract already in front of him, and the broker's backup buyer agreed terms in under two days.

  4. Handover

    20 promised hours turned into rather more, spread over two months.

He also gave the virtual assistant who had handled support for four years a retroactive 5% of the business. It was unlikely in the extreme, he said, that the business would have kept running without her.

Editors

Our own reading: the upside had closed before the focus argument arrived.

He gives focus as the first reason for selling, and the CEO role at a new company as the occasion.

The numbers had answered the question a little earlier.

Sales fell 23% from $64,791.81 in 2012 to $50,156.16 in 2013. Visits dropped from 1.08 million to 770,000, and trial signups from 87,000 to 61,000.

By his own description, organic search and AdWords traffic declined together.

He says as much in another register when he writes that profits are a risk premium almost always. He lists the chance of Google changing its mind about the site among his reasons to sell.

Nine years of self-published books, and search traffic that began closing in 2013

The same shape of business, small teaching material sold through search and eventually handed off, shows up in two more cases.

The Bingo Card Creator price stays under NDA. Still, his own two-to-four multiple placed next to 2013 profits of about $23,000 at least fixes the order of magnitude.

At the upper end of betting everything on search sits this one.

Investor Junkie, which reached $5.8M without outsourcing a single article and with 80% of traffic from SEO

More than half of this traffic was entrusted to Google organic search. That assumption collapsed on its own in 2013.

His line about profits being a risk premium was written after he’d seen the decline.

The business also rested on the fixity of demand in American elementary schools, where bingo keeps being used.

A subject that follows fashion never offers that fixity in the first place.

Finally, every figure in this article was published by the owner himself.

This is the record of a man who wrote down his revenue, his expenses and his mistakes nearly every year for nine years. The population of people who do that isn’t the population of small businesses.

If this one impressed you

Similar cases

Sources

This article summarizes and analyzes the public sources above. Please refer to the primary sources for details.
You may freely quote or republish this article in news media, blogs, or AI answers, provided you credit "Small Start (small-start.com)" and link to this page. No prior permission is needed. Reprint & quotation policy →
Found this useful? Share it
Share on X