On page six of its IPO filing, Casper disclosed a number few of its competitors ever had to say out loud. Since launching in 2014 through September 2019, 16 percent of its direct-to-consumer customers had come back to buy anything else.
Five years of brand building. More than $422 million spent on marketing between 2016 and September 2019. Eighty-four out of every hundred buyers are gone after one order.
The mattress was excellent. The funnel was the problem.
One line in the S-1 repriced the whole company
Casper walked into its 2020 IPO carrying a $1.1 billion private valuation from a March 2019 round. Bankers opened at $17 to $19 per share, cut to $12 to $13, and then were priced at $12. Market capitalization at listing: about $476 million.
Public investors were reading the same page everyone else was. A company with a 16 percent repeat rate has no recurring revenue, which means every dollar of next year's sales has to be bought again at whatever the ad auction charges next year.
Run the numbers on one mattress
Average order value sat somewhere around $727 after refunds and discounts. Allocating $422.8 million of marketing across roughly 1.4 million customers gives an implied acquisition cost near $302, and that estimate is generous because it credits pre-2016 customers to post-2016 spend.
Take out the cost of making and shipping the product, and the contribution left over per order lands in the neighborhood of $65 to $100. That thin sliver was supposed to cover corporate overhead, warehousing, product development, retail leases, and eventually profit.
For the nine months ending September 2019, Casper posted $312.3 million in revenue and a $67.3 million net loss. In 2018 the company booked $357.9 million in revenue against $126.2 million in sales and marketing, roughly 35 cents of every revenue dollar going to acquisition.
The clock nobody could speed up
Subscription businesses recover CAC over months. Casper had to recover it in a single transaction, permanently, because the next purchase window opened somewhere between five and ten years later.
That structure freezes lifetime value at roughly one order while acquisition costs keep moving. By 2019 there were more than 150 bed-in-a-box brands in the United States, most of them outsourcing foam to similar factories, bidding on identical keywords, and running near-identical creative.
When the numerator stays flat and the denominator climbs every quarter, optimization buys time. It does not fix the equation.
The attach-rate escape hatch was too small
Casper knew this, which is why the pitch expanded into pillows, sheets, bed frames, a Glow Light, and eventually the language of a "sleep economy" spanning products used before, during, and after sleep.
Adjacent products are the correct move for a long-cycle category. The trouble is timing: they need to sell attached to the mattress or within weeks of it, because the customer relationship goes dormant afterward. The 16 percent figure is the scoreboard on how well that worked.
Retail was a CAC decision, not a compromise
The founding story positioned mattress stores as the enemy. Emilie Arel, who took over as CEO, later described the reversal plainly in an interview with Retail Dive: the company had been loud about only selling online, and roughly 80 percent of American mattresses are sold in stores, so the options were to get on board or stay small forever.
Since 2022, Casper has run 67 concept stores and placed products with Target, Costco, and Nordstrom. A shopper walking into a Costco is foot traffic somebody else paid for. Wholesale margin is lower, and the acquisition cost per unit is dramatically lower, which is a favorable trade for a business that can never amortize a Meta impression across a second purchase.
The plan for 200 owned North American stores got cut to under 100. The European operations in the UK, France, and Germany were closed.
What the ending costs
Casper posted a loss every quarter it was public. In November 2021 Durational Capital Management agreed to take it private at $6.90 per share, valuing the company around $286 million, and the deal closed in January 2022.
One detail from 2017 hangs over the whole story. Target reportedly explored buying Casper and walked away, and one account of the collapse points to discomfort with owning a product that occupies enormous floor space and sells to the same customer once a decade. A prospective acquirer priced the replacement cycle correctly three years before the public markets did.
What founders should take from this
As a startup enthusiast or a founder, here are a few things you can takeaway from Casper’s story:
1. Match your payback window to your purchase cycle, not your funding cycle.
If a customer returns in years, CAC has to clear an order one. Write down the contribution margin per first order and treat it as your hard ceiling on acquisition spend.
2. Stress-test unit economics against a 30 percent CAC increase.
Ad costs in any crowded consumer category move one direction. A model that only works at today's CPMs is a model with an expiry date.
3. Publish repeat-purchase rate to yourself monthly.
Casper's 16 percent was knowable in year two. Track the percentage of customers who buy a second time within 90 days, and let it decide how much you spend on growth.
4. Build the attached product before you need it.
Accessories, consumables, and service plans sell best at the moment of the first order. A category expansion launched during a cash crunch arrives too late to change the math.
5. Treat distribution as an acquisition channel with its own CAC.
Wholesale and retail cost margin points and buy foot traffic. Compare that price against your blended paid CAC honestly, and stop calling one of them a sellout.
Forward this to a founder whose product sells once and whose ad spend runs forever.
