Blue Apron's S-1 peak customer count did not hold

The record

Blue Apron Holdings filed its Form S-1 on June 1, 2017 ahead of its IPO. The S-1 discloses that marketing expense rose from $14.0 million in 2014 to $144.1 million in 2016, an increase of about 930%, while net revenue rose from $77.8 million to $795.4 million over the same years. The filing's quarterly tables show Customers, its term for paying households, at 649,000 in the first quarter of 2016, rising to 907,000 in the third quarter, dipping to 879,000 in the fourth quarter, then reaching 1,036,000 in the first quarter of 2017. Marketing expense as a share of net revenue rose from 14.8% in the first quarter of 2016 to 24.8% in the first quarter of 2017. The company stated it expected marketing intensity to decline over the remainder of 2017.

What the documents establish

The S-1 is Blue Apron's own issuer-reported account of its customer and marketing trends through the quarter before its IPO; it shows rising marketing intensity accompanying a customer count that had already dipped once, in the fourth quarter of 2016. A second document, the company's Form 10-K for fiscal 2017, filed February 22, 2018, reports that the customer count the S-1 had put at 1,036,000 in the first quarter fell every subsequent quarter of 2017: to 943,000, then 856,000, then 746,000 by year-end, a decline of about 28% from the S-1's most recent reported figure. The 10-K is a separate, later filing; it does not state that the expected decline in marketing intensity failed to materialize, but the 28% customer decline it reports is the opposite of what the S-1 anticipated.

The operating read

A subscription business's S-1 disclosing a customer count at its historical peak, alongside marketing spend rising faster than revenue, is describing a moment, not a trend line. Editorially, a reader evaluating a similar filing should treat the most recent quarter in an S-1 as the company's chosen high-water mark for the disclosure, not as a stable base rate, and should look for whether the cost of acquiring each additional customer was rising or falling at that same moment. Blue Apron's own filing does not identify a single cause for what the 10-K later showed; multiple explanations, including competition and category demand, are consistent with the same two documents.

What to check before you decide

Before treating a pre-IPO subscriber count as durable, check the following.

• Does the filing disclose a full year of quarterly subscriber or customer trend, not just a single quarter?
• Is marketing expense, as a share of revenue, rising or falling in the periods immediately before the filing?
• Has a later 10-K or 10-Q reported whether the trend the S-1 anticipated actually occurred?

The S-1's disclosed customer count was accurate as of its stated date; it was not a forecast, and the 10-K that followed is the document that shows what actually happened next.