How Groupon’s accounting changes hide what’s really going on at the company

I have long criticized Groupon’s accounting practices. As someone who has watched the company closely since the it first filed the paperwork to go public (now scheduled for November 4), I have seen how it has continually adjusted its S-1, often in response to withering criticism.

Analyzing Groupon has been like playing 3-card monte. I’ve managed to successfully follow the queen and track the changes in the company’s accounting. Until now. I’ve lost the queen.

Continually revised accounting

Groupon started off its reporting by claiming that marketing expenses don’t matter and that investors shouldn’t worry about them. It invented a new metric called Adjusted Consolidated Segment Operating Income. If you ignored Groupon’s marketing expenses, Groupon generated a profit of $81.6 million in the first quarter. That metric was so ridiculed that Groupon removed it in the second amendment to its S-1. When that was converted to a more normal accounting metric, Groupon actually lost $98.3 million in the quarter.

Groupon also initially reported its revenue on a gross basis, meaning that it included in its revenue the portion of a Groupon’s value that is shared with merchants. This was clearly in conflict with FASB regulations. So Groupon also abandoned that metric with the third amendment to its S-1. That change dropped Groupon’s 1Q revenue by more than half from $644.7 million to $295.5 million.

But the company’s most recent change in its accounting practices is especially troubling. Released just two weeks before its IPO, Groupon significantly changed the definition of key expense metrics such as marketing expenses, cost of revenue and selling, general and administrative (SG&A) costs. When it made these changes, Groupon did not restate the numbers for the preceding two quarters. We are just supposed to trust that Groupon is now near breakeven, right before the IPO.

At least with ACSOI and gross accounting, all of the correct information was in the S-1. Savvy investors and analysts could ignore the numbers that Groupon wanted investors to focus on and look at the right numbers. With the most recent changes, investors have no way of reasonably comparing its expenses in 3Q with expenses in 2Q and 1Q.

Groupon continues to try to get investors to focus on year-over-year numbers, but those numbers are utterly irrelevant when the company has grown as fast as it has. The quarterly numbers are the only ones that are relevant and Groupon does not provide those.

For example, based on the previous filings, we know that Groupon’s SG&A in was $178.9 million in 1Q and $273.1 million in 2Q. In the third amendment to the S-1, the combined figure for 1Q and 2Q was $407.7 million. But with the latest revision, all we know is that under the new new rules, the combined number for 1Q-3Q is $565.7 million.

Confusometrics

One of my long-standing concerns about Groupon is the extent to which the “Groupon Promise” represents a risk to investors. Under the promise, Groupon customers who are unable to redeem a Groupon or are unsatisfied with their experience can get their money back. Based on the third amendment, it was possible to estimate the percentage of Groupon’s revenue that went to refunds. I pointed out that on a percentage basis, Groupon’s cost of revenue increased more than 40 percent, from 6.8 percent of revenues in the first half of 2010 to 9.6 percent in the first half of 2011.

Sources with knowledge of Groupon’s internal numbers have told me that refund rates are increasing in line with that estimate. I specifically asked Groupon for clarity on this number. Groupon did not respond to that request.

I also gave Groupon the opportunity to comment on its accounting practices for this post. Spokeswoman Julie Mossler responded, “Rocky, we’re choosing not to work with you.” (The company is also currently in its SEC-mandated quiet period surrounding the IPO and often does not respond to media requests for legal reasons.)

After my request, Groupon again redefined the cost-of-revenue metric, including in it many other expenses that would make it hard to evaluate the impact of the Groupon Promise on the company.

If Groupon had consistently reported the expense numbers from the beginning, I wouldn’t have a problem with it. If it restated 1Q and 2Q based on the new new rules, I wouldn’t have a problem with that, either. The fact that it did neither should be a red flag to diligent investors.

In other areas, Groupon continues to hope that investors will make meaningless comparisons. For example, Groupon expresses its marketing expenses as a percentage of gross billings. What investors should really pay attention to is marketing expenses as a percentage of revenue. Groupon’s marketing expenses are 22.3% of gross billings, but 54.8% of revenues. This discrepancy will only increase as Groupon’s share of Groupon deals continues to drop.

Since the beginning, Groupon has not reported key metrics that are essential for investors to assess the health of its business. This includes email open rates (which are likely declining rapidly), churn among customers (likely increasing) and subscriber/customer acquisition costs. (In fairness, the latter number can be approximated using Groupon’s marketing expenses.)

One key number that was in previous S-1s was removed as it got worse: the percentage of each deal that Groupon gets to keep. This number dropped from 42% in 2Q to 37% in 3Q. Over time, I expect this number to drop to 15%-20%.

In other cases, Groupon reports numbers in ways that are very confusing to casual observers. I saw several news outlets report that 30 million consumers purchased Groupons in 3Q. That’s incorrect. Groupon only reports cumulative numbers of Groupon purchasers. Because it’s a cumulative number, by definition, that number can never go down. But it’s reported in the quarterly column. By my estimates, only about 17.6 million people purchased Groupons in 3Q.

Barely passing grade

Groupon is like the consistently failing student who suddenly gets barely enough questions right on the final exam to get a passing grade, without showing any of his work. By not offering restated numbers for 1Q and 2Q, Groupon is not showing its work.

It’s possible that the student was diligent and worked really hard to earn a passing grade. Or it’s possible that the student cheated. Unfortunately, it’s impossible to determine what exactly happened without a deeper audit.

But these factors are important to keep in mind:

• Since the beginning, the company has consistently tried to make its revenues seem substantially bigger than they are and tried to convince investors that its massive losses were, in fact, profits.
• In explaining the business model of Groupon to merchants, Groupon ignores one of the biggest factors that causes Groupons to be unprofitable: the percentage of Groupon customers who are already existing customers.
• Groupon co-founder and chairman Eric Lefkosky has a troubling past with other companies, according to Fortune. In one email, about Starbelly, he wrote, “Lets start having fun… lets get funky… let’s announce everything… let’s be WILDLY positive in our forecasts… lets take this thing to the extreme… if we get wacked [sic] on the ride down-who gives a shit…” Starbelly sold itself for $240 million to Ha-Lo industries, lining Lefkosky’s pockets. Ha-Lo soon went bankrupt.
• Ed Ketz, an associate professor of accounting at Penn State University, told Bloomberg West that according to a model that he ran, there’s “almost a 100% probability” of accounting fraud at Groupon.
• Ketz also pointed out that Groupon’s auditor, Ernst & Young, has not issued an audit report on Groupon’s internal controls. From Groupon’s S-1: “We were not engaged to perform an audit of the Company’s internal control over financial reporting. Our audits included consideration of internal control over financial reporting as a basis for designing audit procedures that are appropriate in the circumstances, but not for the purpose of expressing an opinion on the effectiveness of the Company’s internal control over financial reporting. Accordingly, we express no such opinion.” Two former Groupon salespeople tell me that Groupon’s systems are not robust enough to accurately track its operations. Even in the week of its IPO, Groupon has changed the definition of numbers from 2Q.
• In its roadshow video, Groupon CFO Jason Child says that the company had “over $240 million of cash on the books and no debt.” Potential investors had to rely on the accompanying slide to see that the company owes merchants who have run Groupons $466 million.

The irony is that if Groupon hadn’t tried to push through its unusual accounting measures, it might have gone public sooner and at a higher valuation.

Rocky Agrawal is an analyst focused on the intersection of local, social and mobile. He is a principal analyst at reDesign mobile. Previously, he launched local and mobile products for Microsoft and AOL. He blogs at http://blog.agrawals.org and tweets at @rakeshlobster.

Top photo: Art Prestige Studio / Shutterstock.

Related articles

• Groupon’s tricky S-1 math (venturebeat.com)
• Who gets hurt if Groupon collapses (venturebeat.com)
• Weak technology weighs down Groupon’s sinking ship (venturebeat.com)
• Groupon will IPO in 2 weeks at an $11.4B valuation (venturebeat.com)
• When Groupons are bad for small businesses (venturebeat.com)