Daily Deal Success is All About New Customers

A major concern for Daily Deal space is whether deal promotions are profitable for individual merchants. The argument being: if the average discount is 50+%, and the Daily Deal service takes 30-50% in commissions, then how can a merchant possibly earn a profit afterwards?

A Rice University study attempted to provide an academic approach to this question, finding 32% of Groupon promotions are unprofitable for the merchant in a survey. This contrasts somewhat with a survey we performed this sumer, where 94% of merchants who ran a Daily Deal would like to do so again (versus 40% in the Rice survey). No doubt countless other surveys have been performed, with varying conclusions on either side of the profitability question.

Merchants may not understand the profitability of their own promotions

Daily deal profitability is a complex function of many variables specific to the promotion, the deal service, and the merchant’s margins. A survey is ineffective, since any individual merchant may not fully grasp the ultimate profitability of their promotion, and the aggregated profitability numbers doesn’t address the true answer to whether these promotions are profitable: it depends.

Growing somewhat frustrated with the ongoing controversy over profitability, we built a model to calculate the profitability of any Daily Deal.

Yipit’s Model for Daily Deal profitability

I assumed four main types of buyers: (i) tour bus customers who use their voucher and don’t return, (ii) new customers who become repeat visitors, (iii) “breakage” or people who don’t redeem the voucher, (iv) buyers who are already customers of the merchant.

A few inputs that might not be obvious:

Overage %: average amount a customer spends in addition to the face value of the voucher. Groupon has reported this number to be as high as 60% over the face value.

Cost of Goods Sold %: the marginal cost to the merchant of the good or service being provided. This varies slightly from traditional accounting Cost of Goods Sold – it includes any variable labor and administrative expenses as well.

Conversion to Repeat %: the percent of new customers that redeem the voucher that become repeat visitors of the business. Yipit conducted a survey this summer, which estimated this number at 19%. The value, however, is extremely difficult to estimate and likely varies by category.

Current Customers Already Going %: of those purchasers who are already customers of the business, what percent of people were already visiting the business otherwise, and were not inspired to attend an additional time. This is very difficult to estimate.

Cannibalized GP: the foregone profit per voucher from those purchasers that would have patronized the business anyway.

This model makes several simplifying assumptions, namely that overage is the same for each customer type, that there’s no fixed cost to the merchant for running the promotion (time, training, etc) and that there’s no further branding benefit for being chosen as a deal service’s feature.

You can download an Excel version of this model to input your own assumptions, and estimate merchant profitability for any specific promotion or service.

Key Takeaways

1. The vast majority of value is driven by the creation of new, repeat customers. Even if only a small percentage of purchasers become repeat customers, the amount of value from that group overwhelms unprofitable or breakeven segments.

2. Overage can be make or break. This should be a critical consideration for any local business setting up their promotion: always create an opportunity for to upsell the customer.

3. Breakage helps, but not as much as everyone thinks. The profitability conclusions of this model will vary widely given different promotion terms, deal services, and merchants. This is the first of a several posts in a series that uses this model to examine the profitability of various sides of the Daily Deal space.

  • Pingback: Loci 2010 – Jim Moran of Yipit | Understanding Google Maps & Local Search

  • Pingback: Is Groupon Stores Dead On Arrival? | Yipit Blog

  • Jackpilger2

    Projection is another name for vomiting,do some work and run some actual TESTS and then report those numbers. Jack

    • http://twitter.com/jdmoran Jim Moran

      I did. Most of the assumptions came from surveys we conducted of more than a hundred small businesses that actually ran these deals. They reported those numbers to us.

  • http://www.datamobilitygroup.com Joseph Martins

    I find that many (if not most) of the people who use Groupon are either a) shopping deal to deal, or b) existing customers looking to score a deal at one of their favorite shops. That is to say, those people most likely to use Groupon are not the type of people I would want as customers.

    Unless one’s goal is to chase low margin high volume business built on bargain hunters of questionable loyalty, avoid Groupon like the plague.

  • http://twitter.com/phoneranger Drew Robertson

    I think you might get a different result for perishable goods eg Nets tickets. And for those with high fixed costs where simply making a contribution to overhead is key eg car rentals. In those cases, you care less about repeat visits and more about filling seats NOW.

    • http://twitter.com/jdmoran Jim Moran

      That’s totally right. Most Daily Deals are able to be used at the consumers leisure, for up to a year, so they’re not really about moving expiring inventory. Nets tickets has a different value proposition.

  • Pingback: Yipit’s Jim Moran: Lots of Winners in Hyperlocal | Street Fight

  • Kimberly Ochwangi

    Thanks for sharing!  Great info!

  • Josh Wilson29

    Jim, could you please clarify your calculation of Cannibalized GP?  Because you have introduced the concept of “current customers already going %” but really have not differentiated it from “current customers”, the cannibalization calculation becomes obfuscated.  I would have thought cannibalization would have been a simple (GP of Current Customers Purchasing Coupon) – (Price Value of Voucher * (1-COGS)), assuming that the current customers would have spent $25 at the store regardless.  Even if you get away from the nitty gritty of modeling assumptions, doesn’t $1.75/voucher strike you as a very low estimate for cannibalization?