The Fundamental Problem With Groupon’s (Historical) Business Model

I used to work as a project-turned-product manager for Groupon! I worked there as an intern in 2015, then full time from 2016 to (early) 2021.

During that time (and before and beyond) Groupon certainly went through some struggles! It’s never been able to capture the high of its initial, at-the-time record breaking, fundraising, or of its 2011 IPO. I think the core challenge that it faced is a pretty elucidating example of the challenges of disruption! So I’m going to talk about that challenge, and hopefully highlight things that may be applicable if you find yourself in a potentially disruptive (or self-disruptive) circumstance!

By way of preface, for all of this I can only ever talk about my time there, and I’ll say that I worked on consumer-facing teams so I may be a little sketchier on some of the details of merchant-facing products or issues! This was all 5+ years ago, and I don’t believe any of this is material or nonpublic! In fact, I remember during my time seeing many outsiders making blog posts or youtube videos or reddit posts about this very dynamic!!!!!

I have not followed Groupon that closely, so while I can suspect that some of these things may still hold true today, I truly have no idea! So for the sake of argument, you should assume that all of this may have changed in the intervening 5 years! I literally go to the website or open the app about once a year out of curiosity to see which of my features they’ve removed, and I am still technically a (very very very) minor stockholder! So you know, do with all that what you wish.

Finally, Groupon has historically had many different lines of business. At my time it was predominantly split into Local, Goods (physical products you could buy, Amazon-like), and Travel (flights and hotels, Expedia-like.) While Goods was a significant portion of the business, this just focuses on the “Local” business line, where you go to Groupon to buy a deal from some business local to you, as that was the bread and butter of what Groupon stood for.

Anyway, with the throat-clearing gone: IMO the fundamental problem with Groupon, at least historically, is that its value proposition to merchants and to customers was incompatible.

The classic Groupon deal structure was that a merchant, let’s say Ruthie’s Cat Cafe1, decides to list on Groupon. She would offer a deal of some sort, say $10 for $20 worth of coffee. This means if a customer buys that deal, Groupon takes in $10 and gives the customer a voucher of some sort, canonically a printout with a bar code. When the customer goes to Ruthie’s Cat Cafe, they mention their Groupon and present the voucher to redeem it. Then the barista knows to let the person get up to $20 worth of stuff, charging for the difference if they “spend” more than $20.

Groupon has now taken $10 from the customer, and the customer has $20 of stuff from Ruthie’s Cat Cafe2. Groupon then splits the $10 and keeps $5 for itself, sending $5 to the merchant.

So we net out with:

• Groupon: +$5 in money3

• Ruthie’s Cat Cafe: +$5 in money, -$20 in product, net -$15

• Customer: -$10 in money, +$20 in product, net +$10

On its face……. This is a terrible deal for Ruthie’s Cat Cafe!!!!! And other than like ignoring taxes etc, these are not abstract numbers simplified for the purposes of demonstration, as best I can recall the default deal structure that was commonly used was at least 50% off, with a 50/50 split between the merchant and Groupon. Sure you could negotiate, and over time I believe Groupon cut its take closer to 30%(?), but this is my recollection of the basic structure behind second-half-of-the-2010s-Groupon.

That said, this becomes theoretically interesting when thought of (as I understand it was sold to merchants) as a marketing expense. Businesses often to have to spend money in order to acquire a customer. You have to pay to take out an advertisement, or to send out a coupon, or you prepay for a farmer’s market stall, etc. In comparison to that, there is a genuine benefit to Groupon’s structure where this is marketing with no upfront cost! If you are taking out an ad, you have to put money down first and then hope the customer comes later, which can be annoying for cashflow purposes or prohibitive for someone just starting out who hasn’t made the money yet.

In the Groupon case, the order is reversed, where you get money now4 and only pay the “marketing cost” later when you have to give the customer the “free” product. Also, unlike types of advertising where you pay to put up the ad regardless of how many people actually purchase from it, you only “pay” (in the form of giving a cut to Groupon/free product to customers) if the customer actually buys the Groupon and actually redeems. In that sense it’s both less risky and easier to measure return on investment than other forms of marketing.

So in theory this is potentially fine, and maybe even preferable from a cashflow perspective to comparable marketing options. The way to tell is the basic math that underlies all marketing spend decisions: what is the customer acquisition cost (CAC), and how does it compare to customer lifetime value (LTV) aka how much we will make off them over time? For a sufficiently large purchase, like a car or a laptop or whatever, you may have LTV > CAC on just the one purchase alone! That is clearly not the case in the Groupon example, where you are in effect making net -$15 to acquire this customer. But that’s where LTV and the theoretical promise of Groupon comes in: if you can convince the customer to come back, or tell their friends, and to pay full price, then you can have positive LTV > your CAC. This is exactly how many successful businesses work; it’s fine to be negative on the first purchase if you can make it up in LTV.

So that’s the merchant value prop: a no-money-upfront, only-pay-if-you-get-paid marketing expense, and if you play it right you can use it to acquire customers with positive LTV that you can make back over time56.

From a customer’s perspective it’s simple: you can get a deal! You spend $10 to get $20 worth of stuff! Great!

This can also be a way to try something new, where you may be wary about say indoor skydiving, or a new restaurant that just opened, but if you can try it for 50% off then that might be intriguing.

The problem is that Groupon successfully built the thing that is almost impossible to build, especially build at scale: a strong consumer brand. Groupon meant deals. That’s why you came to Groupon. That’s why you stuck around at Groupon. And that matches with the consumer value prop of the average Groupon deal! You came here for deals, great, here’s a deal!

And even rarer than just building a brand, Groupon did have a loyal customer base!

But look at the behavior that is healthy and positive for merchants, in this structure: ideally they can use Groupon to attract a net-new customer with a no-money-upfront marketing expense in the form of a deal, and then that converts into a repeat full-price (or lower-discount) customer with positive LTV.

Those existed! My Grandma is one of them! She would see a Groupon for some class or spa treatment or cuisine that she’d never tried before, use it as a way to take a low-risk flyer on something new, and then if she liked it would come back and pay full price!

…… My grandma was not the average Groupon customer. The average Groupon customer was deal-focused and price-sensitive. The much more common behavior would be to go to Ruthie’s Cat Cafe, get the free stuff, and never ever come back, instead finding a new different cafe running a Groupon and repeat the cycle. Or to try and buy the same Groupon multiple times and keep coming back expecting to be able to get your $20 of stuff for $10. The latter is why most Groupons had a limit on how often they could be purchased in a given 180 day period!

In this case, the merchant economics completely fall apart. If you aren’t getting repeat full-price customers, you can only make money if your CAC from the deal is LTV positive on that first purchase. But since Groupon was a deals marketplace with a deal-focused customer base, the only way you will get anyone to buy your deal is to make it visibly attractive, usually at least 50% off, and priced competitively to other options. And almost no local business has 50%+ margins! Sometimes you could make this work by using the Groupon to get the customer in the door, and then charging for high-margin add-ons to make back what you’re losing by the discount and Groupon cut. But of course, if you’ve got a price-sensitive customer who just arrived thinking they’re getting a great deal and they’re in fact getting aggressively upsold, that tends to lead to angry customers!

In effect, there is an unsolvable conflict between the desires of the customer (just keep buying discounted stuff) and the merchant (convert the discount-hunting customer into a repeat customer.) Accordingly, many many merchants churned off Groupon, either because they realized it was not serving their goals, or they went out of business!

The aforementioned 180 day repeat purchase limit is a good example of this push and pull, in that from the business’ perspective you don’t want “I can only buy one every 180 days”, you want “I can only buy one ever and then have to come back at full price.” But my recollection is whenever we tried that customers revolted, because from their view the whole point is to get great deals. 180 days was one of the unsatisfying ways to try and split the baby.

An actually very illustrative counter example is the one major blue-chip merchant I can think of that always came back every year: Costco7. I don’t remember the exact deal details (though I bought it once!) but it usually was something like 50% off a one-year Costco membership if you signed up to auto-renew. This worked for the customer because it’s genuinely a deal! You’re saving money! But it also worked for Costco because:

1. The membership itself is almost all pure profit, so they’re not losing money by selling it for 50% off, they’re just making less

2. Costco is legitimately great and also deal-centric! Meaning that even if you acquire a deal-hungry customer from Groupon, they’re still going to auto-renew their membership next year at full price because they like what Costco offers them, unlike Ruthie’s Cat Cafe where full price means they just go try and fleece a different coffee shop. The core Groupon customer behavior is also the core Costco customer behavior.

To Groupon’s credit, it kept trying to find ways to fix this! Trying out different deal structures, both in terms of how big the discounts were, how big Groupon’s take rate was, how you redeem them, how often you could buy them, etc. The one that seemed the most promising to me at the time I left, though who knows what’s happened now, is trying to build deals that are structured to incentivize the repeat visit flow, where you get a deep discount on the first visit, and get smaller discounts for repeat visits. Another alternative that I remember being explored was trying to do time-based deals where you would get the deep discount if you made a reservation or visited during a slow time or slow season (when merchants might be more incentivized to offer discounts so they can at least cover fixed costs) and pay closer to full price if you’re trying to buy for a peak time.

Similarly on the marketing and consumer-facing side8 they would try to rebrand and shift the brand story to focus less on Groupon as a “deals” site and more as a site for “local discovery.” Try to sell Groupon as the place to find your next favorite spa or excursion or restaurant and highlight the quality of the experience as more important the price, and to use that to get a less price-sensitive customer base to use the site.

And the two of these together fit quite nicely in theory! If the customer base is less deals-focused, that can enable more and better kinds of merchants to be on the platform, which can help attract more quality-focused customers, which can attract and keep more merchants, etc etc etc.

But that right there is a chicken and the egg problem. If you’ve got both, you can hopefully build a virtuous cycle. But if you start with a deal-focused customer base and try to add more high-quality merchants onto the platform, they will only do so if they can actually make money off the platform. But:

1. if they make more money because we reduce our take rate, we just mechanically make less money! Better for the platform in the long term to have happier and more successful merchants, but worse in the short term because money is what you need to run the business of being Groupon

2. if they make more money by charging full-price or low-discount, that makes our existing customers mad because they have to wade through stuff that they don’t want (full price/low-discount) to get to the stuff they do (high-discount.) Better in the long term if we can retrain behavior or replace this customer base with another one (especially if the other one is bigger!) but bad in the short term for customer satisfaction and for conversion rate

And on #2 you can run into the challenge where if you have automated recommendation systems, they will see that the lower-discount deals perform worse, which leads them to stop recommending them, which leads them to perform worse, which leads them to stop recommending them, etc etc etc9.

If you tackle it from the other direction, and keep your deal-amenable merchants but try to acquire more quality-and-discovery-focused customers, then they’ll arrive at Groupon and see a bunch of deep discounts for (not universally, but often) merchants that are lower-quality and/or struggling to stay in business. The actual halo merchants that would satisfy this new cohort and this new behavior aren’t on the platform yet, Costco excepted.

So you need a new customer base in order to support the new merchant base, but need a new merchant base in order to attract and satisfy a new customer base. Chicken and egg!

This is what disruption, at least when used academically, basically means! There exists a different way of doing business that not only is (potentially) better in the long run, but you (or whatever incumbent you’re challenging) have to damage your existing business in order to do it. If you continue to optimize the current business model (sell deep discounts to discount-hungry customers) then you will never be able to break out of that valley to whatever is beyond (sell quality local merchants to discovery-hungry customers.) If you do want to break out, that will cause some very real and very legitimate pain along the way.

Disruption is not just “do the thing that is obviously better in all ways”! Academically that’s called a “sustaining” innovation: it just takes what’s already happening and does it more or better. A “disruptive” innovation is in many ways worse, but is in other ways crucially better, and you have to travel through the worse to get to the better.

There were so many lessons learned from working at Groupon (see basically every product management post I’ve made here!) but these big foundational ones really stuck with me: you need to be thinking about your entire product and how it impacts all participants, and sometimes you do need to disrupt yourself if you want to genuinely change your trajectory. At least for the near-five years I was there, we definitely had an understanding of the first part, but we never really were able to commit to the second. And so Groupon never really was able to successfully self-disrupt10.