Coupon affiliates: bad for advertisers and for publishers

Coupon affiliates are in the top three publisher types by revenue in almost every affiliate network. They convert well, they are easy to accept, and the dashboard looks fantastic. So it is no surprise that advertisers keep them on and networks welcome them.

And yet, on closer inspection, they are one of the most destructive parts of the affiliate ecosystem. Not only for the advertiser, which you may already have suspected, but just as much for the serious content publisher who does the real work.

This piece walks through the exact mechanisms.

How it works: the checkout moment

Coupon affiliates live off one specific psychological moment. The consumer has decided to buy something. They are in the checkout, the basket is full, and there it is: "Do you have a discount code?"

That field triggers a pattern that is the same for nearly everyone: open a new tab, search for "[shopname] discount code", click the first coupon site, try a code (often expired), and maybe click through to the "working" one.

At the moment of that click on the coupon site, a new affiliate cookie is placed. In most networks the rule is: last cookie wins. The sale then takes place, and the coupon site receives the full commission.

And the cookie placed earlier by the content publisher who convinced the visitor weeks before? Gone.

The damage to advertisers

1. You pay for a sale that would have happened anyway

This is the fundamental problem: the consumer was already ready to pay. The buying decision had been made. No extra customer was brought in; the coupon site merely diverted the route of a buyer who had already decided.

In the literature this is the incrementality argument: does an affiliate channel deliver net new customers, or does it intercept existing conversions? For coupon sites the answer is rarely positive. Retailers who temporarily switch off their coupon publishers sometimes see conversion rates barely move, because customers simply buy without a code.

Meanwhile the advertiser has paid twice over: first the coupon (margin lost on the order) and then the affiliate commission (an extra percentage on top). For a €100 order with a 10% discount and 8% commission, that amounts to €18 in costs for a sale that could have been free.

2. Code leakage

Advertisers usually issue several kinds of code:

Public promotions: "10% off in June"
Newsletter exclusives: for existing customers or subscribers
Influencer codes: tracked codes per creator
Retention codes: for customers at risk of churning
Winback codes: deep discounts for long-inactive customers

Coupon sites grab all of it. As soon as someone publishes such a code anywhere (deliberately or not), it is on all the major coupon sites within hours. A newsletter code intended for 5,000 subscribers is now used by 500,000 people.

The marketer who wanted to run "clever segmented discounts" sees the strategy undone, and pays commission to a coupon site for every redemption, including from customers who never saw the code through the affiliate channel.

A real example, Wehkamp (2013): as early as August 2013 Wehkamp publicly parted ways with coupon sites, as one of the first large Dutch webshops. The reasoning was direct: Wehkamp already sent discount codes to customers by email itself, so coupon sites added nothing, yet still took the commission on codes they had not distributed. On top of that, coupon sites regularly promote "codes or promotions that do not exist" and win their search positions on "[brand] discount code" through dubious SEO tactics. Wehkamp's conclusion: the value proposition of coupon affiliates did not hold up. Source: Emerce, 19 August 2013 (in Dutch).

3. Crowding out valuable publishers

This is the most underestimated problem. Content publishers (reviewers, bloggers, newsletters, communities) put considerable time into convincing consumers. They write articles, build comparisons, maintain trust.

If the commission structurally ends up with coupon sites instead of with them, their effective earning capacity per customer falls. In the long run they stop promoting that advertiser: the ROI is simply too low.

The paradox: it is precisely the publishers who bring net new customers (people who might never have bought without the article) who leave. The coupon sites, who bring no new customers, stay. The network increasingly becomes a machine that taps existing demand and less and less a channel that creates new demand.

4. Unreliable reporting

If a large share of your commissions goes to coupon sites, your affiliate reporting is misleading. You see high conversion rates, because people who were already ready to buy nearly always convert. But you compare that conversion rate with channels operating early in the funnel. That is comparing apples with pears.

Advertisers who do not spot this cut, on the basis of that data, into campaigns that do add value, because coupon sites "win" on the numbers.

The damage to publishers

This part is discussed less often, but it is at least as relevant.

1. The click-stealing problem

Suppose a content publisher writes an extensive review of a coffee machine. The reader clicks through to the webshop, convinced by the article. In the checkout they hesitate over the price, search for a discount code, click through a coupon site, and buy.

The content publisher has:

Written a 2,000-word article
Taken or bought photographs
Built up domain authority over years
Guided the reader from orientation to a buying decision

And the commission? It goes to the coupon site that served the consumer an expired code at the very last moment.

For serious content publishers this is not an exception, it is a structural phenomenon. Industry estimates (RetailMeNot research, the Awin Benchmark) suggest that 15 to 30% of affiliate conversions in sectors such as fashion, electronics and home are "overwritten" by a late-funnel coupon click.

2. Defensive SEO as a business model

Coupon sites build their reach on one thing: ranking for "[brand] discount code". That is a search term with high intent (someone searching for it is ready to buy) and relatively little competition from the merchant itself.

They invest in that search traffic, not in content. Nor do they need to invest in content: the buying decision has already been made by another publisher. They only harvest the final moment.

Content publishers working for the same advertiser earn fewer commissions the more consumers pass through coupon sites. They are not even competing on the same playing field: they do the work and the coupon sites take the credit.

3. Rate pressure from false conversion signals

Networks and advertisers use conversion data to decide which publishers perform well. If coupon sites structurally show high conversion rates (again: logically, because they close people who have already decided), that can create the impression that content publishers are "underperforming".

The consequence: less budget, lower commissions, earlier termination of partnerships, for the publishers who actually do the most for the advertiser.

This is attribution failure in its most damaging form: not only rewarding the wrong party, but actively penalising the right one on the basis of distorted data.

Why networks accept coupon sites anyway

In fairness it has to be said: for a network, coupon sites are attractive. They generate conversions easily, the numbers look good, and advertisers complain little, at first. The network earns a percentage of every approved transaction.

Only when advertisers start asking whether those conversions were really new, or when content publishers walk away in frustration, do the consequences become visible. But that takes long enough to make accountability difficult.

There is a prisoner's dilemma element too: if network A excludes coupon sites but network B does not, an advertiser may switch to B in order to "see more conversions". In the short term that looks like a loss for A, even though it is better for the advertiser in the long run.

What you can do about it as an advertiser

You do not have to watch passively. Some concrete options:

1. Tie publisher approval to the business model Do not approve publishers blindly on the basis of traffic. Ask: how do they generate visitors? In what context do they promote your brand? A coupon site is recognisable from its business model: its domain name, its content and its traffic sources.

2. Never distribute coupon codes through the affiliate channel Keep codes you create for specific audiences outside the affiliate network. Or use unique codes per publisher, so that you can see which publisher shared which code.

3. Audit conversion attribution Do not only look at total conversions, but at incrementality. A simple test: switch off coupon publishers temporarily and see whether total conversions fall significantly. In most cases the answer speaks volumes. Wehkamp did exactly this in 2013 and concluded that the added value was not there; see the Emerce coverage (in Dutch) for the full story.

4. Choose a network that handles this for you The most effective solution: work with a network that structurally excludes coupon sites and puts publisher quality first. Then you do not have to police this yourself on every application.

Who wins, who loses

If you follow the money, it splits neatly into two parties who win and two who pay.

The winners

The coupon site itself. This is the easiest money in the chain. The site does no selling; it intercepts people who already wanted to buy and who take the final step with "discount code [brand]" in Google. One click at the last moment, and the full sale commission is collected. No content, no brand building, no genuinely referred traffic.
The network. The network earns a percentage of every transaction, whether it runs through a serious content publisher or through a coupon site. For the network's margin it makes no difference who claims the sale, as long as a sale is claimed. That creates a structural incentive to embrace coupon traffic rather than examine it critically: more claimed transactions means more revenue for the network. The advice an advertiser receives is therefore often coloured, aimed at transaction volume rather than at the advertiser's interest.
The advertiser's affiliate manager. Sometimes a winner too, not financially but politically. Someone who has to demonstrate internally that "affiliate works" looks better with a dashboard full of sales than with an honest story about incrementality. Coupon affiliates deliver exactly those attractive figures: high volume, low cost per sale, everything green. That most of it consists of sales that were there anyway does not show up on a management slide. The incentive to look critically thus collides with the incentive to look good.

The losers

The advertiser. Hit twice. (1) They pay away both commission and discount on sales they had already secured, pure margin leakage on traffic with zero incremental value. (2) Worse over time: serious publishers, who do bring new customers, see their conversions disappear because the coupon site jumps in at the last moment. Those publishers therefore send less traffic. The advertiser is left with the party that adds nothing and loses the party that did.
The serious publisher. They do the real work, content, comparison, trust, a customer who would not have come without them, and then see the sale "stolen" on the finish line by a last-click coupon site. Worse attributed conversion, lower income, and ultimately the rational conclusion to invest less in that channel.

In short: the coupon site and the network divide the profit, while the advertiser and the publisher, the two parties who actually keep the market running, pay the bill.

In closing

The data are counter-intuitive: coupon affiliates cost advertisers money they did not need to spend, and they cost publishers income they had earned. The only winner is the coupon site itself, and, in the short term, the network taking a percentage of every commission.

That the system has worked this way for so long says more about how hard attribution is than about how fair the model is. Measuring incrementality has only become accessible to the average advertiser in recent years.

What we hope: that more advertisers and publishers start asking the question that should always have been there, would this purchase have happened without this affiliate?, and adjust their publisher mix accordingly.

At Klatsa we have already made that call. More about this in Why Klatsa does not allow coupon sites.