Quick answer: Across our network over a 90-day window ending 14 September 2026, earnings per click ran from roughly $0.50 in insurance to the low-to-mid teens in financial services, and conversion rates ran from under 1% to the low 30s depending on vertical and partner. Those numbers will not tell you whether an offer is good for you, because conversion rate and EPC are both outputs of three things: where the advertiser set the conversion point, what traffic you are running, and what that traffic costs you. Within a single vertical on our network over this window, EPC across partners spanned roughly fifteen times, and the partner with the highest conversion rate was not the partner with the highest EPC. Use vertical benchmarks to orient. Use conversion rate and EPC read together, against your own traffic cost, to decide anything.
The benchmarks by vertical
Aragon Premium network, 90 days, 17 June to 14 September 2026. Vertical aggregates, banded deliberately. The methodology is stated in full further down.
| Vertical | Conversion rate | Earnings per click | Conversion point tends to be |
|---|---|---|---|
| Financial services | see note below | low-to-mid teens ($) | Shallow – qualified form submission |
| Refinance | high single digits % | $8–10 | Mid – qualified application |
| Personal finance | 2–3% | $2–3 | Deeper – funded or approved action |
| Credit building | under 1% | $1–2 | Event-based – earns after the initial action |
| Rewards | high 20s % | $1–1.50 | Shallow – registration or first action |
| Make money | ~20% | ~$1 | Shallow – signup or first task |
| Market research | ~20% | $0.50 | Shallow – survey start or profile |
| Insurance | low teens % | ~$0.50 | Shallow – quote request |
| Education | see note below | $2 | Shallow – information request |
| Home equity | low-to-mid 20s % | Not comparable | Sale-value model, not payout-per-click |
| Loans | 4–5% | $0.50–0.75 | Mid – qualified application |
| Debt | 5–6% | ~$1 | Mid – qualified consultation |
Note on financial services and education conversion rates. Both verticals show aggregate conversion rates in the mid-40s, and both are inflated by conversions attributed to clicks tracked elsewhere. We checked financial services at partner level: among partners running meaningful volume, conversion rates ran roughly 1.5% to 13%. We are publishing the corrected range rather than the aggregate, and flagging education as subject to the same distortion. Any benchmark table that does not check for this will overstate shallow-conversion verticals.
The conversion-point column matters more than either number. So does the thing the table cannot show you, which is where the traffic came from. If you are still deciding which vertical to build in, our guide to choosing an affiliate marketing vertical covers that decision; this page is about reading the numbers once you are in one.
Why the same offer pays different affiliates different EPCs
This is the part that gets left out of every published benchmark, and it is larger than the difference between verticals.
Within financial services over this 90-day window, across partners running meaningful volume on the same set of offers, EPC ran from roughly $3.50 to roughly $23. Conversion rates over the same set ran from about 1.5% to about 13%. Same vertical, same offers available, same window – a spread of roughly six to seven times on earnings per click.
Insurance is wider still. Across partners in that vertical, EPC spanned roughly fifteen times.
None of that variation is the offer. The offers were the same. The variation is traffic.
Does traffic source affect EPC and conversion rate?
Yes, and it is usually the dominant variable.
Search traffic arriving on a comparison page behaves differently from social traffic arriving from a paid feed, which behaves differently again from email, from display, and from a rewarded environment. Different intent at the moment of the click, different qualification, different likelihood of completing whatever the advertiser counts.
Two partners can run the same offer in the same week and post conversion rates an order of magnitude apart with nobody doing anything wrong. One is buying cheap, broad traffic and converting a small fraction of it. The other is sending fewer, highly qualified visitors from a page built around the exact decision the advertiser is selling into.
This is why a benchmark table is orientation and not evaluation. It describes a blended average across traffic types that have almost nothing in common with each other.
EPC vs conversion rate: which should you optimize?
Neither one on its own, and our insurance data over this window shows why the ordering between them cannot be assumed.
One partner in that vertical converted at over 30% and earned less per click than a partner converting at under 8%. Another recorded no tracked conversions at all across the whole window and still produced the highest EPC in the vertical, because it earned on downstream events rather than on the initial action.
If conversion rate were the thing to optimize, those results would be impossible.
Conversion rate tells you where the advertiser drew the line and how well your traffic clears it. EPC tells you what a click was worth once payout and conversion rate are combined. Neither tells you whether the campaign made money, which is the third number and the one only you hold.
Why is my EPC lower than the published benchmark?
Usually because your traffic mix differs from the blend behind the benchmark, not because anything is wrong.
A published vertical figure averages search, paid social, email, display and rewarded traffic into one number. If your traffic sits at either end of that range, the benchmark will misdescribe you in both directions: qualified search traffic into a vertical dragged down by broad paid social should beat the benchmark, sometimes by several times, and cheap broad traffic should fall below it.
Both can be the correct outcome. The gap is information about your traffic mix, not a verdict on your campaign.
Why a lower EPC can be the better business
Here is what makes conversion rate and EPC misleading when read on their own: neither one is your margin.
A lower EPC is the correct outcome for cheaper traffic, and it can be considerably more profitable than a higher one. If you are buying traffic at a low cost and earning $0.50 a click, that is a better business than earning $14 a click on traffic that costs you $13. The absolute number tells you nothing until it sits next to what you paid.
There are real quality trade-offs in going cheaper. Cheaper traffic usually converts worse, and pushed far enough it stops closing for the advertiser, which is when caps and rates start moving against you. But the objective is not the highest EPC. The objective is the widest gap between what your traffic costs and what it earns, held at a quality level the advertiser is happy with.
Find that point and the whole relationship compounds: the campaign is profitable for you, the advertiser’s downstream numbers hold up, and more budget comes your way. That is the position every serious affiliate is actually optimizing toward, and it is not visible in any single metric.
One honest limitation. We see your EPC. We do not see what you paid for the traffic – that number exists only on your side, and it is the one that decides whether an EPC is good. It is also the most useful thing you can tell an affiliate manager, because it changes which offers and which conversion points we should be putting in front of you.
Why does my EPC drop when my traffic grows?
Comparing the most recent 30 days with the prior 30 days, across four of our largest monetizing verticals, click volume and earnings per click moved in opposite directions in every case.
- A make-money vertical gave up roughly 72% of its volume; EPC rose about 71%.
- A financial services vertical grew volume about 80%; EPC fell about 32%.
- An insurance vertical grew volume about 146%; EPC fell about 49%.
- A personal finance vertical grew volume about 15%; EPC fell about 11%.
That held whether volume went up or down.
I want to be careful about what it proves. Two 30-day windows is a short read, and there are explanations we cannot rule out: the mix of offers inside a vertical changes, partners enter and leave, conversion points get redefined, and seasonality moves things. This is our own account, not the market.
The direction is consistent with what is happening to publishers at much greater scale. In an April 2026 Digiday Media Briefing, Sara Guaglione reported Apartment Therapy’s traffic down roughly 20% year over year with conversions up 34%, and Forbes’ site traffic down 37% in Q1 2026 while its Forbes Vetted affiliate business grew revenue and doubled its conversion rate.
The cause is well documented. A March 2026 Visibility Labs analysis found Google’s AI Overviews appearing on 14% of shopping queries, up from 2.1% in November 2025. Broad, undifferentiated queries are increasingly answered before the click. What still reaches your page carries more intent and is worth more per visit, while pages built to capture general volume are getting less of it.
There is a traffic-cost reading of the same pattern. Scaling usually means buying further down the quality curve, and the traffic you add is both worse-converting and often more expensive at the margin. EPC falls and cost rises at the same time, which squeezes from both directions. The offer has not stopped working; it has been pushed past the traffic that suited it.
What is a good EPC for your traffic?
There is no single figure, which is the main problem with benchmarks that give you one. Four questions get you closer than any table:
Where does this offer’s conversion point sit? A shallow point at a lower rate and a deep one at a higher rate are different businesses. Which suits you depends on what your traffic does after it converts.
What traffic are you sending, and how does it compare to what the benchmark blends? If you run qualified search traffic into a vertical whose average is dragged down by broad paid social, the vertical benchmark understates what you should expect.
What does that traffic cost you? EPC minus traffic cost is the only number that decides whether a campaign is worth running. A benchmark cannot see it and neither can we.
What is happening downstream after the conversion? You are paid at the conversion point, but the advertiser’s experience after it determines whether your rate rises, whether your cap lifts, and whether that advertiser is still buying next year.
The fourth question is the one most publishers skip. The advertiser’s downstream numbers eventually set your price, and a rate change often arrives months after the traffic that caused it. If you are earlier in the process and still building out where offers sit on your pages, our guide on monetizing traffic with affiliate offers covers the placement side of this.
Methodology
Most published affiliate benchmarks do not state their methodology. Here is ours.
- Source: aggregate reporting across the Aragon Premium network. Category and partner-cohort level only.
- Window: 90 days, 17 June to 14 September 2026, for the benchmark table and the partner-spread figures. The volume-versus-EPC comparison uses two adjacent 30-day windows, 17 July to 15 August and 16 August to 14 September 2026.
- EPC is payout per click – what the publisher earns per click, not what the advertiser pays per acquisition.
- Conversion rate is conversions divided by clicks, blending conversion-point definitions that differ by advertiser and traffic types that differ by partner. Both are why the column should not be read alone.
- Partner-spread figures cover partners running more than 300 clicks in the vertical over the window, so a handful of low-volume outliers do not set the range. No partner is identified, and no per-partner payout, revenue or margin figure appears anywhere on this page.
- Banded, not exact. Figures are ranges rather than precise values, and are accurate to the band.
- Known distortion, disclosed: vertical conversion rates can be inflated where conversions are attributed to clicks tracked elsewhere. We found this in financial services and education and have flagged both rather than publishing the inflated figure.
- Excluded: verticals with negligible volume, and two verticals whose reported conversion rates exceed 100% for the same attribution reason.
- Not included: payout rates, revenue, margin, per-advertiser or per-partner figures, or absolute conversion counts. Those are commercial terms, not benchmarks.
- Refresh: quarterly.
Your own numbers will differ from these, sometimes by a lot. That is why we publish them: so you have a real reference point rather than an unsourced average.
FAQ
What is a good EPC in affiliate marketing? There is no single figure, because EPC depends on the vertical, the conversion point, and above all on what traffic you are running. Across our network over 90 days ending September 2026, EPC ranged from around $0.50 in insurance to the low-to-mid teens in financial services, and within a single vertical it varied by as much as fifteen times between partners. A good EPC is one that clears your traffic cost by a healthy margin and holds as you scale.
Does traffic source affect EPC and conversion rate? Heavily, and usually more than the choice of offer. Search, paid social, email, display and rewarded traffic arrive with different intent and qualify at different rates, so two partners running the same offer in the same week can post conversion rates an order of magnitude apart. Any benchmark blends these together, which is why it describes a population rather than your campaign.
Why is my EPC lower than the published benchmark? Most often because your traffic mix differs from the blend behind the benchmark, not because something is wrong. A lower EPC on cheaper traffic can be more profitable than a higher EPC on expensive traffic. Compare EPC against your own traffic cost before treating a gap as a problem.
Is a high conversion rate better than a high EPC? Neither on its own. In our insurance data over this window, one partner converting at over 30% earned less per click than a partner converting at under 8%. Read the two together, and against traffic cost.
Why does my EPC drop when my traffic grows? Because the best-matched intent gets captured first, and incremental volume tends to be both worse-converting and more expensive at the margin. Across four of our largest verticals in the most recent 30-day period, volume and EPC moved in opposite directions every time. Offer mix and conversion-point changes also contribute, so it is worth diagnosing rather than assuming.
How often should affiliate benchmarks be updated? At least quarterly. Payout structures, offer mix, traffic costs and conversion-point definitions move continuously, and a benchmark published without a stated date window and methodology is not worth measuring against.
What should your traffic be earning?
These are network-wide figures across many traffic types, which means they describe a population rather than your campaign. If you run a content site, review property or media brand and want to know what your own traffic should be earning in a given vertical, tell us what you run and where the traffic comes from, and we will look at your numbers against the vertical. If you are weighing an offer that has no history behind it yet, we have also written about how to judge a new affiliate offer with no track record.