Finding a good affiliate offer is a lot like being the only person who knows which cow swallowed a priceless ruby.
You're going to get very rich in the long term, but in the short term you're going to have to feel your way through an awful lot of shit.

Fortunately, unlike the precious-stone-cow-excrement scenario, in affiliate marketing there are ways to reduce the amount of crap you'll have to run through your fingers.
Bad offers, like bad poker players, have tells.
Here's fourteen of them to watch out for - before you start spending money on testing, AKA feeling through the first cow turd.
14: Slow lander
This one's pretty obvious but you'd be amazed how many people don't watch out for it. Test the loading speed of any offer you're considering from the geo you're planning to run it in.
As I've written before, people are horribly unforgiving about loading times in the West, and not much more forgiving elsewhere. If an offer takes more than 3 seconds to load in the USA, Canada, or Western Europe, or more than 5 seconds elsewhere, it's going to bleed clicks and it's going to bleed them harder than a haemophiliac with a razor fetish.
13: Horrible signup process
We're going to go into more detail on some specific ways an offer's conversion process can suck below. But in general, there are a thousand ways to have a terrible funnel.
Get as far into the signup process for any offer you're considering as you can. Proxies to get around geo blocks and virtual machines running Android are your friend here. Imagine you're a user who has just signed up through the traffic source you're considering using. Would you actually get through the signup process?
12: Terrible reviews online
Terrible reviews aren't always a killer, but they don't help.
Reviews on the offer page are particularly bad news. If you're considering promoting an app and the Google Play page is wall-to-wall one-star reviews, chances are that's going to dampen conversion rates somewhat.
11: Just Plain Doesn't Work
Newer affiliates may scoff at the idea that a network would have an offer on its books that is literally impossible to convert. But trust us cynical old hacks, it does happen.
404'ed signup links, broken forms, crashed servers - particularly if it's an older or more obscure offer, it's easy for these things to get overlooked. Test the offer before you spend cash on it!
(Often this is a comparatively simple fix too - so do tell your AM if you find an offer that you'd like to run, but has had a visit from the 404 Fairy. )
10: Broken conversion pixel
Related to the one above - sometimes, and by sometimes I mean "surprisingly often", an offer's pixel won't be installed correctly.
This is another reason to try and run through a signup yourself if you possibly can, to see if it registers on the system. And if you're getting a load of traffic and no conversions at all at a later date, this is also something to query with your AM.
9: If it's too good to be true...
So here's a scenario. You see a new offer on a new network. And its payout is incredibly high. Like $20 for an app install high.
The correct response at this point is not "Holy shit, that's money for old rope! I'm gonna be RICH!".
The correct response is "What's wrong with it?".
Higher payouts are good - but ridiculously high payouts mean either a) everyone involved in the process had a traumatic brain injury to the parts of the brain involving money or b) there's something you don't know, and they do, and it's almost certainly not going to be good for you.
8: Unrealistic payout pricing
On the other side of the coin, just because you're looking for smaller payout offers doesn't mean you should promote a long-form insurance lead gen that's paying $0.25 per lead.
Advertisers frequently try to cheap out by offering a payout so small you won't make your money back. Look at comparable offers and think about the cost of your traffic. Can you actually make a profit off the scraps they're offering?
(If the offer looks good but the payout's terrible, it's always worth having a quick chat with your AM and asking if there's any movement room there.)
7: Cap Lower Than A Hat On A Centipede
Always, always ask about the daily cap.
For newer affiliates - a daily cap is a limit on the maximum number of conversions a day.
Sometimes it's stupidly low. And sometimes it's not per-affiliate, but per-network. If an offer has a cap of 100 conversions a day, a payout of $3, and you're pretty sure five other guys are already running it, that's not going to be a fun time.
6: Already Saturated
There's often a single killer offer that goes through the affiliate industry like an extremely profitable winter vomiting bug. But unfortunately, after everyone's finished their 48 hours of violently spewing money from all orifices, what you're left with is immunity - in the market.
If you've been seeing an offer running everywhere, for weeks, the chances are the low-hanging fruit for that offer have gone (in the geo that you've seen it run in). You won't get easy conversions from running the obvious angles.
The good news is that, just like the winter vomiting bug, immunity is partial. If you can find an untapped pocket of traffic for that offer, or a new angle to promote it on, it'll work just as well as it always did. But if you're just planning to do the same thing everyone else already did for weeks - don't bother.
Incidentally, with offers like this, keep a very close eye out for them opening in new geos. If an offer that has worked very well suddenly appears in a new geo, that's the time to descend upon it like a starving barracuda on an incautious SCUBA diver.
5: Terrible, Horrible, Not Good Trust-Destroying Writing Errors
Bad copy is a bit of a risk for offers in general. They often assume that actually selling the thing they're trying to make money off is our job.
But what you really need to watch out for is the kind of copy error that screams "THIS IS A SCAM! WE'RE TRYING TO STEAL YOUR MONEY!"
Here's a classic example: the iPhone sweepstake that suddenly switches to promising a Samsung on the second page. Or the flashlight that's listed as "75% off", but where the numbers listed aren't actually 75%.
(In general, confusing pricing kills conversion. Watch out for it.)
Note that these trust-destroying errors are at least as common on white-hat offers. Just because it genuinely isn't a scam doesn't mean it can't look a lot like one, particularly to a visitor who has arrived from a random advert they saw on a clickbait site.
4: Non-Responsive Design
2002 called, and it's expressing an ongoing concern with the number of offers that are still stealing its web design.
If you are planning to run an offer on mobile, do not assume the offer's lander design will work on a phone, even if it's expressly advertised as a mobile offer. Check, preferably on an actual phone.
You'll be amazed how many offers just straight-up don't work on any screen smaller than the original designer's 27" iMac.
3: Wrong Language
The offer's targeted to Vietnam. The lander's in English.
This is super-common. And no, it probably won't convert.
However, do watch out - some countries are more bilingual than others. For example, in India English creatives will often work pretty well.
If in doubt, ask on STM whether the offer's language is a problem for the geo. Someone will be able to tell you.
2: Terribad Design
Yes, design can kill an offer all on its own.
Can you tell what elements of the text are meant to be clickable? Does the design of the page actually give you a headache? Does the typography run together and make it look like it's a 12-year-old's First GeoCities Site?
(For those of you too young to remember GeoCities - feast your eyes. )
If so, users are probably going to bounce off it super-hard, and not make you those sweet conversion dollars.
1: CTA Invisible Without Forensic Tools
And on the subject of design - load up the offer page. Can you immediately see where to sign up?
If not, the chances are that your visitors won't be able to either.
Amazingly, it's still the case that some offers haven't realised it's a good idea for their calls to action to be visible.
This one is less of a hard-and-fast rule than others here - and do remember that long-form landers convert well even if the CTA's waaaay down the page. But as a rule, if you can't find the signup / buy now button, your visitors won't be able to either.
And that's it! Check for these errors and you'll probably be able to cut the number of offers you have to test in half. Any comments or questions? Or do you have more tips for spotting loser offers? Post 'em below
Very good post, I think I have experienced all 14 already and there is no sign of them stopping coming back
And what an awesome title photo LOL 
Thank you!
caurmen strikes again!
Similies on point.
Great stuff as usual, caurmen. Thanks!
Thanks caurmen!
Great guide - the attention to detail when picking out offers is definitely valuable when getting thing started.
Specifically appreciate the shout out to Geocities haha
Thanks, I jsut experienced this today with an offer I was about to run--4+ second loading time on each page of a two-page sign-up process, not to mention it's been an hour and I still can't find the registration email--not even in my spam folder. I've had a couple of offers where I wondered if I was just being too critical...well maybe I wasn't. Thanks for the list.
Epic thread from Caurmen as usual 
I enjoyed this post
That GeoCities reference got me! Great tips tho thank you.
Good list but I had great offers DESPITE some of the things listed.
Slow lander being one of them, especially working internationally sometimes what's fast to you is slow for them and vice versa, therefor if you want to be god level optimizing get yourself cheap amazon vps GEO close & pull up the pages in RDC to see how it responds locally. I know there's sites that measure this type of shit for you but I'd rather go this route sometimes.
Same goes for terrible sign up process & reviews. If your offer is the only one in the niche, that niche could be so wonderful that those things don't ruin it. If reviews truly mattered then how come so many people killed in garcinia? This holds extra weight in international niches where the amount of offers is so limited.
Same goes for already saturated point. Just because everybody is doing something in one place doesn't mean they're doing it in another, in fact MOST OFTEN do they not do it in another place. many times I was struggling to break even in one geo & then a simple geo change turned the campaign into a 500% ROI thing in matters of hours.
Same goes for spelling errors, I've ran same copy for years on landers that I eventually got to reading (was originally ripped, and I never really read it) and found TERRIBLE mistakes. Didn't prevent it from making absolute fucking killing.
Same goes for #2 and #1 I guess.
Those things are extremely important, but it doesn't mean it won't work.
@pierrecabas - Yep, completely agree about testing the offer speed in the geo you're planning to run it in. As mentioned in the post, you have to do this - just because an offer loads slowly for you in the USA if the offer's in Singapore, that's not a sign at all that Singaporean visitors will have the same experience.
And saturation is definitely geo-specific too. Obviously if an offer's saturated in the USA that doesn't mean it's an automatic lose in Brazil. (In fact, a saturated offer opening in a new geo is great news.)
Spelling mistakes are a complex one. My experience is that they matter much less on landers than offer pages. The closer to the conversion (which normally requires trust) the more important they seem to be.
Good points - thanks for the comments!
There is this website, I used to use it to see the offer's page, It's perfect with many features like the carrier you want to use in each country and it is worldwide, The bad thing is that the limited edition is verry limited like if you wanna test with 3g connection it only give you one test and the pro edition is expensive for beginners like me
https://affilitest.com