The Scaling Fallacy: Life As An Affiliate Whack-a-Moler
Jack creates 10 campaigns in 10 countries on TrafficJunky:
Here are his morning stats:
United Kingdom: +$12
JACK’S TOTAL PROFIT: $138
Bill creates 1 campaign in 1 country on TrafficJunky.
BILL’S TOTAL PROFIT: $90
Which affiliate would you rather be?
Come on, it’s a no-brainer.
Bill is in a much better position.
Let’s look at what each affiliate has to do in the morning:
Jack has to:
- Check stats for 10 different campaigns.
- Optimise creatives in 6 different languages.
- Search for the best offers in 10 different countries.
- Manage bids against hundreds of competitors.
- Put out fires all day before he can work on any new campaigns.
Bill has to:
- Check stats for 1 campaign.
- Optimise creatives in a single language.
- Track the best offer in a single geo.
- Manage bids against a handful of strong competitors, and many more weak, distracted competitors (like Jack).
- Find new sources to test his campaign.
Now imagine the two affiliates hear about an amazing new traffic source. We’ll call it Exoclick for the lols.
They both want to scale their campaigns to it.
Bill, having much more time on his hands, goes first.
He knows Germany inside out. He knows what works. Scaling is merely a case of busting open Voluum, tapping Duplicate, and changing the tokens:
His stats now look like this:
BILL’S TOTAL PROFIT: $154
Jack, on the other hand, is starting to sweat.
He can no longer make sense of his stats without spending 40 minutes staring at Excel.
He has 20 campaigns in 10 countries on 2 traffic sources.
And his stats look like this:
United Kingdom: +$3
United Kingdom: -$88
JACK’S TOTAL PROFIT: $2
Jack can only manage so many campaigns effectively.
Even then, it only takes one rogue offer (in this case, the UK) to wipe out his progress in the other markets.
Jack, like many affiliates, has chosen to stack the deck in somebody else’s favour. He needs everything to go right, or his model is inefficient.
Let me tell you:
In affiliate marketing, fucking rarely does something ever go right — let alone all of it.
Why is that?
Every market requires a complex understanding of conversion rates, clickthrough rates, and the dozens of metrics that wreak havoc on them. Even this may prove irrelevant if you fail to catch hot offers early on their upswing.
So how are you going to manage the above, MULTIPLIED BY TEN, when your attention is DIVIDED BY TEN? (Plus porn.)
There’s a pretty simple solution.
Look at what happens if Jack only runs campaigns in one country — Germany*.
The stats completely change.
His profits would be $98.
(And his ad spend considerably lower.)
*This is an example, not an endorsement for advertising in Germany.
The stats improve, but that is purely superficial. What we should really be interested in is what gives us the best chance of success and true ‘scaling’ going forward.
It’s the revitalising effect of removing so much deadwood that has the real impact.
Perhaps Jack’s morning to-do list would start to look more appealing.
No more staring glumly at the ‘Monday Headbanger’ with rows of campaigns disintegrating in puffs of smoke.
His blood pressure would drop, his sense of direction might return.
More of his working hours would be spent looking for opportunity rather than putting out fires day after day, every day, until there’s nothing left to save.
This is the price an affiliate pays for spreading himself too thin.
* * *
Volume is sacred in this industry.
It goes against instinct to start culling campaigns — especially those that are profitable — and it isn’t made any easier by the pines of your affiliate manager (“Revenue, revenue, revenue!”).
But remember: volume is irrelevant if it’s fractured across 20 offers in 10 countries.
You can’t leverage the sheer number of your campaigns. It’s your dominance in each individual market that counts.
On the flip side, achieving dominance in a single market is easier when you don’t have a desktop scattered with half-finished landing pages for every single nation in the EU.
Attacking a single country with brute force and finding something that works?
Now that is a scalable business plan.
Pussy-slapping 15 nations in to converting at a 10% ROI then screaming like a banshee as you realise Nation 16 lost the entire fucking lot with one dead offer?
Not so scalable, mon ami.
* * *
1. Spreading yourself too thin will cannibalise any profits you might have made.
2. Spreading yourself too thin will prevent you from scaling the campaigns that are actually worth scaling.
There is a balance, of course.
You do want to explore new markets. I can’t emphasise that enough.
Without exploring, you won’t find any profits period.
Newbie American affiliates who refuse to exit their home market are probably familiar with consistent big fat zeros. That’s because domestic arbitrage… is a bitch. There’s a long line of capitalist pigs that got to the table before you.
If you are going to flee in to foreign markets, any experiment should be carried out with a degree of control.
Foreign campaigns should be planned, not ripped with reckless abandon.
Do some calculations on the required CTR/CVR metrics before you even start.
(You’ll know if you’ve planned a campaign — you won’t have French translations in one tab, German in another, and a Turkey flag occupying Photoshop.)
You shouldn’t hesitate to scrap these campaigns if the profit achieved is barely worth a pot to piss in.
So you find a country somewhere close to the Russian border that produces a steady $10/profit per day.
Should you pursue it?
Should you fuck.
Maintaining a $10/day campaign is like pulling down your pants, cracking open your skull, and taking a hearty dump on the sacred membranes.
A complete waste of mental resources.
The only possible reason to focus on these campaigns is if you don’t have the capital to compete elsewhere.
And in that case, your objective should be to raze the marketplace, pillage whatever you can, then move on to richer pastures.
* * *
(Yes, there is some here. Eventually. I put it at the end to confuse the undeserving bastards who didn’t make it this far.)
1. Pick a very small number of countries to operate it.
2. Ensure those countries have the same language so you can recycle creatives.
3. Kill any campaign that requires time to manage but can’t satisfy your income objectives — regardless of profitability.
4. Don’t scale at all until your existing campaigns have settled from peak ‘just-been-launched’ performance in to stable, reliable numbers. (Or you’ll be left with an empire of sand castles.)
5. Understand and respect the one metric that Voluum/CPVLab can’t calculate: opportunity cost.
Whatever you work on today, it comes at a price.
13 CommentsLeave a comment
always a great lesson…. even if i did have to read to the end 😉
I generally agree with you, but I disagree with everything you’ve said in this post. I would rather be Jack, hands down. What if the expansion potential wasn’t in Germany, but rather in Belgium or Australia? Guess what, Bill is SOL. Worst case scenario, all Jack has to do is pause the campaigns that aren’t profitable and then focus all his attention in the new market. Pausing a campaign takes a few seconds. You’re giving contradicting advice. On the one hand, you say volume is king and you should always be looking to scale. That’s exactly what Jack is doing by exploring additional geos. Not Bill. Next, why in the #@&$*# would you pause a campaign that is profitable, even if it only nets you $10/day? If it’s not worth your time, simply let it run without interference until it’s no longer profitable. This takes zero effort to maintain, other than quickly glancing at your profit column once per day. You might scoff at $10/day, but imagine if you had 10 such campaigns going. That’s now $3k/month for which you have to do no work.
Like I said, you do need to explore different geos to find one worth chasing, but running a ‘portfolio’ of campaigns across multiple countries is always harder to scale than one country keyed in and portable to other sources. Yes, volume is king — that only applies if it’s concentrated in a single market. You get no advantage by fragmenting it across the globe.
If I had 10 successful $10/day campaigns running, that’s exactly the type of situation that leads to managing a portfolio of unscalable campaigns. The margins slip with time. There’s no such thing as $3K/month in CPA arbitrage where you don’t have to do any work.
Even if you’ve mentally prepared yourself to pause campaigns when they lose profitability, there’s always a temptation to keep the small earners alive by switching an offer, changing a LP, etc etc. Just not worth it for $10/day.
i notice you talk a lot about adult dating. is there any other big niches now outside of the adult world you’d say would be worth working with in CPA?
Keep the content coming you’re the man!
I think Finch was a little extreme in his example to make a point and teach a lesson, but overall the philosophy is good.
Here’s the major flaw in your thinking: you believe these small campaigns take no effort.
Each campaign in adult means you have to upload creatives every few days (banner blindness), you have to deal with offers pausing or capping, and you have to worry about potential bidding wars.
Wasting time and energy on $10 a day campaigns is what prevents from spending time to make that $90 Germany campaign, into a $900 a day campaign.
Once a campaign is large enough, it only takes a few tweaks to 2x the campaign. Being the best means you have competitive EDGES. How are you suppose to have the EDGES and ADVANTAGES if you’re worried about $10 a campaigns?
There’s two different mindsets in affiliate marketing. The guys who make millions a year have a DOMINATION mindset, while “Jack” has a “survival” mindset.
The money is in mastering a few countries / traffic sources, and being the best in them.
When you look at mobile offers, there’s a huge array of non-adult stuff — and growing fast. For the web specifically: gaming, business opportunity, weight loss, muscle building, skincare, and the entire financial industry. Dating has been my market of choice for a long time (+ more recently adult), but it’s by no means the only niche. I think it’s probably the most accessible though.
Thank you for great post.
You are talking about geos here, but how about traffic source. What is the best way to test them for newbei?
For example I have spent 166 usd 4746 clicks 12 conversions cpa 0.9, try apps and mobile web targeting, didn’t do optimisation. The network have bot traffic according to my tracker.
I plan to do publisher filtering, choose optimal ctr/cr creatives, redo best ones, cut off unprofitable and play with targets. Do I plan to waste my time?)
PS I am at the beggining in mobile affiliate. This is my fifth campaign. Only one a bit profitable(5 usd per day).
If you’ve launched 5 campaigns and one is profitable, then you’re doing better than most.
Every traffic source is different, as is every offer you choose to run on it.
Before you run a campaign on a new traffic source, you should do an audit of what other advertisers are running on there. Is there affiliate activity? If so, there’s your proof that the platform works. Next step is to research it heavily. Contact the network and extract as much info as you can on what other performance marketers are doing — countries, niches, offers, etc — then compare it to what you see with your own eyes. Access the network’s websites using a proxy in different countries. Try to nail down any signal of recurring offers, countries of high affiliate activity, etc etc. Once you have all of this mapped down, then you can go ahead and start spending money on your own campaigns.
As for bot traffic — depends on what type of bot traffic you mean. If it’s fake clicks and you’re running CPC, then send a report to the network and demand a refund. If it’s dud impressions, blacklist as many of the offending targets as you can. Mobile networks still have a long way to go with this problem to be honest. Hence the need for a bigger testing budget vs. web campaigns.
In my case there is huge competition on this traffic source and I need to test different angles and offers. Also I have CTR below 0.7 – this is not good for traffic volume.
To sum it up
– research yourself
– ask rep(just to know if you could trust him)
This guide could help research http://iamattila.com/iamteacher/exclusive-iamattilas-ultimate-spying-guide-for-mobile-media-buyers.php
There were only one bot for this time, but I am monitoring its activity. Bot – publisher that don’t click on LP. The bad thing with voluum is that you couldn’t see and maintain raw data. So you couldn’t do reporting to publisher. IMT is better in this case. Really think to move on it(in case it has SQL access). The thing is that newbei is seeking for optimised server for low price)
Well said. The key takeaway from this post is FOCUS. With laser focus you will be surprised how much you can get a single campaign to go from break even to profitable. The scaling shouldn’t happen until you have a rock solid steady base with killer split tested creatives.
Thanks Finch, you always deliver quality. I have been struggling a lot with scaling and i am afraid it might run me to the ground. As a noob i was happy to start getting the hang of things and be able to create a campaign and get it to profit but i ca’t seem to go any further.
I mostly run Pop/Redirect and Landing page and interstitial ads. I can’t scale to bigger similar traffic sources as they all require $1000 plus commitment which i can’t commit to to at the moment. The few that i have tried end up doinglike $10-$20 profit /day for a short while and then just dies off after a while.
(1) Is it advisable to port campaigns like this to very different traffic sources(pop to banner) ?
(2) Any advice on how to go about this will be most appreciated