Home > Paid Traffic Sources > Facebook & Instagram

---

Getting $$$ales but something is missing to make them PROFITABLE (14)

---

Thanks for stopping by...

So here is the deal, last couple weeks I've been trying to crack the affiliate game.

The result so far is that I can make sales and conversions but I am really struggling to get any campaign profitable.
I am simply writing this to see if we together can see WHAT I AM MISSING.

The offer converts, affiliates are running the same offer on the same traffic sources AND "you do not have a traffic problem",
which tells me that there is ONE crucial part need to fix asap to get this going.


Let's get into it...


Traffic source: Facebook
Offer type: Ecom offer - mass appeal.
GEO: US

Testing Strategy:
Going straight for PURCHASE optimization.
(Never had luck with all the other ones in the last 2 years)

All campaigns are started with 3x CPA Per AD SET and I
usually let ALL ad set run 50% of the budget before looking at the metrics.

Using top 2 landers that i find during research via Adplexity and fb spy tools + similar web.

3-4 different ad images/ videos per ad set gathered during research also.


So let's start with Ecom Offer 1

Payout: $35/ sale

Stats so far:
Link clicks from FB to LP (Gathered from 2 ad accounts) : 1782
Clicks to the offer page from LP: 637
Landing Page CTR: 35%
Conversion: 37
CR%: 5.81%


Rev: $1295
Ad Spend: $2094
Profit: -$799
ROI: -39%



Dashboard overview on the adsets for this campaign

Some things i am noticing here:
Frequency is over 1.00 even though i am excluding ALL past visitors and purchasers. Explain that?

CPMs on some of the campaigns are not so sexy, making the click price even unsexier.
CPC is too high? I need to get the CPC under .90 to be in the green. And trust me when I say this,
I've tried tons of different creatives. None of them seems to get me under 1 buck.


Overview of the Ad level in FB

There is a few ads that stick out here but whenever i kill of the "Losers" the winning ad stops performing.


What have i tried in order to optimize this?
- Tried different ad copy headlines = No difference in performance
- Tried even more new ad images = No difference, the ad starts by getting cheap clicks and then the price goes up to +$1.
- Tried different ad account = No difference
- Tried running the same ads from another fan page = No difference
- Tried different angles (4 different angles) on LP = No difference


So I am not sure where to go from here.

I feel that if i jump to another offer the same stuff will replicate it self.
Because this is whats been happening for me the last 1.5 years when running ads on FB.
I've done a lot of the Shopify game, and i could get sales but never profitable.

Shoot away your thoughts, critiques etc.

---

That was the first thing I noticed as well when looking at the stats: CPC is too high.

It may help if you posted some of the ads you've been running, to get some feedback. (And since they're not getting the results you want, you wouldn't be giving anything away anyways.)


Wow. That basically sums up what I was about to suggest.

What's your targeting like?

Also: How many other ad accounts have you tried?

Makes me wonder if using some really catchy image to increase CTR would get things rolling - to get the campaign "unstuck" from the rut it's in.

What's the best relevancy score?

I'm no FB expert by any means - but want to get the discussion rolling. I'm also moving this thread to the FB forum section to get more eyeballs from FB veterans. Hope that helps!



Amy

---

CPC is high but your other stats aren't too bad, are you just targeting geo/gender/age?

Mobile only or desktop too? Newsfeed ads or everything? (is something eating up a lot of your traffic cost and not converting? Android?)

Nothing wrong with the pixel we use purchase or lead only too.

Just want to make sure you tried 3-4 images and how many adsets? These were just straight ripped from research? Need to try a lot more haha especially new ones FB has not yet seen, the profit often lies in the creative and it will require a bit more testing but usually a winner stands out fast you don't have to follow those exact x$/adset spends

Go deeper into your targeting... will be more work but it should be profitable eventually. Also keep in mind the time of year Q4 is most expensive for ad buying. Lots of people complaining about ROI issues lately but the mass broad approach is getting a lot harder to tame the traffic unless you got some good advantages you can never be sure what the other guys are getting, maybe they own the offer? Are they running right now or is the ads you're seeing from a while ago?

---

I wouldn't worry about the CPCs too much. They're not that high, especially for conversion campaigns optimized for purchase conversions.

How many ad sets do you have live at once? It doesn't look like you're giving your ad sets enough budget to pull in 50 purchase conversions per week. This will definitely cause them to go off the rails.

Try running only one top of funnel ad set with all of the budget going into it. Use a broad audience, exclude retargeting traffic, and toss in your top performing ads. The broad targeting will likely lower your CPC too. Hell, if you have a lot of pixel data I'd try a 10% lookalike audience or open targeting.

Also, I noticed that you have two ad sets with higher CPMs and lower cost per conversions. I'm guessing those are retargeting. Have you tried optimizing your retargeting ad sets for landing page views? It can be a very inexpensive way to drive users back into the funnel. They're already in your funnel, so you don't need to pay a premium for Facebook to optimize ad delivery for conversions.

Consolidating all of these ad sets should help quite a bit.

One other thing to consider - the "Initiate Checkout" conversion is halfway between Add to Cart and Purchase and has been performing surprisingly well for my e-commerce ads lately. It may be worth a test for you.

---

Nice tip, gonna test that one out too didn't even notice that as an option

---

Hey Guys.

Thanks for chiming in with info but apparently i stopped running this offer like 1 week ago because i couldn't get it to move.

Although there is some really valuable info from the comments here.

So why not spin this thread even more with a new offer i am running that has been giving me the best love so far of all the offers i've tried since starting aff marketing.




Let's get into it...




Traffic source: Facebook
Offer type: Ecom offer - back pain solution product.
GEO: US
CPA: $30 per sale


Angles used:
I researched and saw that everyone was running the exact same angle towards
peeps with back pain.
So i did some further research and found out that
elderly are prone to fall risks when they are outside walking.
Depper i dug and found out that this offer would be perfect for this demographic.

So the angle used is to "Prevent fall risks in elderly by using the offer product)


Testing Strategy:
Going straight for PURCHASE optimization.

Started with 13 ad sets at $15 each, targeting different interests all from age 55+
Men and female
Auto placements,
7-day click view


Each adset started with 4 different images with SAME headline


NOW this offer has been running since Tuesday this week and i've tried a lot of different things to improve the metrics.
Below you will see a before stat and after stat.

Total sales since Tue: 9 sales which brings me $270 in revenue.

LP CTR beginning: 21% --- After changes: 37% = Changed some copy, and instead of using the "Check availability button" i added 3 buttons age so they can click their own age.
CPC (Link Clicks) Beginning: $1.60 ---- After changes $0.96 = What happened here was that i killed the bad performing ad images and added 2 new that are now running.
Link CTR Beginning: 2.69% --- After changes 5% ish = see above.

Offer Conversion Rate: 5.15%

So 9 sales so far with $270 in rev
Spend: $366
Roi -25% and -$90

Here is a snapshot of the ad sets running:



So i am not sure where to optimize more here?
I have increased all my metrics since day one and still aren't profitable.

Is the angle weak? I mean i'm getting 35-37% view throughs to the offer.
The offer is converting as it has over 700k visitors last month according to similarweb, and my aff managers says it's one of their hottest offer right now.

What would you do here?

---

Why not test multiple angles? More images? More headlines?

Looks like a promising offer with promising stats - so it's not the offer.

Why not brainstorm to come up with at least 5 angles + 10 images per angle + 5 headlines per angle to see which combination works best?

Your creatives will create a ceiling that no amount of optimization can break through - it's starting to sound like that may be the bottleneck.



Amy

---

Amy thats some fantastic input there.

I have one question regarding new angles.

So you have let's say 5 new angles.
Do you then change the ENTIRE pre-sell page or only the headline and the first sentences to match the angle?

---

That's up to you. Go through all your funnels step by step and see how does it "feel" to you. Personally I always have certain "variables" coming back in the ad, prelander and offer page (this is not always possible ofcourse, since advertiser has to adjust stuff on the page "just" for you.), this way the consumer has the feeling it's a natural flow.

If you run a certain ad + prelander with same style, but your offer page doesn't match at all with colors/layout etc, big chance your CR will be shit!

Other thing; am I drunk or your CPC is waaaaaaay to high? No matter which targeting audience, geo or vertical you pick, a CPC ideally should be between 0,01 and 0,10ct. Ofcourse this depends on so many factors yes. But if 1 dollar is possible, 0,50ct should be possible too and even 0,10ct IMHO.

---

I would start testing the different angles without changing the presell page. At some point in the past, when I was learning how to make this AM thing to work for me, I've spent way to much time on this in an attempt to optimize the whole flow, so it's 100% consistent from the ad all the way to the offer ... in the end, it didn't make all that much of a difference. Sure, it did have some impact, but even without being completely consistent, I was able to quickly spot what angles show more promise ... then I could optimize the whole funnel for those that performed the best.

---

Stickupkid and Matuloo have both made very good points.

On the one hand, it's good to just change ONE variable and keep the rest constant like Matuloo suggests - for example test different ads with the same presell page. But as stickupkid has pointed out, ONLY do this IF the ad -> presell flow is natural.

The good thing about testing one variable and keeping the rest constant, is you will see which one is the best candidate for that variable. Bad thing is that if the "constant" part of the campaign is ineffective, then you can test lots of candidates for that one variable and STILL not see profits. (e.g. Say you want to test multiple images for the same headline + ad copy + presell page, if your headline + ad copy + presell page is a bad combination, you can test a million images and still not get the campaign to convert well.)

What I would suggest in general, is create a separate presell page per angle, for multiple angles (say 5), and then develop a separate set of ads for EACH angle/presell.


For each set of ads, you'll need to test image + headline + ad copy. Use any testing approach that makes sense to you. The following is an example approach:

1)Write a single ad copy and 3-5 headlines, and find 10 images.

2a)Produce 10 ads from the 10 different images, choosing one of the 3-5 headlines for each ad. Run the ads to collect some stats, then identify ads that are performing the best, take their headlines and images and create combinations and test those.

For example, let's say 2 of the 10 ads seem to be doing well (with hopes of breaking even or better; or showing low CPC / high CTR): Ad6 = Headline2 + Image9 and Ad7 = Headline1 + Image5

I would then create ads from all combinations of the headlines and images and add them to the test, i.e.:

Ad11 = Headline2 + Image5

Ad12 + Headline1 + Image9

You get the idea!

2b)Alternatively, you can product an ad for every headline+image combination - so for example 5 headlines x 10 images = 50 ads. This approach can be more effective in uncovering the best combination, but will also take a lot longer and require a considerably bigger budget. If you DO go this route, it would help to pause ads that have high CPC / low CTR early on to save on budget. (Do keep in mind though that the best-converting ads aren't necessarily the ones with the highest CTR / lowest CPC; however, if you see ads that have overly-high CPC / overly-low CTR, you may be able to tell that they probably won't turn out profitable - so pause them to save money.)


The above are just some ideas. The key is that when you KNOW you have a good offer, don't give up after testing only a single angle (however kickass it seems to be!) and a few ads. There are lots of ways to get creative when brainstorming angles - for example:

-We have an entire forum section on angles - go through some of the best threads.

-Read Cashvertising (or if you have, flip through it again).

-Identify online forums and facebook groups where your target audience hangs out, to identify pleasure and pain points. Borrow their language/jargon when crafting your ad copy and presell page copy.


Have fun!



Amy

---

Here's one tip worth testing to keep the ad -> presell flow natural. Have a script that dynamically customizes the page with the same or similar image/headline/offer from the ad.

With FB, you can automatically pass the adid variable to the presell page, to let the presell lp know which ad was clicked.

This allows you to have multiple highly targeted pages with just one initial page.

I'm in the process of rigging up the javascript to make this work 100% dynamically, but I've had FB traffic lp CTRs as high as in the 60-80% range using a manual version of this method.

---

Continuing from my post above - I forgot to mention one thing: After finding one or more angle+image+headline combos that work well, feel free to test different ad copies for each of the winning angles to improve performance further.

That's a great tip! I've observed positive results from this as well.




Amy

---

Yep I agree with you Mr. Baffoe! We have a simple way using airtable which will generate parameters on our links according to which angle/geo is used and creative selected in the table we created which will dynamically change the lander to respect the same copy from FB ad to lander.

The CTR is much higher, CVR is still similar

---

Home > Paid Traffic Sources > Facebook & Instagram