Ask Me Anything - Facebook Ads [2M+ Ad spend]

[Pritesh]

Hey guys,

I am here to answer any questions you might have about Facebook ads.
I have around $2M+ Ad spend experience on Facebook (85% agency, 15% personal). Total paid ads experience of $7.2 M+ if you include Google Ads.

I am here only to talk about Facebook ads, as my experience with Google Ads is extremely one dimensional and I cannot give you any advice on that despite it being the majority of my ad spend.

I have run all types of ads on Facebook from app installs to dynamic product ads, from whitehat to cloaked ads, from lead gen to conversion, for ecom stores to small businesses almost everything. I have quit the agency to start my own and right now I am running only eCommerce ads for passive income and lead gen ads for offline service-based businesses.

I reserve the right to respond to questions that might force me to reveal my ad strategies, I particularly will not entertain any questions that directly ask me for my personal ad testing techniques, etc. I have my own approach to doing things, and revealing it would just not be in my best interests. I can, however, provide you critique on your ad testing approach.

Please do not ask me long-winded questions or stack 3+ question in a single post, I urge you to value my time.

All answers are my opinion and based on my experience, YMMV.

Finally, I am also working with a select group of individuals who need lead gen services for their business for video testimonials for my new agency, you can check it out here: FOR HIRE - Looking for Clients/Leads/Sales For Your Business? I am Your Guy

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[Pritesh]

I'll go with one.

If i wanted to get cheapest US traffic with low budget, how would i go about it?
By cheapest i mean something $0.05 cpc and not really targeted.
The fact that its US based would be enough.

Thanks for AMA.

Let me preface by saying non-targeted traffic can never get you stable cpc metric, there are just so many factors involved in it that you'd be looking at $0.05 cpc one day and $0.90 the next. However, there is a way to almost guarantee stable cpcs within your range.

Here's a campaign structure I'd do if getting the cheapest US traffic is my goal. Note this involves a lot of testing, but once you find the perfect combo, you'd be unstoppable.

First, you make a sandbox campaign to find out your winner interest + creative combo.
In the campaign, I'd first test on just one interest which creative delivers the lowest CPC.

So just one ad set (PPE Objective) with one big interest 1Mil+ audience and test 5 creatives. I'd also test run a dynamic creative adset if you want to test headline and ad copy as well. Duplicate the adset 3 times so you get statistically significant results (You don't need to do this if you're budget-constrained)

Now you must have a single post id which is a clear winner. Now every single adset and test we do going forward will use this post id and nothing else.

Next, we test what objective brings you the cheapest CPC. PPE Or traffic.

In the same sandbox campaign, make an adset with traffic objective optimized for landing page views. Again, duplicate the adset 3 times so you get statistically significant results (You don't need to do this if you're budget-constrained).

now it would be clear which Objective-Adset-Ad combo delivers you the lowest CPC. Turn everything else off and verticle scale that adset by duplicating it in the same campaign 2-3 times.

Repeat this process for more interests, you can use the same post id on different interests as well rather then repeating the entire process if you're budget-constrained.
Now you have a bunch of Objective-Adset-Interest target-Ad combos, here it's upto you which you want to turn off or keep it going.

it is important to find the winner post id first, because the more social proof you stack onto a post id, the cheaper your metrics get.

Finally, if you're looking at doing ad arbitrage using FB ads, think again, you'll get banned sooner or later.

 

[Pritesh]

The FB ads interface is pretty complex - do you use tools like AdEspresso/Qwaya to simplify the setup? If not, why?

No I don't.
Honestly, the Facebook ads interface is the most easiest interface, you don't even have to bother with what 90% of the stuff does unless you're doing some advanced stuff. It is miles ahead of Google Ads when it comes to user-friendliness and dumbing things down.
Just take the Facebook Ads Blueprint course by Facebook itself or watch a few live campaigns builds, it looks more daunting than it is, trust me.

If you lost all your knowledge and had to start from scratch, as a total noob on (Facebook) Ads:

How would you go step by step to learn and get good at it?

Or different way:
How would you teach a noob like me to get good at Facebook Ads?

PS: thanks for the AmA

When I started, Facebook's own Ads Blueprint wasn't a thing. But now, if you want complete handholding through the entire process there is no better course. Obviously, when you get the essentials off I'd recommend learning from experience and consuming some advanced ad strategies.
I'm always an advocate of do first fix later, don't just consume content, start doing it while you learn.

Facebook's own learning catalog is quite comprehensive (it also has a few advanced ad strategies now): Facebook Blueprint : Blueprint

 

[Pritesh]

Thank you @Pritesh

I changed my ads from traffic to conversions, as you suggested in the other post.

I'm spending slightly more, but sales are also more numerous. Now I'm only using that goal.

Glad to hear you've taken action on my advice and seeing results right away.
Yes, the cost per click metric will obviously be high because you're going after good traffic and not traffic that will just "window-shop" your products.

Now I am taking new photos of the product for do the tests, but sometimes I have some difficulties. If you had to choose photos of products, which do you think perform best? Those of the product on a colored background (more "professional"), or that of a product that is used (perhaps made by the smartphone, which seems "amateur")?

From my personal experience, 9/10 times an amateur photo of the product will always outperform a professionally made product image on a colored background or white box.

A professionally taken photo just looks like a stock image and falls in the banner blindness category. I'd also try adding some weirdness or try to induce the feeling of "something's off" with the ad creative, I know it's pretty vague, but the more weird or outrageous your product image is the more people will stop their scrolling and actually look at your ad text. If you can't do that a simple red border to the image or a red arrow pointing to the product also works.

 

[Pritesh]

I hope my question isn't out of line though. Did Facebook make you a media partner after spending $2 Million in ads? Are there any exclusive benefits that comes with having such huge spend history with Facebook? I'm curious

No, no facebook rep assigned for my personal account, the agency I used to work with does have a FB rep assigned. I do however have Google rep assigned to me.

FB reps come in different tiers and the higher you spend the more knowledgeable reps you get assigned. The reps assigned to us barely provided any valuable input and in most cases had much lower knowledge about running Facebook ads than even the new recruits at my previous agency. The only visible feature after being assigned a rep was unlocking a few audiences targeting options, but they weren't a huge advantage, in fact we never used them.

What's the general flow of starting a new ad campaign for a new product or service? I've heard that you have to test ads and build audiences first. Can you please clarify the process?

The testing process is the most crucial thing you focus on for launching anything new, but the process depends on the scale of the business.
Normally I'd test 100+ variations of adset+ad for ecommerce, but running ads for small businesses wouldn't require more than 5.

About your "test ads and build audiences first." I'm assuming you're talking about seasoning the pixel?

You can refer to this post, where I've broken down a testing process for traffic, for other objectives the process is somewhat similar. I find a winning post id, then scale it while testing other factors.

What’s the simplest and high ROI campaign you’ve run?

Realtors, just one converted lead for them is minimum $6k so the metric is pretty arbitrary as I've been involved in many niches. For a general niche lets say Ecommerce: 5x tracked ROAS after $20k spend was my highlight adset.

Have you successfully run ads on FB for B2B services? If so, what kind of conversion rates did you hope for and did you achieve the conversion rates you expected?

Online B2B services (web design, etc) usually have pretty low ROI for me or breakeven and dependent on customer lifetime value.

If we're talking SaaS B2B apps on the other hand, they usually have good ROIs 2-3x top of the funnel (not factoring in stick time or lifetime value)

 

[Pritesh]

1) For a totally new e-commerce business selling dry food stuff which has just launched, which no one knows about, what would your advice be with regards to FB ads? Should I run ads with Brand Awareness/Traffic objective or Conversion objective? Or something else?

2) While I know how to use the FB ads interface, should I outsource the whole ads strategy to those freelancers on Upwork? I was thinking those guys are well versed in getting traffic. What's your thoughts on this?

1) I've given a few ad strategies in my previous replies you can use them to test new products. I'd personally run Conversion > Optimized for Landing page views for starters to test which ad creative performs best and then go on from there.

2) I wouldn't know, never worked with a freelancer before.

I was thinking those guys are well versed in getting traffic

Your objective is to get sales right? Anyone can drive you traffic, heck if you pay FB and run traffic campaigns they will throw traffic on your site all day long, will they convert? Highly unlikely.

For someone that doesn't have their own product and they just want to make $1k/day (more or less) passively - what should they do / what offer should they promote?

Man, that's as broad as you can get asking a question. No one can tell you what to sell, you need to do research on it yourself and then test it yourself.

But, in my opinion, Tangible products, and services that fall in "need" category would obviously perform better than those that fall in "wants" category.
If you find a pain point in a niche, Facebook ads is the perfect medium to promote it. It has so much data on people, you can literally hit your audience with their pain every day until they give in and buy your product or service to curb that pain. But, finding that pain and niche is something you'll have to do on your own.

 

[Pritesh]

2 Questions

Have you run many ads into Messenger/Messenger Bots and if so how have those gone?

And I have someone who wants to run ads in the real estate wholesaling niche but they don't have an email list or any audiences to go off of. Any suggestions for figuring out how to target something like that aka distressed homeowners?

Yes, I've run automated message bot ads, they were a good solution for people who are hesitant to give out personal details via lead forms. I cannot, however, comment on the quality of the leads.

For real estate question:

1. You can target interests including Zillow, NAR, etc
2. There is also a secret interest not many people know. FB had an interest category of "Likely to move", but due to some political pressure they had to remove it, but FB wasn't gonna give up on juicy Realtor ad spend. Now the interest to target to find these people is "House Hunting". I shouldn't be giving this away, as the term is not easy to find, but there you go.

Yea that makes sense, I guess I could also use the education > degree > business/law, etc
In terms of image and video content for the recruiting ad is there anything super different than marketing a service business?

I´d create high-quality production videos about what benefits there are for you working with this specific law firm, for instance: parental care, remote working, high salary, paid vacation, etc.

If you're targeting nationally, go for a high production value video, but for just targeting locally, I wouldn't recommend spending much time on it. Just put your benifits in the ad body itself, that should be more than enough.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[Pritesh]

Hi @Pritesh, I have 2 questions.

I've just started using FB ads 2 months ago, and have had some success.

I will ask a question and share a strategy I use. Curious to hear your feedback.
I understand that there is no 1 certain answer for this system, but maybe you can find something I am doing wrong. Thanks.


I) What is the best way to use FB ads to prove a product is worth selling?

1) Run targetted traffic to a landing/sales page
2) High CTRs, engagement, and ATC events are good signs that the product will make it

II) What is the best strategy to first test, and then build and scale conversion campaign for selling a physical product?

Here's what I would do (assume sales page is already created, and FB Pixel is installed).
Also, assume I am aware of and using custom and LAL audiences when the opportunity presents itself.

1) Find the best audience using simple creative showcasing the product.

* Traffic objective - 10 ad sets for 10 single interests. Use 1 same creative for each ad set.
* See which audience gets you the best CTR and cheapest CPC, and which one has more engagement

2) Find the best ad variation

* Traffic objective - run 10 different ads within a winning ad set (step 1)
* See which creative gets the best CTR, engagement
* repeat the same process for copy/headline/call to action

3) Switch to Conversion campaign using the best audience (found in step 1) and best ad variation (found in step 2)

4) Run for View Content objective.

5) Switch to ATC objective once you get 15-25 ATC per week with VC objective

6) Switch to PUR once you get 15-25 PUR per week with ATC objective

7) Scale by gradually increasing the budget and watching the frequency don't go above "2"

When you mentioned product, I'm guessing its an eCommerce business?

I would highly suggest against running PPE/Traffic ads for testing products.

Directly running conversion ads are good, but it's very VERY expensive just for product testing, a lot of people with budget to blow for testing use this as it's a guaranteed tell if a product is worth pursuing.

For someone with low budget for product testing say $100-300 per product, your process is pretty good, but replace the traffic campaign directly with Conversion > Optimized for View Content directly from the start.

Also, I wouldn't stress too much about 15-25 metrics before switching to the next conversion in pixel. That used to be the thing to do in 2016-2017 not anymore, tbh you can run directly purchase campaign straight from 0 purchase conversions in the pixel if you feel like it.

What I'd do for conversion objective is:
At the very start, on a fresh pixel. Conversion > View Content just to test interests and creatives. Pick the best performing combos.
Next directly run them in a Conversion > Purchase campaign.

Your goals should be as followed:
1) Find the Creative with the best CTR and lowest CPC. A single post ID.
2) Next focus on finding best performing VV 75%/95% LLAs, interests, VC LLAs rather than climbing the objective ladder like you mentioned.

Btw, I've been noticing its pretty hard to consistently churn out ROAS like a few years ago with ecom ads, you'd want an AOV of at least $45+ to run fb ads profitably these days.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[The-J]

To get it going, it usually takes average testing of around 100+ ad creatives and 20-30+ Interests. Compared to running ads for other businesses that's a lot, but the scalability of ecom is also huge so it pays off for the client, not for me.

The ugly truth.

I've managed the spend of over $7 million on Facebook/IG combined for about two dozen companies over the past 4 years.

People who strike it big with their first creative can be said to have gotten lucky. But even they aren't immune to creative testing, because that killer creative burns out within a month or so. (Yes, you can bring it back, but you gotta wait a while to do so and generally on a different non-overlapping audience).

It's not uncommon to have one killer creative that burns out and then the rest of the creatives that are tested don't do nearly as well. This is why a testing protocol is absolutely paramount, and why some portion of ad budget needs to go to testing at all times. There's no getting away from it.

The best possible scenario an advertiser can have is to have a set of awesome creatives (10-15) that all work extremely well... on the broadest possible traffic. That puts your ad in the top .01% of ads on the entire platform, and that's how real scale is found. Broad appeals with a variety of creatives constantly being shuffled. Back when I was working for a skincare company, we had an ad that spent literally $4 million and reached almost 20% of the entire female population of the USA over the course of 2 1/2 years.

But guess what... EVEN THAT BURNS OUT! And it's not even possible for the vast majority of companies, which cater to a specific niche. And those burn out even quicker.

 

[Pritesh]

Huh, that's interesting. Slightly off topic, but:

Do you have any insights into what the average cost per conversion is in that space? $100? $500? 1000?

Close friend of the family just got a job as a mortgage broker and I always wondered how well online ads work in such a competitive local space.

I don't have a big data set of running ads for realtors (just four so far). Cost per lead for buyers is pretty low, you can say for $6-15 you'll get a "qualified lead". For home sellers, it goes up to around $30-70. Now how many of those leads converted I have no idea, both were during my agency days and I did not interact with clients directly, that job was for a different team.

We did manage to bring down seller lead cost to around $15-20, but I am not sure if you want to call them qualified leads though. We ran what you can call a "resource bait" ad, offering listicle guide on how to increase their property value and collected leads that way. You can say they have an affinity to sell house, but to call them qualified leads is pushing it.

Do you have any successful SaaS b2b funnels you can share?

Sorry, copywriting sales funnels is something I am good at and I wouldn't want to lose my edge. Moreover, I hate to admit it, but it's pretty one dimensional. I found a call out technique that delivers everytime and I replicate that in multiple niches.

 

[Pritesh]

From your service thread you mention you don’t work with e-commerce.

Is it generally especially difficult to run e-commerce ads profitably when compared to other business types? Is there another reason why e-commence clients aren’t your bread and butter?

Compared to the ideal businesses I work with as mentioned in the thread, running ads for eCommerce businesses take a LOT of time, effort and testing to be worth it for me.

And let's be honest, unless it's a patented unique product, why would I bother running ads for someone else, when I can dropship myself? I do have my own ecom stores generating me easy passive income.

The initial testing is huge time and money investment for ecom and its very risky for me to charge a retainer every month when the client can easily cut me off as soon as I get a winning ad and keep the ads running themselves.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[Pritesh]

What about ecommerce takes so much more time and effort do you think? Assuming it’s a successful products and brand already. What makes it so much more complex than services?

To get it going, it usually takes average testing of around 100+ ad creatives and 20-30+ Interests. Compared to running ads for other businesses that's a lot, but the scalability of ecom is also huge so it pays off for the client, not for me.

 

[Pritesh]

The ugly truth.

I've managed the spend of over $7 million on Facebook/IG combined for about two dozen companies over the past 4 years.

People who strike it big with their first creative can be said to have gotten lucky. But even they aren't immune to creative testing, because that killer creative burns out within a month or so. (Yes, you can bring it back, but you gotta wait a while to do so and generally on a different non-overlapping audience).

It's not uncommon to have one killer creative that burns out and then the rest of the creatives that are tested don't do nearly as well. This is why a testing protocol is absolutely paramount, and why some portion of ad budget needs to go to testing at all times. There's no getting away from it.

The best possible scenario an advertiser can have is to have a set of awesome creatives (10-15) that all work extremely well... on the broadest possible traffic. That puts your ad in the top .01% of ads on the entire platform, and that's how real scale is found. Broad appeals with a variety of creatives constantly being shuffled. Back when I was working for a skincare company, we had an ad that spent literally $4 million and reached almost 20% of the entire female population of the USA over the course of 2 1/2 years.

But guess what... EVEN THAT BURNS OUT! And it's not even possible for the vast majority of companies, which cater to a specific niche. And those burn out even quicker.

Couldn't agree more.
I personally test a bunch of creatives and angles and then shortlist 5-10 and run them 1-2 at a time in a rotation. When I find an angle that works, I start making subtle variations to keep creatives fresh.

Also, a wide audience targeting is the way to go, more so now. I usually test with 5-10 Million audiences and then after the pixel is sufficiently seasoned, you can take off all the targeting options and fb will find you buyers. Non targeted adsets usually outperform Interests or even LLA audiences over time.

What you said about striking big with first creative is also true. If you laser-target an extremely niche audience, who are passionate about your niche, obviously those people will resonate with your product and probably buy it. But, you won't be able to scale it unless you have a wider hook to target people beyond those passionate few.

Guys, when you look to hire a "media buyer" or someone to run ads for your dropship store, please think about this:
Everyone in the entrepreneurial space knows how easy it is to setup a dropship model ecom store, if someone is telling you they are good at running ads for an ecom store, why wouldn't they do it themselves? These guys simply have watched a few courses and want to test the methods with your store and your money. Do not fall for it.
If someone is good at churning out 100 different ad angles and audiences to test your product, and willing to run ads for your store, they are lying!

I finally understand why people with extreme low budgets don't get much out of FB ads. It's a game that whales with boundless $$$ play.

Continuing from my above points, either you have the time or the money.

I didn't have big pockets when I started advertising my ecom store, but whatever I earned I poured all into testing ads, testing angles and audiences. It took me at least 4 months and around -$1400 loss (after calculating ad spend, Cost of goods and fees, store revenue etc) after I saw my first profitable month.
I scaled it very slow, but I did my testing relentlessly. What profits I made, I again invested it into running more tests. That store survived more than 14 months now.

 

[Pritesh]

I'm curious : during that 4+ months of testing ads without much success, did it occur to you it might be a product (need) issue and not so much about the ads? Did you felt like giving up at one point and find a better product?

Nope, I selected a niche and tested a few products in that niche. I've chased shiny objects a lot in the past, from my research I have seen products in that niche working for other brands in the past and even today, so it's just the messaging and audience I needed to nail down.

In my opinion, any product can be made to work if you find the right hook for it. There obviously may be some highly target problem solver product that a pocket of an audience is hungry to buy, that would work right off the bat. But such products can't be scaled unless you figure out a mass appeal hook which works on a wider audience.

 

[Pritesh]

How would you go about setting up ads for a single product e-commerce business?

I'm using a WordPress landing page with a Shopify checkout link, and have already wasted $400 on Google Ads. Lots of clicks but no sales.

My product is unique in the USA but I'm worried people will just buy a cheaper solution used or just aren't interested in the product. It's a disposable version of a durable good. Meant to be used temporarily or just as a cheaper alternative because of lower manufacturing costs, shipping, etc.

I wouldn't waste money on Google unless you know exactly what you're doing. Most app/site placements in Google are botted clicks and is a sure shot way of losing money.

Like I said int the first post, this AMA is about facebook and not Google.
If you really want to stick with Google, place your ads on Youtube, those get you genuine visitors.

What do you think of FB lead ads? Particularly for the real estate niche.

I'm an investor not a realtor and I've seen other investors say this is the ad type that performs best for them.

Makes sense to me as FB wants to keep the user on their platform (more $$$ for them) and the lead from is pre-filled with the user's FB account info so it lowers the hurdle of them entering info manually. But I'd love your take on it.

Also I'd appreciate any feedback on the following FB ad campaign I'm about to run.

I'm about to start a campaign to find motivated sellers. Most investors target the same audience with the same "we buy houses cash" type offer (wholesaling and flipping). I'm doing a different model where we buy with terms not cash.

The wholesalers basically submit low-ball cash offers that are only right for a small sliver of the market (desperate, has lots of equity, needs cash fast, house might need lots of repairs), but I can hit a much bigger market (any equity amount, nice houses in nice neighborhoods) and pay more as long as the seller can give us some time to pay them off. So our USP would be "We Pay Full Price For Your House".

I just want to drive inbound leads at an acceptable price and then I know how to pre-screen them over the phone.

Location-wise I'll be targeting a few counties in my state. Can't do pindrop as it would cross into other states. Not sure if it matters.

Demographic wise is where I'm stuck. I've heard a lot of conflicting advice on this. FB has all this data but I don't want to overtarget and I'd rather let FB find the right people for me, and then I can make lookalikes off of that. I don't have a list to upload or any warm pixel audiences to start with.

Here is what one real estate FB ads guy does. He makes 4 groups

• No interests
• Loan related interests (mortgage companies, refinancing, etc I guess)
• House related interests (zillow, trulia, etc)
• All interests (loan + house ones combined)

View attachment 32091

He runs each one at like $5-10/day ($20-40 total). After a few days he sees if there's a clear winner, and then he'll trim the losers and duplicate the whole thing to scale. Do you think that's a decent way to go?

I plan to test ~5 different headlines, images, copy using FB's dynamic ad stuff. I can do it manually if you think it's worth it for better data.

I should know pretty quickly if I'm getting leads from the lead form or if they call. If the campaign bombs I can pivot without losing my shirt.

Is this plan good or is it a$$? Poke holes and tell me where I can improve.

P.S. Thanks for this thread, I've been learning what I can about FB Ads in order to run this campaign. That 'house hunting' interest tip is something I hadn't heard of.

I'm assuming most of them have an agency run their ads?
Lead ads are the best performing objective to get clients for any physical/tangible business and is the one I use. But, you need a system in place to deliver the filled leads to your email/phone instantly. If you're not so tech-savvy, you need to work with someone to set this up.

Warm leads convert miles better than cold ones. If you want to do it yourself, check the integrations/CRMs FB lead ads support to capture leads and implement it.

About your targeting question, I' wouldn't touch interests at all, to be honest, it's a real estate niche and just one conversion for you is worth a lot more than let's say someone in other B2C niches. Facebook's algorithm is a powerful thing, no doubt narrowing by interests will get you leads in the short run, but it will die down in a week max particularly if you're narrowing it further by county as well.

My suggestion, just 2 adsets: Male- age range- locations/zipcodes- leave interests blank.
Female - age range- locations/zipcodes- leave interests blank.

Obviously you will lose a lot of money at the start, but once you get around 10 or so leads, facebook's algorithm will know your exact target audience and these ad sets will not die until 5-6 months+

Getting leads from facebook for real estate is the easiest thing to do, if you're learning to do it yourself, don't give up. The rewards are insane and don't chicken out in spending money upfront to make the algorithm learn your audience. Every wannabe real estate advertiser targets Zillow, etc the interests you've shown, agencies do so as well to get them quicker leads. If you're in it for the long haul, follow my targeting and over the long run you'll be swimming in leads.


@AniM Edited the end section.

 

[sparechange]

Do you put your left shoe on first, or right shoe?

 

[Benedict]

If you lost all your knowledge and had to start from scratch, as a total noob on (Facebook) Ads:

How would you go step by step to learn and get good at it?

Or different way:
How would you teach a noob like me to get good at Facebook Ads?

PS: thanks for the AmA

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[Vadim26]

Hi @Pritesh, I have 2 questions.

I've just started using FB ads 2 months ago, and have had some success.

I will ask a question and share a strategy I use. Curious to hear your feedback.
I understand that there is no 1 certain answer for this system, but maybe you can find something I am doing wrong. Thanks.


I) What is the best way to use FB ads to prove a product is worth selling?

1) Run targetted traffic to a landing/sales page
2) High CTRs, engagement, and ATC events are good signs that the product will make it

II) What is the best strategy to first test, and then build and scale conversion campaign for selling a physical product?

Here's what I would do (assume sales page is already created, and FB Pixel is installed).
Also, assume I am aware of and using custom and LAL audiences when the opportunity presents itself.

1) Find the best audience using simple creative showcasing the product.

* Traffic objective - 10 ad sets for 10 single interests. Use 1 same creative for each ad set (**).
* See which audience gets you the best CTR and cheapest CPC, and which one has more engagement

2) Find the best ad variation

* Traffic objective - run 10 different ads within a winning ad set (step 1)
* See which creative gets the best CTR, engagement
* repeat the same process for copy/headline/call to action

3) Switch to Conversion campaign using the best audience (found in step 1) and best ad variation (found in step 2)

4) Run for View Content objective.

5) Switch to ATC objective once you get 15-25 ATC per week with VC objective

6) Switch to PUR once you get 15-25 PUR per week with ATC objective

7) Scale by gradually increasing the budget and watching the frequency doesn't go above "2"


** The reason I use traffic objective to test is that link clicks are cheap and I accurately judge my add performance.

 

[Pritesh]

Thank you! Exactly what I was looking for. Will give that a try.
Yes, an Ecom Business.
One more question, though.
What are your thoughts on running PPE for winning ad set just to get social proof? (before switching to conversion and start selling).

I used to do PPE exclusively for creative testing in the past. Put $20 for a day, dump in 10 creatives and it used to find a winning creative within a day. But, from my experience lately, it hasn't been giving out clear winners.

Also, I'd run PPE for social proof after you get your winning post id using conversion objective. If you haven't noticed normally PPE ads don't have a call to action, but when you feed a post ID from a conversion ad to the PPE adset, it will show you a call to action.
Don't go overboard with PPE daily budget though, they are literally the bottom of the barrel traffic and if more of them start visiting your website, your pixel data will be tainted beyond redemption.

Spoiler: Greyhat advice below, use at your own risk:

There are *other ways* to get social proof like likes and shares if you catch my drift and you won't need that many tbh. Just make a few friends comment on the post praising the product and like them with your page so it shows at the top of the comments.
Filter every single word in English dictionary on your facebook page so no one could comment on your post unless you manually unhide it, etc etc. Tons of stuff you can do.

 

[Pritesh]

Can you run a facebook ad in spanish text in Phoenix, AZ USA ?

Sure, Facebook has a language filter as well as location filter. You can even add in interest for more laser-focused targeting options.

Great post, thank you @Pritesh

I know that you have to test it and depends by sector / copy / etc., but in your experiences for an e-commerce work better the carousel ads or the single image ads?

Video ads > Carousel > Single Image.

 

[Pritesh]

For video ads, are you referring to those expensive production ones, with models and entire video crew?
Or can a slideshow saved in video format be considered one as well?

No, for ecommerce product demo videos with text above describing features work best.

 

[Pritesh]

If I'm testing 5 ad creatives (ad designs), and I'm a new tiny eCom biz, how much should I spend on testing each one? Some folks say $5 / day for 7 days is enough to see how each one performs. What's your thoughts?

Are they in a single ad set? 1k Impressions per ad creative should be enough to show you the winner in terms of CTR and CPC, but make sure that each creative reaches the 1k mark.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[Matt_2190]

Thank you @Pritesh

I changed my ads from traffic to conversions, as you suggested in the other post.

I'm spending slightly more, but sales are also more numerous. Now I'm only using that goal.

Now I am taking new photos of the product for do the tests, but sometimes I have some difficulties. If you had to choose photos of products, which do you think perform best? Those of the product on a colored background (more "professional"), or that of a product that is used (perhaps made by the smartphone, which seems "amateur")?

 

[Pritesh]

Dont they ban you from this type of targeting, its like targeting hispanics ?

Why would they ban you for using their own targeting options?

Just don't call them out specifically in ad copy like for example "Hey Hispanics in California" that will surely trigger the FB bot and ban your account for indiscrimination.

Question : I remember reading people talking about a technique whereby you run a FB ad with Traffic Objective, then run an Engagement Objective to the same ad for social proof at the same time, to further boost the ad's reach. Is there such a thing?

Like I mentioned earlier, those techniques worked great in 2017, but nowadays you don't want to send junk traffic to your website to taint your pixel. There are other ways to get social proof and frankly you don't need much of it

Anyways, testing it can't hurt right? So here's the process I used to do to get social proof and in turn lower cpc/CTR metrics.

Directly run PPE campaign on that post ID. You'd get a bunch of dirt-cheap social proof and traffic to your website. Simultaneously run a conversion-based campaign to take advantage of lower metrics. Be sure to monitor comments though, as you'll get a bunch of low-quality audience visiting and commenting on your post.

 

[Pritesh]

In my Ads table, I see the above. Is 'Below average / Bottom 10% of ads' something to be concerned about?

I wouldn't be too concerned as long as its getting me conversions.

 

[The-J]

Guys, when you look to hire a "media buyer" or someone to run ads for your dropship store, please think about this:
Everyone in the entrepreneurial space knows how easy it is to setup a dropship model ecom store, if someone is telling you they are good at running ads for an ecom store, why wouldn't they do it themselves? These guys simply have watched a few courses and want to test the methods with your store and your money. Do not fall for it.
If someone is good at churning out 100 different ad angles and audiences to test your product, and willing to run ads for your store, they are lying!

I disagree. It's not actually easy to make a dropshipping store work long term. It's really hard and highly unlikely. Takes half a day to set up, and maybe you're profitable out the gate, but if you pick a good product, chances are someone else is going to figure it out too. If you have a single dropshipping store that's been running profitably and consistently for over a year, it means you've picked a good product and have excellent ads. Not everyone who knows how to split test knows how to pick products, nor do they know how to actually make good creative.

Many of the guys who SUCCEEDED with dropshipping and now doing Facebook ads... guess what? They're doing it because their original store failed. And why did their store fail? Because they couldn't compete anymore. The business model wasn't good enough, their offer wasn't good enough, and they eventually ran out of money to test. You can only bleed for so long.

It's not that they're lying. It's that they know they can't compete and they'd rather use their tactics for an entrepreneur who has already figured out the offer side. There's more money in that compared to a dropshipping store. That's the reality. It's 2020: dropshipping is too easy and the entry barrier is gone.

Of course, there are people out there who want to get their feet wet after buying a course. You definitely don't want to hire those people.

Never hire a media buyer or agency to start up your business for you. Instead, hire a media buyer or agency when you've got an offer working and an idea of what actually works for the offer. The hire will take a lot of time off your hands and allow you to start doing other things, like expanding into new channels, investing in product development, and courting investors.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[Pritesh]

@Pritesh - curious why you would scale the adset by duplicating it vs increasing the budget? I've seen this recommended a few times but seems to be contrary to what facebook recommends and based on facebook docs this strategy implies a negative impact to delivery due to audience overlap across the ad sets. Could you elaborate on the reasoning behind this? Thanks

Unless you have a very tiny audience, audience overlap doesn't matter. You will not be outbidding your own ads as Facebook doesn't allow it compared to other networks like Google.

The term you're looking for is "Pattern Interrupt". I have used it in real life display signage, and while it's hard to quantify its impact in my gut I know it works. All I want a potential customer to do is to pause and take notice, then I can swoop in to begin the close.

Great thread btw!

In advertising, usually in Facebook, we call it a scroll stopper, but I didn't want to use technical terms to explain it.
In video ads, you need a scroll stopper in the first 3 seconds. In image ads, it's usually done by red circling something irrelevant in the picture so people stop and take notice. Fiddling with high contrast/saturation, etc there are many ways to stop a person scrolling through their newsfeed to take notice.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[Pritesh]

Right, so I launched the campaign after I posted. So far I've gotten 4 leads at ~$9/lead.
1 has turned into a conversation and 2 I haven't been able to get a hold of yet. The last one had fake info as far as I can tell.

Lead quality should be much higher than outbound but yes they are still cold.

At least the campaign is running and I'm getting a good CPL. We'll see if the numbers hold but follow-up is the new problem I need to work on.

I am working on setting up a CRM and some automation so I get an email notification and maybe send the lead an immediate text/email. I am tech savvy so I'm not worried about doing any of this. Using zapier and already got my email notification set up but still need to pipe the info into a CRM.

As far as targeting goes. I ran with the 4 adsets I mentioned previously. The "mortgage-related" adset got 3 leads and the "no interests" got 1. I'm not sure how many impressions I need to make a statistically sound decision but I'm thinking of pausing the other 2 and letting the ones that are working ride. Do you think this is a good move?

Thanks for the advice on targeting but FB Housing ads don't allow you to target by gender or age anymore. I think they changed it a year or so ago due to some discrimination issues. You also can't target zip codes for the same reason. Just counties (and they don't have all of them lol) or cities (with 15+ mile radius which sucks if you're near a border).

I am targeting multiple counties + cities within driving range. I want to make sure the audience size is big enough.

I've heard other investors using FB say they get better results the wider they go, even targeting the whole state or multiple states, doing business virtually. One day I'll get there lol.

Right now I just need to figure out how to optimize and scale what I have right now, and follow up quickly.

$9/lead is very good considering the ad just started. Over time the algorithm will optimize automatically since your campaign's objective is leads.

Do not pause your adsets, just put testing adsets on a $5-10/day budget and let it run for at least 3 days. Do not fiddle with adsets, I know it's hard seeing the money go, but this is where 90% of new advertisers fail. Do not edit anything whatsoever in adset, if you want to run a new test, duplicate it and run that.

About your impressions, usually 1k impressions should've you a good result if you're testing absurd interest targeting. But this should only be done when you've run out of laser targeted interests. But again, judge your ads by only conversion metrics, and let them run for at least 3 days even at low budget.

You can target by zipcode, and you can even make a callout ad saying Hey Arlington home owners. I never had a problem with those. Just don't discriminate in the ad saying hey Hispanic home owners, etc.

Facebook will do the optimizing for you if you spend enough money to get the data. Most people don't. If it's a small city, you don't even need >$1-2k to reach every single person with your ad. That's why I usually suggest not using interest targeting if it's a wide appeal business.

 

[Pritesh]

So I got to 1000 impressions on all 4 of my adsets and I haven't had any more leads come in.

I'm planning to pause the 2 ads that got 0 leads and then maybe duplicate the campaign with the 2 that did work. Hopefully they pick up a different segment of the audience pool. Do you think that's sensible @Pritesh ?

I also want to test some new creative.

What's the best way to split-test in Facebook? 2 ads in one adset or 2 different adsets?

You want multiple ads in a single adset to test, otherwise how will you judge the performance if two different ads are targeting two different audiences.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[Kid]

I'll go with one.

If i wanted to get cheapest US traffic with low budget, how would i go about it?
By cheapest i mean something $0.05 cpc and not really targeted.
The fact that its US based would be enough.

Thanks for AMA.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.