We dare you to sit down and witness Depesh lay out his nine takeaways for ecommerce marketing campaigns and not learn something that you can apply immediately.

He’s put together the best tips he, himself, gathered from auditing and running 100s of Facebook ad accounts and generating over $100M since 2005. It’s a full-suite approach covering unique angles to audiences, creatives, ad set optimisation, funnels, automation, and of course, bulking AOV and LTV for long-term profitability.

This was one of the most jam-packed presentations of AWA, be sure you take notes and take full advantage of his Q&A post-speech.

Speech by Depesh Mandalia | CEO, SMC and ZASR Digital

Depesh Mandalia Speech Transcript

What’s going on Bangkok? How’s everyone doing? Yeah. So I can see a few confused faces, right? Like, what the hell is going on?

This is not a chef show. We are going to talk about Facebook ads.

But there is relevance behind why I’m dressed like this today. And it actually goes back to this picture here.

This dude was about 14, 15 years old. And he wanted to be a chef.

Now, in my house, my dad’s view was that men don’t cook. Like, women do all the cooking.

So he said to me, “You know, you’re not really gonna get a job out of cooking. Men don’t do cooking. It’s not really for you.”

But I loved it. I loved playing around with recipes and ingredients and flavours and things like that.

So I was a bit disappointed. I figured, you know what, I’m not gonna get a job as a chef.

So then I went into something completely different, so computers and marketing and things like that.

Well actually, now that I’m doing marketing, this is actually what I do.


Depesh Mandalia - SMC And ZASR

Depesh Mandalia – SMC And ZASR

I actually cook up recipes and ingredients with Facebook ads. So, I don’t know if Manu actually made it to this show, but a big shout out to Manu, I don’t know if he’s here, for this image here.

This is what I do. When it comes to marketing, I played around with so many different channels. So Facebook is the channel that I’m playing around with right now.

But it comes down to recipes and ingredients. And that’s gonna be a key theme of what I’m gonna be talking about today.

Now at the moment, I run a Facebook ads agency, so SMC up there. I also run a Facebook ad training program with ZASR.

Started In 2005

But actually, my story in ecom started in 2005. So, I’ve been in this game for quite some time now.

I was following the corporate ladder, earning my stripes as a website manager, marketing manager, etc. etc. And I thought that was my career, you know, straight into corporate.

But in 2009, I hit a really low point. I lost my job. My second child was due. And for 2 months, I couldn’t find a job.

Now for me, that was a worse time for me. I fell into depression. And I really found it hard to get out. But actually, my saviour was affiliate marketing.

Because at that point, I figured that if I can’t get a job, I need to create my own income.

Made My First Sale

So in 2009, I would spend hours and hours on forums and reading blogs and hacking WordPress and trying to get all these affiliate links going. And about 6 months later, I made my first sale.

For those of you that do affiliate marketing or even ecom, when that first sale comes in, that’s like the most amazing, amazing feeling.

So I got there and I got the bug. Actually, two years later, I hit a 7-figure revenue through affiliate marketing. And I wouldn’t have done that had I not lost my job.

So for all of you, whatever you’re going through, if you’ve been through something similar, you will come out of it.

There’s always a reason for these things happening.

Well actually, if I fast forward a year on from that after I made my 7-figure revenue, I lost it again because I made some really, really silly mistakes.

My one tip for you guys that are doing affiliate marketing and lead gen, build a list.

That’s one thing I didn’t do. I was driving SEO traffic, I was driving like millions of visitors, making loads of cash out of it.

But then all of a sudden Google made a change in 2012 and I lost all my traffic.

Facebook Ads

So in 2012, I started playing around with Facebook. Facebook ads at the time, it wasn’t as informative as it is right now.

Actually, I don’t know if anyone’s run Facebook ads back in 2011-2012. You could only run it through a plug-in on Chrome.

Hands up. Who’s actually seen that interface? I don’t know how many OGs we’ve got here. Yeah, that was difficult.

So whatever you think is wrong with the ads manager right now, you can 10x that, six years ago.

Actually, through test and learn, 18 months of failure, the only thing I thought that you could do with Facebook ads were literally just boosting your post.

And I just couldn’t figure out. Like, why would someone buy out of Facebook? It just didn’t make sense.

I actually came from the paid search world, where you bid on a keyword, you put your text on there, and you get the click and drive them into your website.

So I was treating Facebook exactly the same.

$1K Ad Spend, Zero Revenue

I spent 18 months trying to figure this thing out, Spending money, making mistakes until I found the right recipe.

So I had the ingredients. I kind of figured out that I knew how to do this stuff.

But then when I figured out the right recipe for Facebook ads back in 2014 and 2015, I managed to spend $6 million to generate $26.5 million in 18 months.

We took an ecom store, print-on-demand, from $800,000 to $26.5 million purely through Facebook ads. And at the time, it was probably the fastest-growing ecom store in the world at the time.

That got me a lot of interest from Facebook.

So, Facebook approached me and started to get some insight into the kind of, how do you run Facebook ads, what are you doing that we don’t know.

Actually, at the time, my team had actually built some software on top of the Facebook Ads API, which would automate and analyse everything.

So a lot of these kinds of tools that you see in the marketplace right now, we were doing that back in 2015. We were well ahead of the curve.

Getting Involved With Facebook

That got me involved with Facebook. So, I do advisory services, I get access to alphas. I feedback on their products, I give candid feedback on their support from the marketing managers.

So, I don’t know how many of you guys have got reps. There’s like whole different levels of okay reps and really good reps.

Unfortunately, the good reps, they’re in high demand. So you don’t really get too many of those.

And so my kind of work with Facebook is really feeding back what’s actually happening on the ground. So I’m working with Facebook every day on the ads platform.

My team is running multiple campaigns. And I’m giving Facebook this direct feedback and I’m getting some good stuff back from them.

I don’t know if you saw that the guys that are organising this event put this post out.

Is Facebook Still A Viable Platform

Depesh Mandalia - Post

Depesh Mandalia – Post

And they said, “Look, what’s some kind of big numbers that you’re doing right now?”

At that time, that was probably September time, we’d actually spent $100,000 and generated $1.3 million in return from an e-commerce store.

The average order value was about $200. So these kind of numbers is absolutely insane.

A lot of people are looking at Facebook and saying, “You know, where it is going right now? Is it still a viable platform?”

Absolutely it is.

But it’s only if you understand the right ingredients and you pour together the right recipes that you can start to get numbers like this.

And a lot of ecom people that are afraid to experiment with higher ticket prices. A lot of drop shippers, print-on-demand people will be playing around with $20-30 products and really scared to kind of push it up.

But there are different ways to actually get more value out of Facebook.

So I’m hoping that you can take away at least one of the nine things I’m going to talk through and go and implement it today.

Actually over the last 30 days, so consider this as November, this is Q4, this is the peak. Everyone’s stressing about, you know, Black Friday, etc.

The same account, we took that spent to $260,000 in the month and generated $4.3 million from an ecom store.

I don’t know how many other people can get these kinds of numbers in peak. Even I was surprised that we were able to do this.

But it’s because we follow a formula. It’s because we follow some of the things I’m going to take you through today. And when you start to repeat that success, it starts to build momentum.

Ingredients. Recipe. Experience.

So what I talk about is ingredients, recipe, and experience.

You can take the best ingredients for an ecom store. As long as you know the recipe, then you can start to bring it together. You can start to make sense of it.

So let me give you an example. Who here likes pizza? Awesome.

Pizza is my favourite dish. Absolutely.

So you look at this image here.

Depesh Mandalia – Pizza

Depesh Mandalia – Pizza

Really nice, fluffy base, gorgeous red tomato toppings, and fresh ingredients on top. It’s a really appetising pizza.

So, what are the ingredients? The ingredients are flour and water and vegetables and all that kind of stuff to bring your pizza together.

The recipe is going to be pretty much standard. Cook the base, add the toppings and shove it in the oven.

But even something like that can go wrong. And, you can end up with something like this.

Depesh Mandalia – Pizza

Depesh Mandalia – Pizza

And the point I want you to take away is, you know, you’re going through these sessions all of today, tomorrow, possibly Friday, and you’re learning all these techniques.

Some of them you will take away and you’ll create a perfect pizza. But some of them you’ll start off like this. You will make mistakes.

I’ve done it, I’ve done this myself. In fact, the first pizza I made when I was about 15 wasn’t far off this.

But I learned and I kept practicing. And I took that forward into those learnings and I made sense of it as well.

Higher Return On Ad Spend

Depesh Mandalia – Revenue

Depesh Mandalia – Revenue

That’s how we were able to make these kinds of numbers that we do with Facebook ads. Whether we’re spending low £1,000 and generating £30,000.

Depesh Mandalia – Revenue

Depesh Mandalia – Revenue

Spending £13,000, generating £138,000.

These higher return on ad spend performances are possible when you understand how Facebook works inside.

And I’m going to take you through 9 of the key points I’ve pulled out from the last 3 months that I really want you to kind of take away.

Depesh Mandalia – Revenue

Depesh Mandalia – Revenue

Another example, £90k and £460k revenue.

These are all ecom stores.

And then the example that I gave earlier, $100,000 to $1.3 million.

Depesh Mandalia – Revenue

Depesh Mandalia – Revenue

Now, before I go into that and there’s been a lot of scaremongering about Facebook.

So I read a post a couple of months ago talking about Facebook Q4, it’s gonna get too expensive. There are too many competitors. Facebook is changing the ads platform.

You’re not going to be able to target lots of different things that were previously in ads manager.

Depesh Mandalia – Facebook

Depesh Mandalia – Facebook

And Facebook is plateauing in a lot of countries like the US, for example.

So, where is the opportunity? How can people like me and many other people that are making Facebook a success, continue to make money and profit out of it by understanding some of the principles of Facebook.

Where Is The Opportunity

If you keep the user happy, Facebook is happy.

There’s a reason why Facebook go after affiliates and cryptos and nutras and ecom stores, they’re not against you personally.

They don’t want you not to advertise on the platform, they want your money.

But if the user is unhappy, then Facebook is unhappy. And you won’t get the performance you need. Your account will get banned, your ads get disapproved.

You don’t get the same reach, you get punished with higher CPMs.

Most of the accounts that I was running through Q4. We had super low CPMs, like $10-15. All the way through Q4.

So when lots of people are stressing out about $50, $80 CPM, you’re doing something wrong. You’re doing something wrong that’s not pleasing the gods of Facebook. Like that’s the key thing.

Zuckerberg doesn’t sit there and say, “Well I’m gonna ban that account, I’m gonna ban this.”

It’s all automated. There are rules. And you either follow them or you don’t, and that’s absolutely your choice.

I’d do it. Not just because I work with Facebook because I want to make money. And so you have to kind of make that decision.

I completely understand blackhat and I know that that’s way over here, and there’s a whole different game that goes on there.

But if you don’t want to lose your account, if you want to keep your ads approved, then you need to be thinking about doing things in a way that’s gonna work for Facebook.

Before I continue, I’m gonna take off my dressing. There we go.

Let’s get started. Who’s looking forward to hearing the nine e-commerce takeaways? Yes? Awesome.

#1 Navigate Or Die

So number one. I’m gonna talk about navigation.

A lot of you will be experienced in Facebook ads and I’m sure that when you’re in ads manager, you know exactly what’s going on. But a lot of you will go into your ads account and you will see this.

Depesh Mandala – Navigate or Die

Depesh Mandala – Navigate or Die

It’ll be a complete mess of data and controls and knobs and all these kind of things. And you don’t know where you’re kind of focusing on.

And so you kind of launch all of your columns and you’re trying to analyse, you’re trying to figure things out.

Depesh Mandalia – Navigate or Die

Depesh Mandalia – Columns

Well actually, if you’re running an e-commerce store, what’s the one metric that matters?

Return On Ad Spend

Return on ad spend. Everything pivots off return on ad spend.

And then I break it down to the level below. So what drives return on ad spend? It’s the cost per acquisition and average order value.

Depesh Mandalia – ROAS

Depesh Mandalia – ROAS

They’re the two metrics that I focus on.

And then I might break it down further and I say, right, how do I influence cost per acquisition?

Depesh Mandalia – ROAS

Depesh Mandalia – ROAS

It’s how much in my paying per click and what conversion rate am I getting? How do I influence average order value when I’m looking at what the average selling price is on the store?

And how many of those items I’m selling. Now all of a sudden I’ve just simplified e-commerce metrics for you.

So the thing here is, and I work with hundreds and hundreds of different people, it really takes them a lot of time to really get the guts of how to make their e-commerce store work.

Once you understand the metrics and you know where the problem is, then you can focus on the right area.

So, if your problem is CPM and costs, then you need to work Facebook a lot harder.

If your AOV is too low and you’re really struggling to get the CPA to make your return on ad spend profitable, then you need to work on your store.

Maybe start bundling up your offers. Start charging more for your goods, start selling more in your basket.

Once you start to break it down, it becomes a lot more simplified.

Conversion Rate

Now there’s one metric, which actually, Facebook don’t even give you in ads manager and that’s the conversion rate.

So, if you’re just relying on ads manager for your reporting, you’re missing out on one big key metric, which is how many of my clicks are converting into customers.

And so what I do is I pull that out into a Google sheet. Then I analyse it right there.

Depesh Mandalia – Example

Depesh Mandalia – Example

So this example here. I will look at my results at the top level by day. And then I’ll colour-code the results that come in.

Now as I explained, return on ad spend is a factor of AOV and CPA. And so I have those columns on one side.


Then, CPA is a factor of conversion rate and cost per click.

So any given day, I know exactly where the problem is and I know what to do about it.

If my cost per click has gone up, then maybe I need to start refreshing my ads. Because they’re not as performant as they were before.

Or if my conversion rates dropped, maybe there’s a problem with my page load time or something like that.

This is the simplified dashboard that I run for all of my accounts. And at the top level, I know exactly what’s going on. The cool thing about this is if you’re running ten ad accounts or a hundred, you can just have them all in one view.

And it’s super simple to scan it, especially with the colour-coding, to figure out what’s going on.

If you want this spreadsheet, I’ve got a link at the end of this presentation. You can grab this spreadsheet and start using it today.

The second spreadsheet that I look at is calculating what my CPA and my AOV needs to be to hit my return on ad spend target.

So this is another spreadsheet I’ll give you at the end of the presentation.

Depesh Mandalia – Example

Depesh Mandalia – Example

And what you can do with this is to model how hard you need to work on each of those metrics.

It’s just a really simple way of figuring out, right, if I increase my AOV, what does that mean for my ROAS versus actually if I increase my CPA.

And I can tell you, it’s infinitely easier to work on your AOV than it is your CPA. This is where people get stuck with Facebook ads and say, “You know, Facebook is difficult, it doesn’t work.”

CPA: Harder Metric

You’re focusing on the harder metric, which is CPA.

If you can get your AOV up, you know, you can continue maintaining your CPA, that’s where you find scale.

So the second one is, a lot of people come to me and say, “How do you do fast scaling?”

You hear lots of strategies. Manual bidding, duplicating ad sets, multiple accounts.

There are so many different ways to hack the system.

But one of the simplest ways of scaling is actually going global.

#2 Scale Global

A lot of you will be focusing on one market at a time. Let’s test in the US and see how it goes or maybe in the UK or your native market and see what kind of performance you can get out.

Depesh Mandalia – Scale Global

Depesh Mandalia – Scale Global

But actually, if you look at the top e-commerce countries by GDP as I’ve got illustrated here, you’ll see there’s a whole range of opportunity here.

And so I’ve highlighted Netherlands and Italy. Time after time, it doesn’t matter what store I’m testing. They’re generally always profitable.

Actually, there’s also an outlier here, which is Poland. Poland has higher e-commerce adoption than Italy and it’s closer to Spain.

Sure, the population might not be as big and you’re not going to scale necessarily to 8-figures, but when you start to stack up these countries and you start to go horizontal, then that’s where you can make money.

Facebook Stats

So a few stats from Facebook. Poland has 16 million people on Facebook of which 14 million are active in a given month.

epesh Mandalia – Poland

epesh Mandalia – Poland

Hungary has 5.7 million.

Depesh Mandalia – Hungary

Depesh Mandalia – Hungary

The Czech Republic has 5.2.

Depesh Mandalia – Czech Republic

Depesh Mandalia – Czech Republic

And there’s a whole range of these countries where you can start to profit.

So to give an example, I was scaling an ecom store over through Q4 and we got our spend up by actually going horizontal. The return on ad spend, we were able to replicate in literally every country were going in for.

Bunch Up Your Countries

And the simplest way to do this is to bunch up your countries, even go with the English copy. And sure, some of the countries need specific payment providers.

But if you’re testing the market, you don’t need to translate your website, your ads, think about currency. Testing those markets with what you’ve got and see what kind of reaction you can get back.

Depesh Mandalia – Purchase ROAS

Depesh Mandalia – Purchase ROAS

Like for example one of the countries that stood out was Ukraine and Slovenia and Romania. They’re not countries I would’ve logically said, “Yes let’s go and spend some money there and see if it’s gonna work.”

But they worked. And I wouldn’t have found this out unless I’d been testing.

Depesh Mandalia – Location

Depesh Mandalia – Location

I generally run a worldwide campaign or at least with my top European markets, maybe top global markets. Run them in a single ad set.

And then when you get your results to come through, segment the results and see what kind of results you’re getting by country.

Is there a breakout, is there a country you can actually take out and market separately? And is there a reason to perhaps go into dedicated language for that country and actually put the effort in to translate?

So a lot of people get put off because you know, translation is quite hard. But go and test see if it’s worth your time first.

#3 The 4-Funnel System (Start Dating)

So the third one I want to take you through is what I call, the 4-funnel system. And this is about dating your prospect.

When a lot of you run your ads, you see your prospect as someone that you want them to buy straightaway.

It’s like going on a date. So let’s say you just met someone, you’re on the first date. And regardless of what your goal is, you might want to marry them, you might want to take them to bed.

Whatever your goal is, if you’re going to go straight after that in the first date, you’re probably not going to be as successful as if you build that relationship up.

Depesh Mandalia – Example

Depesh Mandalia – Example

So if I give you an example. Let’s say, Scarlett Johansson, George Clooney, they’re on a date.

The goal of the first date is to get to the second date. And this is true about Facebook ads as well.

A lot of people put so much effort into that first ad. Buy now. 60% off. I’m sure you’ve all seen the ads or, especially, from dropshippers.

And there are lots of templated ads out there where they’re just trying to get that sale from the first ad. Then they complain because Facebook ads aren’t working.

Well surprise, surprise, your marketing isn’t working. And this is the big problem that I find.

Breakdown Of The 4-Funnel System

Depesh Mandalia – The 4-Funnel System

Depesh Mandalia – The 4-Funnel System

So the 4-funnel system, what it does is it breaks up your prospecting audiences at the top, your lookalikes, your behaviours, and your interests.

And then I built a warm layer below that. So warm for me is someone who’s engaged but not necessarily clicked.

They’ve viewed a video, they’ve engaged with my page, they’ve engaged with the Instagram.

And what I’m doing with those is, because they’re one step down, they know the product, they know the brand at this point.

So I can be a bit more direct response at this level.

And then my hot layer is literally website retargeting.

But the fourth layer, I see most people missing out is existing customers. Facebook is not just a customer acquisition tool. It’s also a re-acquisition tool.

And if you’re not running ads to your existing customers alongside your email campaigns and your Messenger bots, then you are literally leaving money on the table.

Setup: Cold, Warm, Hot, Existing

Depesh Mandalia – Example

Depesh Mandalia – Example

So this is kind of a typical view of how my account might look with the cold, warm, and hot setup.

I use those exact names. So cold, warm, and hot in my campaign setup.

And then what I’m doing is just literally filtering. So I can filter by, all campaigns that have cold in there, warm, and hot. And I can see the return on ad spend and performance dripping down that funnel.

You can literally see it here, wherein my cold audiences, my ROAS is about one. And then it’s about two and a half in my warm audiences. We’ll go down the funnel.

And then finally at the bottom of the funnel, which is 3.6. And this funnel averaged out about 2.5, which was just on target for what we were after.

A lot of you won’t be doing this, a lot of you won’t be building relationships in your marketing.

But I really encourage you to think about things differently. Because yes, CPMs are going up. Yes, there are more competitors on Facebook.

But if you go back to marketing principles and you get the marketing right, then you will be able to cut through.

So the next one I want to talk about, this is one of my favourite parts of Facebook.

#4 3N Ads Formula

So, the 3N ads formula. 3N is about the three nudges.

What do you need your consumer to do when they see your ad in the newsfeed?

#1, they need to stop and take notice of it because you can have the greatest image or the greatest copy, the greatest landing page in the world, but if someone doesn’t click your link or if they don’t even see your ad, then it’s completely waste of time.

So the first thing is getting them to stop scrolling.

The second thing is to engage. So your copy, your visuals, everything has to hook them into the ad itself.

And then the third thing is to take action. The action could be to buy now. It could be to learn more, it could be a lead generation, it could just be to click play on the video.

They’re the only 3 things that matter when it comes to an ad.

So I’ll just give you a few examples of how that looks.

Collection Format

Depesh Mandalia – Example

Depesh Mandalia – Example

So, this ad here through Q4 was the most successful ad that we ran.

This ad template is called a collection format. I think Facebook recently have now kind of rebranded it as an instant experience.

But one of the key things that this does so well is it mixes brand with the product. So, with this kind of ad format, we can have a brand video that runs in your newsfeed. And then we’ve got products below it.

So it’s a mix of dynamic product ads that you have carousel flicking through with branded videos. But beyond that, there are a few psychological principles that we also apply to this ad as well.

The first thing we did is when you see a face in your newsfeed, and that face is looking directly at you, you have no choice but to look them straight in the eye.

Because we are pre-programmed to make eye contact with people. Regardless of trying to understand their impressions, you will make eye contact.

So, when you meet someone, the first thing you look at is their eyes.

And when you look at this ad in your newsfeed, you will look at these people in the eye. It works time and time again to increase engagement.

Lifestyle Image

The second thing is a lifestyle. So this is a lifestyle image. So, you know, this is selling ski-wear.

A lot of you will be showing the standard images of ski-wear. And maybe, having a 360 of the actual apparel itself.

But this is a lifestyle. People don’t buy ski-wear, they buy the skiing experience. And this is so different to how a lot of people advertise on ecom, because people don’t buy the thing that you’re pushing to them.

They want the outcome. Like what’s the benefit of this product?

So that’s what I focus heavily on, at the top of the funnel, is talking about the outcome, not the product.

And the other thing about this is they’re also in situ as well. So this particular brand, they spend tens of thousands bringing influencers to the slopes, wearing their apparel, and looking cool, having fun, etc.

And so we can play with that in the newsfeed because that’s exactly what we’re selling.


Depesh Mandalia – Example

Depesh Mandalia – Example

This second example is a retargeting ad.

Really, really simple but the hook that gets them to stop and take notice is not just saying, “Here’s a five-star review.”

It’s actually putting the five stars in there.

So I’m a big fan of emojis, not overly working them but they’re great at just visually getting people to stop their scroll. And just to take notice.

And when done in a decent way like this, it can work.

Or this example here, where I ran an ad and I’ve used two hooks.

Depesh Mandalia – Example

Depesh Mandalia – Example

Number one, I ran it in Q4, and so I’ve called out Q4. And secondly, I’ve called out a celebrity. And I’m obviously making Gary Vee look cool in this picture.

But this ad works because it disrupts people’s patterns.

And actually, another thing that we do is to take out the color. Use a black and white ad. Don’t overdo it but that can also interrupt the design pattern on your news feed and get people to stop scrolling and to take notice as well.

#5 Graduation Testing

So this part I’m going to talk to you about is testing and how I use fast testing to scale up as well.

Now, apologies if anyone’s hungry, I’m going to go back to the pizzas because I love pizzas.

When you first tried the pizza, when you first like, okay, let me ask you a question. Who here likes pineapple on their pizza? I do.

I was reading a debate online and it was talking about how people got really, really worked up. Like pizza and pineapple don’t go together.

People are really, really feeling passionate about that. I love it. I think it’s a great enhancement to the pizza.

But in order to figure out whether you like that pizza, did you have to eat the whole pizza? No.

All you needed to do was to take one bite and figure out if that pizza was right for you. And that’s exactly how I do testing.

I don’t need to spend $50, $100, $500 testing out ads and ad sets and things like that. I do bite-size testing, I do fast testing.

Because what I’m looking at is not how much have I spent on the ad, but how many impressions have the ad actually had?

Relevance Score

So if anyone knows about delivery insights or relevance score, Facebook only needs 500 impressions to figure out what your relevance score is or to give you delivery insights.

Statistically, that is enough data for Facebook that 500 people have seen this ad.

So why are we testing so much with ads and pushing 5,000, 10,000 impressions on ads? Because I can guarantee you once an ad has started and very rarely does it get massively better.

If an ad starts off bad, it’s probably going to stay bad because Facebook has figured out this is the place in the auction.

And so when I’m doing testing at the ad level, I’m only looking for 1,500 impressions. 500 is what Facebook needs. I’m not a machine, so I need a bit more data.  But I’m looking at 1,500 impressions.

And I’m looking at my early metrics. So what’s my cost per view content? What’s my cost per add to cart?

Because if they’re off the scale, if it’s way too high, if I’ve spent $10, $20, $30 dollars and I’ve got no add to carts, then I kill it.

Because that ad is not working.

It might not be working because of the ads being bad, it might be not working because the ad set is bad. Or actually, might not be working because when Facebook threw it into the auction, I ended up with bad traffic.

So I might duplicate the ad and try again. And this is kind of how I go through testing really fast as well.

Price List For The Optimisation Events

I did an experiment actually, a couple of months ago. What I wanted to do is figure out what is the actual price list for the different optimisation events in Facebook?

So what I did is, on one of my ecom stores, I ran the same ad for the same product across multiple different ad sets. And optimised them for different events.

Depesh Mandalia – Example (CPM)

Depesh Mandalia – Example (CPM)

So for example, purchase, payment info, initiate checkout, add to cart, and view content. And I also ran a couple of page post engagement campaigns.

I wanted to see what kind of CPMs came back. And it was literally graded from, kind of, lowest quality traffic to the best quality traffic, and the prices matched as well.  So the more expensive CPMs were purchase.

But here’s the problem. If everyone’s bidding on a purchase, then everyone CPMs are going to go up.

Why are you not trying out purchase payment info or initiate check out or add to cart?

Actually building a journey and building a conversation with customers and actually getting them on to your site for a lower CPM and then converting them.

So it’s just a different way and a different angle of taking this.

Campaign Setup

Depesh Mandalia – Example (Campaigns)

Depesh Mandalia – Example (Campaigns)

When I set up my campaigns, I have a test campaign where I do all of my testings. And when I find a winner I move it to my scaling campaign.

The reason I do that is that when you have lots of testing going on in a campaign if that campaign continues to collect good and bad data, it can have an ongoing impact on your ad set.

And so, when I find winners, I move those out to a scale campaign.

I might start with $50 or $200 a day. And see how fast I can scale that up. If it doesn’t work, then I go back to the drawing board.

Facebook Split Testing

The other cool thing which I don’t think enough people are using is Facebook split testing.

Has anyone actually use split tests in Facebook? Okay, so not enough people. There are only a few hands here.

It’s a really amazing tool.

Facebook split test, you can take a single audience, and you can run multiple ads or multiple strategies against the single audience. And Facebook will make sure that the audience will only see that one ad.

So when you run a normal ad set, Facebook was cycling those ads to one user. But this is a proper ABC way of testing.

And here’s an example that we ran for Black Friday.

Depesh Mandalia – Example (Black Friday)

Depesh Mandalia – Example (Black Friday)

We ran three offers to existing customers to figure out what’s the offer that best resonates and gives us the best performance. That we then go forward to our prospects on Black Friday.

And this actually gave us really deep intent behind offer three here, had the highest return on ad spend through split testing with our existing customers.

So we knew that’s what resonated with them and that’s what we went live with on Black Friday. That was part of the secret of our success. To be able to test that in advance and actually take it forward.

#6 Keep The Aisles Clear

Now, how many of you guys have actually been into a shopping center and the aisle is cluttered?

There are all things, kind of, all over the floor and you’re trying to make your way to the products, you’re trying to make your way to the cart.

And in fact, a lot of people will experience this in Bangkok. If you go and walk outside, there are people hassling you and they want to sell your stuff. They want to, I don’t know, they want to get you in their cab and all that kind of stuff.

That’s really frustrating but the fact is a lot of ecom owners are doing exactly the same on their stores.

Let me give you an example. An extreme example.

Depesh Mandalia – Example (Website)

Depesh Mandalia – Example (Website)

This website here is a complete mess.

Can anyone tell me what this website is about? On an immediate glance, you can’t.

Well, it’s actually a car rental website. And you wouldn’t think that.

Now let me tell you another thing. This website here is one of the highest revenue websites in the UK for car rental.

I’m not saying that you go and set your ecom stores up like this. But what I am saying is this person here knows her customer. She polarises her customer. So they either love this site or they’ll hate it.

And those that love it are her ideal customer. So don’t try and sell to everyone because it’s really hard to try and appeal to everyone.

Focus on your avatar, like who’s the person you’re selling to.


And the other thing I really, really want you to think about is when you’re running your ecom landers and you’ve got all these pop-ups and things going on, it is so annoying.

Depesh Mandalia – Example (Pop-ups

Depesh Mandalia – Example (Pop-ups

I guarantee you guys find it annoying, yet you still want to run it in your ecom stores because you want the data. You want email capture, you want people to like your Facebook page, you want people to subscribe to this, etc. etc.

You’re blocking people from doing what they are there to do. View your product, decide if they want it, add to cart, and buy.

So think about how you’re actually keeping that journey slick for users, but it’s not actually just for users.

There’s a big part of the Facebook algorithm that puts a big weighting on user satisfaction. So the user value.

The BEAR-V Algorithm

Depesh Mandalia – The BEAR-V Algorithm

Depesh Mandalia – The BEAR-V Algorithm

It’s a BEAR+V algorithm, it’s the advertiser bid and the estimated action rate.

So the bid is either automatic or manual depending on what you’re doing.

The estimated action rate is, Facebook says, “How likely is this person to take this action?”

This action could be purchase, initiate checkout, whatever you’re trying to do.

But then it looks at the user value. Are people bouncing from your website? Does your website load fast enough? How far do they go down, etc., etc?

And this plays a big part on your CPMs.

So when you’re complaining about high CPMs, look at your store. Is it loading fast enough? Are you making it slick, are you blocking people from actually buying through your store?

Because that has a big impact on your CPMs as well.

I’ll give you an example that we had a few months ago, were actually, one of our key leading products ran out of stock.

Depesh Mandalia – Out Of Stock

Depesh Mandalia – Out Of Stock

And we were stressing about Facebook and thinking, “You know, why’s conversion down, etc., etc.”

We figured out it was a stock issue, we managed to fix that. But that had a knock-on impact on our costs.

Because when something goes wrong on your website, Facebook doesn’t look at your stock and say, “Alright, yeah, I get it. You’re out of stock, so yeah, that’s fine.”

Facebook says, “I’m sending this traffic but it’s not converting, so maybe I’ll go and send this traffic because that traffic’s poor.”

Now all of a sudden, a good ad set goes bad because Facebook is looking at different traffic.

Any problem on your website, get it fixed ASAP because it does have a big impact on your bidding and on your ad set as well.

Loading Problems

The second example here is a website which had a ton of loading problems.

Depesh Mandalia – Loading Problems

Depesh Mandalia – Loading Problems

And we spent weeks and weeks trying to figure out how to get this website to load faster. It was on Shopify, had a ton of plug-ins.

If you’re running Shopify or WooCommerce and you’ve got plug-ins installed, they have a big impact on your load time.

We literally doubled ROAS by fixing up the page load time. That’s all we did.

Now, 3 days later and you can see in the top of that chart, this store owner actually put more plug-ins on the site and slowed the site down, which was really, really annoying.

But the fact is that your page speed does have a big, big impact as well.

#7 Ads Sniper

The 7th thing I want you to take away is a strategy that I call Ads Sniper.

When a lot of you are, kind of, running ads and you’re, kind of, figuring out what’s working, what’s not?

This is like the simplest strategy that you can put in place is to queue up your ads, look at your return on ad spend, and just kill the ones that are not performing.

You can see here, there are some ads that are below the average.

Depesh Mandalia – Ads

Depesh Mandalia – Ads

I’m looking at, for example here, I’ve grouped a number of ad sets together and I’m just looking at the ads that are underperforming and cutting them. Because they are losses for you.

And also, those losses have a negative impact on your ad set because the more those losses pick up, the more negative data your ad set picks up as well.

It makes it hard to scale and grow out of that as well.

System And Automation

Now I don’t know how many of you have watched The Founder. I found it a fascinating film.

So The Founder is talking about the foundations of McDonald’s and when it started off. One of the really cool things I loved about it, was the speedy system.

So McDonald’s literally invented the fast food business.

And this is when McDonald’s actually used to serve real meat. Not the McDonald’s that you know now. This is like 60-70 years ago.

What they handle is a real dedicated system and automation and everything just worked like clockwork. And that’s how I run my Facebook ads.

Everything is run by systems and automation.

Depesh Mandalia – Ads

Depesh Mandalia – Ads

And a big part of this is cutting bad ads, promoting good ads, increasing budgets, and using automation.

Like, take all that stuff that you do repetitively and automate it.

So if you’re seeing, for example, I look at last 3 days performance and if my ROAS is good, then I’ll bump the budget up.

Or I’ll have an alert, I use a tool called Reveal. I’ll send a little alert into our Slack channel. And an alert will go in and it will tell us that we can actually start to scale up even faster because these ad sets are working really well.

I don’t want to be spending time in my ads manager and neither should you. You want to be spending your time marketing.

So let the tools and automation do all the stuff that’s not really great use of your time and focus your time on research and testing and your customer.

I’m a bit tight on time, so I’m going to go a bit quicker here.

#8 The Reverse Sales Funnel

Number eight is the Reverse Sales Funnel. So here’s an example of how a lot of ecom store owners handle their business.

Depesh Mandalia – The Reverse Sales Funnel

Depesh Mandalia – The Reverse Sales Funnel

So a lot of you are like the hare here. So I don’t know if you’ve heard about the story of the hare and the tortoise, which one won the race?

This one actually goes to prove which one won the race but lot of you go for the first sale and then you stop. Because you’ve made your acquisition.

But that’s not how you win with ecom. Ecom is steady and slow, finding opportunities and growing up. But you don’t stop after you get that first acquisition.

You carry on going, you go and monetise that customer further and further and further. Because it’s far cheaper to retain a customer than it is to go and acquire again.

And this is the thing that stresses me out so much about people on Facebook ads. It’s yes, go and acquire your customer but keep on monetising them down the funnel.

Hot To Warm To Cold

Lifetime value is such a big thing about making Facebook ads work as well. And so what I call is the Reverse Funnel, so we’ve talked about previously going from, kind of, cold, warm to hot.

But on the back of that, once the customers purchase, they go from warm, from hot and warm to cold. And so you have to get them at the right time.

Depesh Mandalia – The Reverse Sales Funnel

Depesh Mandalia – The Reverse Sales Funnel

So 0 to 30 days after purchasing, maybe there’s an upsell or cross-sell. Even just a simple thank you video, maybe you get some referrals out of them.

If they haven’t purchased off the first 30 days, think about different strategies.

So new arrivals for example, generally, works well if your store has that. And then if they go into the kind of 60 to 90 days, they’re going to start tipping off.

And then maybe you start doing different things as well.

Keeping The People Engaged In Your Brand

So it’s all about kind of, monetising people and keeping them engaged in your brand.

Here’s an example of one I ran over summer, this was for a candle business. So the cool thing about candles is they burn out and people need new candles.

Depesh Mandalia – Candles

Depesh Mandalia – Candles

What we figured out is most people buy new candles between 30 and 60 days from their first purchase.

We ran a real simple retargeting campaign to existing customers that purchase between 30 days and 180 days.

And in that month, we spent $1,500 on it and generated $40,000 in return. The cost per acquisition was $2. That’s absolutely insane.

We were looking at this and saying to the client for many months, saying we need to do this, we need to do this, get the sign-off.

Just think how much money was left on the table if these customers didn’t come back. This is such an important part of making Facebook ads work for you. And really squeezing as much value out of your customers as possible that I just see lots of people missing out on.

And this kind of ad set is super, super simple to set up.

Depesh Mandalia – Setup

Depesh Mandalia – Setup

People have purchased in the one last 180 days, exclude people that have purchased in the last 30 days. And that’s it, that’s your audience. It’s super simple to set up, really quickly.

I encourage you if you’ve got an e-commerce store and you can get a future sell, go start testing that.

#9 Delight Your Customers

And the last one I want to share with you is delighting your customers.

Now, Facebook made a fundamental shift forward for e-commerce about 6 months ago and they’re now pushing this even harder.

Depesh Mandalia – Feedback

Depesh Mandalia – Feedback

Customer feedback, I don’t know if you guys have seen it on your stores yet, but it will really start to roll out on more and more stores. And what Facebook are doing, is they’re doing your customer service job for you.

They’re asking your customers, “Are you happy with this store?”

And I find it insane that people will go and check their customer feedback but they won’t ask their customers themselves.

They are your customers. You need to take responsibility before Facebook do and find out if your store or your store experience has a problem.

NPS: Net Promoter Score

And so what I use is a methodology called Net Promoter Score, NPS.

Has anyone heard of NPS? Awesome. So a few people.

NPS is a way of scoring the quality and satisfaction for your user right after they purchase, and it’s really simple.

So I use a tool called Delighted.com. There are lots of different tools out there.

It’s a simple question. “Would you recommend this brand, this product to your friends and family?”

It’s a score between 1 and 10. And the score calculates between those that gave you 9 and 10, so they’re your promoters. Those that give you a score between 1 and 6, they’re your detractors.

You take one from the other and you get your score.

Depesh Mandalia – Example

Depesh Mandalia – Example

So here, for example, it’s a scale of -100 to +100. 69 is pretty good.

I’ve seen businesses getting -40, -50, and guess what they’re the companies that fail. They’re the companies that go under.

Because they continue to spend money on acquisition. They disappoint their customers. Their customers don’t come back and they can’t monetise them.

And that’s where they struggle.

User Satisfaction

This is what Facebook is trying to get you to do, to improve your customer experience. But you should do it yourself.

Now, one little tip from this, which works time and time again is when you’re running your Net Promoter Score surveys.

If you actually pull out all your promoters into an email list and upload that as a custom audience and create a lookalike, time and time again and that will be your best audience.

Because these are people that love your brand.

So when you’re thinking about building custom audiences and lookalikes of your best customers, it always comes down to quality.

In this example, it was about 2,000 emails and these were people that absolutely loved the brand.

Depesh Mandalia – Example

Depesh Mandalia – Example

We were able to scale up the lookalike campaign because we found even more people similar to that as well.

Now, apologies if it’s felt a bit like this, where I’ve been hosing you with knowledge and information, but it’s been a really tight compressed amount of time.

But one thing I really want you to do and I’ve been in your seat, you know, I’ve attended conferences. Don’t go away.

Do Something And Make A Difference

And next week write up your notes and can I think about what I’m gonna do.

Whether it’s this presentation or any of the other ones, you go and implement the stuff that you think is going to make a difference like, right now, today, this evening, tomorrow, whenever it is.

Because I’ve done the same myself.

You go home, you forget some of the notes, your momentum is gone, and all that excitement is gone. And then you kind of lose out.

And like, I didn’t come all this way just to show you my chef’s hat and show you my cooking experience. I want you guys to actually improve your ecom results.

Because Facebook ads, who knows how long it’s going to last, how long we were going to be able to monetise it in the way that we are right now.

It could be 12 months, it could be 2 years, could be 3 years. But you need to take action right now. You need to start doing that test and learn.

And if you are successful, you might be able to bake a cake like this, which is what I do with Facebook ads.

Depesh Mandalia – Bake A Cake

Depesh Mandalia – Bake A Cake

And just remember ingredients, recipe, and experience.

The ingredients will change based on your store and your experience.

The recipe I’ve shared some of them with you and it’s about experimenting.

But the experience is what’s going to really cut through.

So like I said, I’ll give you some of the downloads on depeshmandalia.com/awa18. And there are contact details on there.

Depesh Mandalia

Depesh Mandalia

If you’ve got any other questions for me at all as well. It’s been an absolute pleasure and thank you for listening to me today.


Thank you very, very much. Let’s go over into the Q&A. So what kind of questions do we have? Do you translate your ads and landing pages when you target European countries?
Sure. So do you translate your ads and landing pages when you target European? Not when I’m testing. Because English is such a universal language. Like I see a lot of people strategising as you’ve seen those results back there, we were running English ads in Slovenia, in Ukraine, in Romania, and Poland, etc. And we’re driving good revenue. Some of these countries, we then find as a breakout and then we’ll translate. And we’ll do maybe currency or localised landing pages and carts and ads, etc. But for the sake of testing don’t go all in and translate everything. Test and see what the market is saying. If you’re not gonna be able to scale up in English, sure. But find the potential and then scale it up from there.
Is there somebody who wants to ask a question in person? Okay. Yes, please.

Mind Shift From Running Paid Search To Running Facebook Ads

Sure, so the question was, what was the mind shift from running paid search to running Facebook ads? And for me, it was about with paid search, I was bidding on intent. So someone searching for LCD TV or laptop or anything like that, they’re giving you an intent. Your goal is to get them out of the search result, into your lander, and try and convert them. Create a landing page relative to what they’re searching. And for me, it’s a lot simpler because you know what they’re searching for in most cases. So you build a funnel based on that. With Facebook, you’re creating intent. Like Facebook works at a higher level depending on how you’re running your campaigns. And so, the mind shift was moving from trying to get people out of the newsfeed and onto my site to only pulling out the right people from the newsfeed onto my site. So I don’t want a 100% click-through rate because no audience targeting is gonna be 100% spot on. I want the right people to click-through. So for example, where the ski-wear brand that you saw earlier, we were getting click-through rates of 0.9%, 1% but our conversion rates were 6 and 8%. That means more to me. And a lot of people focus on click-through rate. Only 2, 3, 4% click-through rate and the conversion rate drops because you’ve got the wrong people. So that’s the thing that kind of shifted it for me.
Thank you very much.