The Meta pixel and Conversions API, set up properly

Your Meta pixel and your Conversions API data is the most valuable thing inside your ad account right now, and most likely you’re not using it correctly, and it’s probably not set up correctly. In this video I’m going to go over everything: how we properly set up a Meta pixel and a Conversions API, the best practices, why it actually matters, and how you should structure it with your landing pages. The basis of this, to get right into it, is that if your pixel data is bad, your leads in the future are bad as well. If your pixel data is good, your leads in the future are going to be good as well. What you give back to the pixel is what you’re going to get in the future.

A lot of people think your pixel is just there to track and measure how many leads you’re getting. That is completely false. That’s like saying you’re going to use a machete to cut a banana. That is not the point at all. The pixel doesn’t measure your leads. It decides who your next leads are. Every time a lead event fires, when you’re optimising a campaign for a specific lead event, you’re telling Meta: this is what I want, go and find more people like this. People who are in this industry, who click this button, who submit this application, who have this behaviour on your landing page. So you’re going to get a lot of people like that.

That is why, when you’re running campaigns, you’ll see a lot of very similar people coming from your ads over time. There aren’t a lot of outliers. It isn’t very broad. You’ll get very similar industries, very similar revenue ranges, requesting amounts and similar answers, because if your pixel is set up like this you’re sending pixel data back to Meta saying: get me more people like this. Give me more people like John’s Trucking. Give me more people doing $50K a month. That is what you’re saying to Meta. You’re basically saying go and find more people like them. If it fires on every form submission, Meta hunts for people who submit forms.

If you have it set up incorrectly, this can also work against you. If there’s no qualification and the pixel fires on every single form submission, for everyone, people making zero dollars a month, it’s going to send you more people making zero dollars a month. So it’s a double-edged sword. The main thing is that you should only be firing it after qualification, a one-time passcode and a mobile or email verification. That way you have a good cost per lead and a good-quality merchant, and you’re sending good data back to Meta. It’s a balancing act. Same spend, same creatives, but you can get a completely different lead if your pixel is set up correctly. If it isn’t, you’re just going to be draining your budget and you’re not really going to be making the most of your ad account. That is why I said at the beginning that the most valuable asset within your ad account right now is your pixel, because it stores all of the data. It’s not your creatives, it’s not your campaign, it’s not the numbers in your Ads Manager. It is your pixel, because that is what carries all the information and knows exactly who your customer is.

Now, what everyone does, and what you should not do, is this. This is what most people have set up, and it is why most people say Facebook leads are garbage: because they have their pixel set up incorrectly. I’ve seen it time and time again. I get on a lot of consulting calls and people have their pixel set up completely incorrectly. It’s just so wrong. And that is why they are paying $60 for a merchant that isn’t even qualified. So you have your Facebook ads, your creatives, your landing pages, everything like that. Then form submission: the person submits how much money they make, how much they’re looking for, what industry they’re in, their information.

And then a lead fires, regardless of whether they make zero dollars a month, whether they’re a startup, whether they’re a trucking company or a medical spa, whether they make $100K a month. It just fires on every single submission. You’re basically telling Meta: find me more people who submit a form. That’s it. There’s no other criteria, no other barrier to entry. And that is bad, because you’re telling Meta to just find people who are submitting applications, and Meta is going to do exactly what you tell it to do. It’s going to find the cheapest possible way to get you form submissions, from anyone. That is why your results are going to be a race to the bottom. You’re going to start getting very low-quality leads, because you’re letting anyone through. It’s like a gate, and you’re letting anyone through. More people who submit forms, but not merchants who fund.

The leads this gets you are startups, $5K-a-month businesses, people who are over-leveraged, third and fourth positions nobody wants, merchants who have defaulted before, and spam — you can get that anywhere, but it is included too. So you are intentionally telling Meta to get you low-quality leads. Meta is not stupid. Meta is a great multi-billion-dollar advertising platform. It knows exactly what you want and it is going to do exactly what you ask. Setting up your pixel like this is prompting it to give you bad-quality leads. And that is why you should be doing this instead. There should be two exits off the form, and only one of them talks to Meta.

So you have your Meta ads, you have your landing page, same as before, and then you have your gates: your form, your conditional logic. This is your minimum criteria. This is what we run as our minimum criteria. You can have lower, you can have higher, whatever you want — it’s your funnel. They submit the application, and if they’re qualified they pass and go on to the page I’ll go over in a second. If they fail, they go to a disqualification page. No one-time passcode, nothing. And that data is never sent back to Facebook, because you do not want more people like that. You do not want more unqualified people.

You can obviously still call these people. You can sell them on. You can still service them in some way, shape or form, but you do not want to optimise your campaign towards them. You already paid for them, so you might as well call them and try to get them funded. But the main thing is that you do not want to optimise your future leads for them. After someone completes a form and meets all the requirements — they make $50K a month, they’re requesting, let’s say, $30K, everything is good — they do a one-time passcode. A six-digit code goes to their phone number and back into the application. They have a verified email address. All of this is good. They are sent to a qualified landing page, where the lead event on the pixel fires and is sent to Meta.

That is saying: tell Meta to get more people who are likely to go onto the landing page, fill out the application in this way, submit the application and go on to this page. That is precisely what you are telling Meta, and what you tell Meta is what you’re going to get from Meta. So this will get you leads that are actually doing $30K a month, six months in business, first or second positions only, clean statements, no negative balance days, asking for a good amount, and verified mobile phone numbers. You need to set this up. This is the bare minimum. The disqualified merchant still lands somewhere real. You could still service them in some way, shape or form. They just never become a data point that goes back to Meta.

The truth is that a monkey can install a pixel. It’s literally a few clicks on your screen. That is not the hard part. The part that actually matters is making sure it fires and only sends back the information that is actually useful to you. So number four is the actual build. Meta sends traffic. There are two gates, and green passes both. The red stops at the first, and only what passes a gate goes back to Meta. What that means, long story short, is that there are different funnel events: there’s a qualified form submission, there’s a funded deal, there are offline events, there are submissions to the lender. All these different things that you can send back to Meta as data, with different values and different scores for the lead, and that influences what Meta is going to bring you next.

As you can see here, this is Meta. This is the creative engine: your ad accounts, your ad creatives, your budget, your landing pages, everything. They submit an application. If they’re disqualified they still go to a landing page, they just never get sent back to Meta, because you don’t want to optimise off that. But if you keep going, you qualify a form submission like I showed you above, and you send that back to Meta. You can take it a step further. Inside your CRM you can have offline events: if a merchant submits an application, you can optimise towards that, and you can send hashed data back to Meta so that it is accurate and it knows exactly which lead it is talking about. You’re basically saying: look, Meta, give me more people who submitted a true funding application. And then you can send that back.

The thing with this is that you need at least fifty conversion events a week. That is why you can only go further into the funnel if you are spending more money on ads. Then you can go all the way to a funded deal, which obviously means you need to be funding fifty deals a week for that to make sense, because that is how many conversion events you need. So that is how that works. Basically, every step of the way you can send back more data to Meta to make it more accurate, and to make sure Meta knows exactly who you want on the application.

Events Manager is just the inbox, and the green things land in it. Everything Meta learns comes back through the loops. Two loops carry good data back; the third one carries nothing, because you don’t want to send that data back. Now, what you’ll notice is that there is a pixel, and then there is the Conversions API. The pixel is what is happening on the actual phone; the Conversions API is happening on the server side. For example, here you can see the browser pixel, the lead, the ID. It is the same thing with the Conversions API, but you also carry the hashed data.

After the iOS updates — honestly, I don’t even know, a few years ago now — this kind of got screwed up. And then there is the server side, which always goes through no matter what. What happens is that when the pixel fires, both of these go through. Sometimes the first one doesn’t go through, but either way, when you send the information back to the Events Manager it deduplicates. There is no duplicate: it matches this ID with this ID, and then it matches everything up in the Events Manager and takes all that data. The browser copy is the merchant’s phone telling Meta. Phones can get blocked — the iOS updates, the ad blockers, the cookies — so sometimes it never arrives. The server copy, which is the Conversions API, is your system telling Meta. It cannot be blocked, and it is sending that hashed data. Both of them carry the same event ID, so either way it lands in the Events Manager.

Step five is basically the step-by-step explanation. You can actually just download this document, put it into Claude or whatever AI you use, and have it guide you step by step through setting up the Conversions API and the Meta pixel. Step one, you create the pixel in Events Manager. One pixel per funnel: don’t share a pixel between a funding offer and a credit repair offer, or you’ll break both. Step two, you install the base code on every single page of your landing page — the initial heading page, every application page, the thank-you page, everything. Make sure that it is working. Step three, you build the form with qualification in mind. You want that conditional logic for how much money they’re making, how much they’re requesting, whether they’re US-based, what state they’re in, all of this. You only want to send the good-quality merchants back to Meta, which is easy with any landing page software.

Step four is that you build the two destination landing pages, like I showed you above: the qualified and the unqualified page. Only the qualified page should be firing the pixel and sending the data back to Meta. Two separate URLs, not one thank-you page with different text. Step five is putting in the one-time passcode. You can use something like Twilio — that is what we use — to have the one-time passcode sent to the prospect’s phone number in order to actually submit the application. Step six, fire the lead on the qualified page and nowhere else. Only the lead event you’re optimising the campaign for in Meta should be on the qualified thank-you page. Not on the submit button, not on the form start, not on anything else. It fires once they have verified their phone number, because you are trying to optimise as far down the funnel as it makes sense to.

Step seven, add the Conversions API for the same event. You’re going to have to generate an access token and add that into the landing page as well. Send the hashed data — email, phone, first name, last name, state — plus IP and user agent. That is what pushes Event Match Quality up, and it is why you verify phone and email in the first place. This is easy, but you want to make sure you set it up correctly. Every landing page builder is going to be a little different, so use the guide. Step eight is deduplication. You don’t really have to set anything up; you just have to make sure it is actually working. When you run a test in Events Manager, make sure it is flagging the event ID and bringing the two together, like that graphic. You’ll run a test and it will bring in the event from the pixel and then from the API, and you need to make sure it deduplicates. If it doesn’t, there is an issue: every lead is counting twice, your cost per lead looks amazing and the optimisation is running on fiction.

Next is testing everything. You have to run it through the funnel. If it fires on a disqualified lead, something is wrong and you need to fix it. Nothing else matters until that is fixed. And then step ten is that you set the campaign to optimise towards the lead event, and the qualified page also fires a webhook into your CRM so the rep is dialling inside sixty seconds. That way you are only going to get more of the high-quality merchants you already worked for.

Step six is what to send back by budget. Depending on the budget you are running these campaigns at, you are going to be sending back different stages within the funnel. The main thing is that you need at least fifty optimisation events per ad set per week. If you have anything less, Meta is not going to have enough data to optimise the campaign and find the people you actually want. So if you are under a hundred bucks a day, you should just not be running ads. You need a bare minimum of $100 a day. If you run $100 a day to a thousand bucks a day, you optimise for a lead.

A qualified lead can be them making $30K a month, $50K a month, $100K a month. You can move that slider up as you go on, so that you skew the pixel and it is only getting you higher-quality leads. But, like I said, you need at least fifty per week. After $1,000 a day that you are spending — so $30K a month — that is when you can start to add offline events, like funding application submissions and funded deals. But, like I mentioned, you still need fifty events a week, and that is why it only works at higher volumes. This lead event is where most of you are going to be, and where you can skew it: $30K a month, $50K a month. You can change the credit score qualification, so that you only send back people with a 650 credit score or more, requesting $50K or more. Whatever you want.

And then, as I mentioned, there are the funded deals. You set up the funded event back to the original click using the hashed data for the email and the phone number of that specific lead. That only works because you verified both of them on the form. You basically set a funded deal as a Purchase, which is the actual conversion event, and then there are different values: $60K, $250K, $10K. Each value is going to weigh differently in Meta when you send that data back. So if John’s Trucking only funded for $20K and Frank’s Trucking funded for $100K, there is going to be a different value there. You send the hashed data — the email, the phone number, the state, all of that — back to Meta so that it is as accurate as possible. Like I said, the fifty a week is really important, so that it has enough data to actually optimise off.

Number eight is a very slept-on strategy, which is using Microsoft Clarity. This gives you very good clarity on what is happening on your landing pages at each step. It gives you heat maps of where people are looking, where people are spending time, where people are dropping off. And it is also free. You just install it, and this is how it looks on your side. You can see where people are clicking, whether people are clicking back, whether people are trying to see things, how long people are on landing pages. It is a heat map. And you also see where most of your traffic is: are they mostly mobile, mostly on desktop, iPad, whatever it is.

And this is how we found out that bank statement uploads were killing good merchants. We saw that when people were getting to the bank statement step, they were just leaving. We watched them reach the step and leave. So we removed it, significantly lowered our cost per lead, and we were still able to get high-quality merchants. It was a win-win for us, and without Microsoft Clarity we would not have found it. Step nine is the biggest landing page mistakes in general, with tracking and everything. If you have one thank-you page for everyone, that is a big red flag, because you are sending both your qualified and your disqualified leads to the same URL, and you cannot have a properly set up pixel if you are doing that.

Optimising for funded deals at only five a week: it is great to optimise for funded deals, it is awesome, but if you are not funding fifty deals a week it is not going to work. You need big volume, a big budget, to actually do that. Any offline events, like I mentioned, need at least a thousand bucks a day from what we have seen. Number three, a Conversions API with no hashed data. That defeats the whole purpose. You need the hashed data, because if not, Meta cannot attribute anything to those specific leads. Number four, if you have one pixel and multiple different offers, multiple different products that you are selling, different angles in general — if you have a debt consolidation offer plus an MCA offer under the same pixel, it is not going to work. They are going to contradict each other and both of them are just going to not work well.

And then, of course, not using Microsoft Clarity: you are just guessing. It is all based on emotion, and there is no specific data you can back up and actually know what to make decisions on. So those are the biggest mistakes I usually see whenever I hop on a consulting call. And then number ten is just a checklist you can use when you are actually setting this up properly, or resetting it, or if you want to make sure your setup is working correctly. Go through this checklist and have all of that going.

Like I said, this document is going to be linked below. You can upload it to your AI and walk through it to set up your Conversions API, your tracking and everything on your landing page. Just ask it questions and it can reference this document as a guide. But that is everything. Hopefully now you understand why the pixel and the Conversions API are so important. It is, for the third time now, the most valuable asset within your ad account, especially if you have spent money on it already. So make sure that it is set up correctly. It is not rocket science. Make sure that you are conditioning it well. If you have bad leads being sent back through the pixel, that is hurting your pixel big time. So make sure you have good hygiene with your pixel.

If you have any questions, let me know. Hopefully you found some value in this video. If you are an MCA brokerage with ten or more reps that is looking for more high-quality, real-time and exclusive leads, and you have made it to this point in the video, click the link below and book a call with us. We sell them on a pay-per-qualified-lead basis, all using the strategies and concepts I talk about in my videos, and from there we can show you how it works and show you the system. We deliver real-time, exclusive merchants doing $30K a month in revenue, requesting $15,000 or more, with four months of bank statements, six months in business and a mobile-verified phone number — all the things I talked about in this video. And we deliver them straight to your CRM. So if that sounds interesting, book a call below and we will talk.