Solved: Should I Replace My Facebook Pixel After Pandemic?

Hi there,

Thanks for reaching out!

I had a look over the situation you described. It's a really common problem, actually, especially for businesses that saw a big shift in their customer base over the last few years. It's easy for your ad account's data to get a bit muddled when the type of person buying from you changes so suddenly. Your instinct to segment your customers and focus on the pre-2020 group is spot on – that's exactly where I'd start thinking.

You've asked a few really good questions about using a new pixel, and I'm happy to give you some of my thoughts and guidance on how I'd approach this. It's a bit of a deep topic, but getting it right can completely change your results. The short answer is I would strongly advise against starting a fresh pixel. The longer answer is about refining the data you have, not throwing it away.

We'll need to look at your Pixel and Data Strategy...

First off, let's tackle your main question: should you ditch the old pixel? My honest advice is no, definately not. The single biggest asset an ad account has is its historical data. Your pixel has spent years learning who visits your site, who browses, who adds to cart, and most importantly, who buys. Even with the "bad" customers from the COVID-era, there's a mountain of valuable data in there that the Meta algorithm uses to find new customers. Starting from scratch would be like wiping the memory of your most experienced salesperson. You'd be back to square one, and your costs would likely go way up while the new pixel goes through its learning phase all over again, which can be a painful and expensive process.

Think of the pixel as a dumb collector. It just records events. It doesn't judge whether a customer was "good" or "bad". The real inteligence comes from how you, the advertiser, use that collected data to build audiences. Your problem isn't a "contaminated" pixel; it's that you're probably letting Meta's algorithm look at the entire pool of data to find new people, including the low-quality segment you want to avoid. The solution is to be more specific and tell the algorithm exactly which slice of that data to model its new audiences on.

Your idea of running two pixels side-by-side is creative, but it's an unneccessary complication that I think would cause more headaches than it solves. You'd be splitting your data, which means neither pixel would have a complete picture of performance. This would make it harder for Meta to optimise delivery, potentially driving up costs on both accounts. You'd also have the nightmare of managing two sets of tracking and reporting. We can achieve the exact same goal – focusing on your high-value customers – in a much cleaner, more efficient way within your existing ad account and with your single, data-rich pixel.

I'd say you need to master your Audience Segmentation...

This is the core of the solution. You're already on the right path creating segments from your pre-2020 customers, so let's expand on that and build a really robust strategy. You need to become the master of your audiences, telling Meta with surgical precision who to target and, just as importantly, who to exclude.

When we take on a new account, especially for an eCommerce client, we build the audience strategy around a clear funnel structure: Top of Funnel (ToFu), Middle of Funnel (MoFu), and Bottom of Funnel (BoFu). Your audiences should be organised this way too.

Here’s a breakdown of how we prioritise audiences, from coldest to warmest. This is the playbook you should be working from:

So, how do you apply this? Here are your actionable steps:

Step 1: Create Your "Gold Standard" Custom Audience.
Take your list of pre-2020 customers (a CSV file with email addresses or phone numbers is perfect). Go to the 'Audiences' section in your Meta Ads Manager, click 'Create Audience' -> 'Custom Audience' -> 'Customer List'. Upload your file. Meta will match these users to their profiles. Name this audience something clear like "Pre-2020 High-Value Customers". This is now your gold-standard source of data.

Step 2: Create Your "Exclusion" Custom Audience.
Now do the exact same thing, but with a list of all the "bad fit" customers who purchased from 2020 onwards. Create a separate Custom Audience from this list and name it "Post-2020 Exclusion List". This list is just as powerful as your good one, because it lets you tell Meta exactly who not to show ads to.

Step 3: Build Your Hero Lookalike Audience.
Go back to 'Audiences' and this time click 'Create Audience' -> 'Lookalike Audience'. For your source, select your "Pre-2020 High-Value Customers" Custom Audience. Start by creating a 1% Lookalike for your primary target country (e.g., United Kingdom 1%). A 1% lookalike is the smallest and most similar group of people to your source audience. This will be your most potent cold-traffic audience. I remember one eCommerce client where getting their audience source right was a huge part of how we helped them achieve a 1000% return on ad spend.

You probably should rethink your Campaign Structure...

Having the right audiences is half the battle; the other half is using them in a structured way. A messy campaign structure can kill performance just as quickly as bad targeting. You should mirror the ToFu/MoFu/BoFu funnel.

A simple, powerful setup would be:

1. Campaign 1: Prospecting (ToFu)
   Objective: Conversions (Purchases). This campaign's job is to find new customers. Inside this campaign, you'll create different ad sets to test your audiences against each other.
   • Ad Set 1: Your new "1% Lookalike of Pre-2020 Customers".
   • Ad Set 2: A different Lookalike, maybe 2% or one based on "All Purchasers".
   • Ad Set 3: An interest-based audience you think is relevant.
   In all these ad sets, you must add your "Post-2020 Exclusion List" to the exclusions. This is critical. You're now actively cleaning your targeting. You also need to exclude all recent purchasers (e.g., last 30 days) and all your website visitors (e.g., last 30 days) because you'll target them in the next campaign.


2. Campaign 2: Retargeting (MoFu/BoFu)
   Objective: Conversions (Purchases). This campaign's job is to bring back people who have already shown interest.
   • Ad Set 1: Website Visitors (e.g., last 30 days), excluding purchasers.
   • Ad Set 2: Added to Cart (e.g., last 14 days), excluding purchasers.
   • Ad Set 3: Social Engagers (e.g., last 60 days), excluding website visitors and purchasers.
   The creative here should be different. You might offer a small discount, show testimonials, or answer common questions to get them over the line.

This structure allows you to control your budget properly, deliver the right message to the right person, and clearly see what's working. The golden rule of testing is to let the data decide. Don't turn off an ad set after a day. Let it spend at least 2-3 times your target cost-per-purchase before you make a call. If your target CPA is £30, let an ad set spend £60-£90 before you decide it's not a winner. Be patient and methodical. A proper testing framework was key to hitting a 691% return for one women's apparel client on Meta and Pinterest ads.

You'll need to go beyond just the ads...

One last thing to consider. Sometimes, the problem isn't just the audience targeting. The ads can be doing their job perfectly by bringing people to your site, but if the site itself isn't resonating with the right kind of customer, they won't convert.

Ask yourself honestly: has your website, your product descriptions, your photography, and your pricing kept pace with the type of high-value customer you want to attract now? The "COVID money" customers might have been less discerning, but your ideal pre-2020 customers might expect something more.

Look at your store through their eyes:

• -> Do your product photos look proffesional and high-quality?
• -> Is your product copy persuasive and focused on the benefits that your ideal customer cares about?
• -> Does your website look trustworthy? Do you have clear reviews, contact information, and a professional design?
• -> Is your offer still compelling? Has the market changed?

Improving your website's conversion rate by even a small amount can have a huge impact on your ad performance. Better on-site experience helps to naturally filter out the bad-fit customers and makes every pound you spend on ads work harder. Strong, persuasive copppy and a clear, trustworthy website will attract the customers you want and repel the ones you don't, putting less pressure on your ad targeting to do all the heavy lifting.

This is the main advice I have for you:

To pull this all together, here is a table summarising the actionable plan I would recommend for you to implement. This is a step-by-step process to take back control of your ad account and start attracting the right customers again.

As you can see, getting this right involves quite a few moving parts, and it goes much deeper than just the pixel. It's about a complete strategy that covers data analysis, audience psychology, campaign architecture, and funnel optimisation. It's not just about setting up an ad and hoping for the best; it's about building a predictable system for attracting the right people.

This is what we do all day, every day for our clients. We take the guesswork out of it and implement these kinds of proven structures to drive real results. If you feel like this is a bit overwhelming and you'd like an expert eye on your ad account to build a more hands-on plan together, we offer a free initial consultation. We can go through your specific setup and give you some clear, actionable steps forward.

Hope this helps give you a new direction!

Regards,

Team @ Lukas Holschuh