Solved: Dwindling Ad Performance - Should I Start From Scratch?

Hi there,

Thanks for reaching out! I had a look at the situation you described, and honestly, it's a really common problem, especially for accounts that have been running successfully for a long time. It’s incredibly frustrating when you see performance just fall off a cliff like that after being so consistent. I'm happy to give you some of my initial thoughts and guidance based on my experience with this sort of thing.

First off, my gut reaction to your main question – should you start from scratch? – is almost certainly no. I know it feels like a drastic problem that needs a drastic solution, but starting a new ad account is usually the last resort, a real nuclear option. It might seem like a fresh start, but you'd be throwing away five years of valuable data, and that's a huge asset you don't want to lose. The issue is very unlikely to be the account itself being 'broken'. It's much more likely a mix of other factors that have built up over time. We just need to figure out what they are and tackle them systematically.

I'd say you shouldn't start from scratch just yet...

Let's talk about why hitting the reset button is probably the wrong move. When you start a new ad account and a new pixel, you're essentially telling Meta you're a complete beginner. All that history, all the data about who your best customers are, what they respond to, and how they behave on your site – it's all gone. Your pixel would be completely cold, which means the algorithm has zero information to work with when it tries to find you new customers. The initial 'learning phase' for your new campaigns would be incredibly expensive and slow, and you'd likely see even worse performance than you're seeing now, at least for a good while.

Think of your pixel data as a highly detailed map of where to find treasure. For five years, you've been adding landmarks, marking paths, and highlighting the best spots. Starting a new pixel is like throwing that map away and starting with a completely blank piece of paper in a new, unfamiliar forest. It's a massive step backwards.

The problem you're describing – a sudden, sharp decline in performance after a period of high spend and success – almost always points to audience saturation and creative fatigue, not a technical fault with the account. You've likely shown your ads to the most responsive people in your audiences so many times that they're no longer effective. A new account won't magically find new people; it will just start the same process over again, but from a much weaker position. Instead of starting over, the better approach is to revitalise what you already have. We need to do a bit of a deep clean and rebuild your strategy from within the account you've spent years building. It’s a fixable problem, it just requires a bit of a structured approach rather than a panic button.

We'll need to look at what's really going on...

Before we change anything, we need to do a proper diagnosis. It’s like being a detective at a crime scene. Where are the clues? Where is the drop-off happening? You need to look at the entire customer journey, from the moment they see your ad to the moment they complete a purchase. This is where your metrics are your best friend.

Let's break it down into stages:

1. The Ad Level (Top of Funnel):
Start by looking at your core ad metrics within Meta's dashboard. How have they changed since before February 23rd?

-> Click-Through Rate (CTR): Has your CTR dropped significantly? If people are seeing your ads but not clicking on them anymore, it's a massive red flag that your creative has gone stale. This is what we call creative fatigue. An image or video that was getting you great results six months ago might be completely invisible to your audience now. They've seen it too many times, and they just scroll past.

-> Cost Per Click (CPC): Has your CPC shot up? This is often linked to a low CTR. Meta rewards engaging ads with lower costs. If your ads aren't getting clicks, Meta will charge you more to show them. If both your CTR is down and CPC is up, it's almost definately a creative problem. Your ad copy and your images/videos need an urgent refresh. This is the first and most common culprit.

2. The Website/Landing Page Level (Middle of Funnel):
Okay, so let's say people are still clicking the ads, but the sales aren't coming through. Now we need to follow them to your website. You need to look at your website analytics (like Google Analytics or Shopify Analytics) and compare the data.

-> Lots of Clicks, but low "Add to Carts": If you're getting plenty of traffic to your product pages but very few people are adding items to their cart, the issue lies on the product page itself. This could be a few things. Are your product photos compelling enough? Are your product descriptions persuasive? Is the pricing right? Perhaps a competitor has launched a similar product with a better offer, or your shipping costs are too high. People are interested enough to click, but something on that page is putting them off taking the next step.

-> Lots of "Add to Carts", but low "Initiate Checkouts" or Purchases: This is a classic one. If people are adding to cart but then abandoning the purchase, the problem is usually in your checkout process. Are there unexpected shipping costs that only appear at the last minute? This is a huge conversion killer. Is your checkout process long and complicated? Do you require people to create an account? You need to make it as seamless and trustworthy as possible. Things like trust badges (Visa, Mastercard, PayPal), customer reviews, and clear return policies can make a big difference here. Your site might not look as trustworthy as it needs to, and that hesitation is enough to lose the sale.

I remember one client selling cleaning products. Their sales had slumped, and they were convinced their ads had stopped working. We looked at their funnel and found that while their CTR was okay, their Add to Cart rate was terrible. We worked with them to get new product photography and rewrote the descriptions to focus on the benefits, not just the features. That simple change, combined with a better audience structure, led to a 633% return and a 190% increase in revenue. It wasn't the ads that were 'broken'; it was a block in their sales funnel.

Here’s a quick way to think about it:

By breaking it down like this, you can pinpoint the exact stage where your customers are dropping off. That tells you where to focus your energy, rather than just blindly starting a new account and hoping for the best.

You probably should rebuild your audience targeting...

This is the big one. After the creative, your audience targeting is the most likely reason for the performance drop. When you're spending $2000 a day, you burn through audiences very, very quickly. Even if an audience has millions of people, the number of people within that audience who are *actually* likely to buy from you is much smaller. You've likely reached all the low-hanging fruit, and now Meta is struggling to find more people like them, which drives your costs up and your conversion rate down. This is audience saturation.

So, we dont need a new account, we need a new audience strategy within your existing account. As a consultant, this is one of the first things I overhaul in a client's account. Most people aren't structured or specific enough. Here’s how I would prioritise it:

The Funnel Structure: ToFu, MoFu, BoFu

You need to think of your campaigns in terms of a funnel. Top of Funnel (ToFu) is for cold audiences who've never heard of you. Middle of Funnel (MoFu) is for people who have engaged with you somehow but haven't gone to your site yet. Bottom of Funnel (BoFu) is for warm audiences – your website visitors, people who've added to cart, etc. You should have seperate campaigns for each stage.

1. Bottom of Funnel (BoFu) - Your Hottest Audiences:
This is your retargeting. These are the people most likely to convert, and you need to be very specific here. Don't just lump all "website visitors in 180 days" into one audience. Break it down!

-> Highest Priority: Added to Cart (last 7-14 days), Initiated Checkout (last 7-14 days). These people were *this close* to buying. Hit them with ads that overcome objections. Maybe a small discount code ("Complete your order and get 10% off!"), a reminder of what they left behind, or customer testimonials.

-> Medium Priority: Viewed Content / Product Page Visitors (last 14-30 days). These people showed interest in specific products. You can retarget them with carousel ads showing the products they viewed, plus some similar ones.

-> Lower Priority: All Website Visitors (last 30-90 days). This is a broader group. You can show them your best-sellers or brand-building content.

You also have your previous customers. These are perfect for creating Lookalike audiences, which we'll get to, but you can also run campaigns to them to encourage repeat purchases.

2. Top of Funnel (ToFu) - Finding New Customers:
This is where you've likely seen the biggest drop. Your old reliable audiences are exhausted. It's time to build new ones.

-> Lookalike Audiences (LALs): This should be your number one tool for prospecting. But you have to use them correctly. I often see clients making LALs from their entire website visitor list. That’s not specific enough. You want to create LALs from your *best* customers. In order of priority, you should test LALs of:
1. Your highest lifetime value customers (you can upload this as a custom list).
2. All Purchasers (180 days).
3. Initiated Checkouts (180 days).
4. Added to Cart (180 days).
Start with a 1% LAL in your main countries, then test 1-3%, 3-5% etc. A 1% LAL of your actual buyers is far more powerful than a 5% LAL of people who just visited your homepage.

-> Detailed Targeting (Interests): This is likely where you've been operating. It's time for a rethink. Don't use broad interests. If you sell high-end hiking gear, don't just target "Hiking." That's too broad. Think deeper. What magazines do serious hikers read? (e.g., Backpacker Magazine). What brands do they trust? (e.g., Patagonia, Arc'teryx - if available for targeting). What specific activities are they into? (e.g., "Thru-hiking"). You need to find interests that are more specific to your ideal customer and less likely to be used by all your competitors. Group these into themed ad sets and test them against each other.

I remember one client who sold outdoor equipment, and their campaigns were struggling. They were just targeting "outdoors" and "camping." We helped them build a new campaign structure based on this funnel model and got them testing niche interest groups like specific national parks, gear blogs, and smaller competitor brands. That restructure alone was enough to get them 18k high-quality website visitors and turn their performance around. It just took a more thoughtful approach to audience building.

You'll need to test new creatives and offers...

So, we've covered the 'who' (audiences), now let's talk about the 'what' (your ads). As I mentioned in the diagnostics part, if your CTR is low, your creative is tired. After 5 years, it's easy to fall into a pattern of what you *think* works, but tastes and platforms change.

Creative Testing is Non-Negotiable
You need to be constantly testing new creatives. This doesn't have to be a massive production every time. Here's a simple testing framework:

-> Test Different Formats: If you only use static images, you absolutely have to test video. It doesn't have to be a Hollywood production. Sometimes, a simple User-Generated Content (UGC) style video shot on a phone can outperform a polished, expensive ad because it feels more authentic. We've seen this with several SaaS clients where simple UGC videos showing the software in action got amazing results. The same principle applies to eCommerce. Show real people using and enjoying your products. Also test Carousels and Collection ads, which are fantastic for eCommerce.

-> Test Different Hooks: The first 3 seconds of your video or the headline of your image ad are everything. You need to stop the scroll. Try testing different hooks for the same ad. For example:
- Hook 1 (Problem-focused): "Tired of [problem your product solves]?"
- Hook 2 (Benefit-focused): "The secret to achieving [desired outcome]."
- Hook 3 (Social Proof): "See why 10,000 customers are raving about this."

-> Test Different Angles: Your product probably has multiple benefits. Try creating ads that focus on each one individually. One ad might be about the quality and craftsmanship, another about the value for money, and a third about how it makes the customer feel. You don't know which message will resonate best with a new audience until you test it.

What About Your Offer?
Sometimes, the issue isn't just the ad, but the offer itself has lost its punch. The market moves fast. Take an honest look at your main competitors. What are they offering? Have they started offering free shipping on all orders? Are they running a "buy one, get one" deal? Is their base price now lower than yours? A drop in sales can sometimes be a sign that your offer is no longer competitive. It might be time to test a new offer, like a bundle deal, a first-time customer discount, or a free gift with purchase, to re-ignite interest.

For a women's apparel client, performance had plateaued. We implemented a robust testing strategy across both Meta and Pinterest. We tested new image styles (moving from studio shots to lifestyle shots), new video formats, and new ad copy. This systematic approach to creative renewal, combined with the right audience structure, generated a 691% return. You can't let your creative get stale; it's the engine of your campaigns.

This is the main advice I have for you:

So, to pull all that together, I know this is a lot to take in. The key takeaway is that you shouldn't panic and delete your account. The problem is almost certainly solvable with a structured, methodical refresh of your strategy. You have the asset – five years of data. Now you just need to use it more effectively.

I've detailed my main recommendations for you in a table below. This is the action plan I'd suggest you follow.

I know that implementing all of this, especially when you're also trying to run the business, can feel like a full-time job in itself. It takes a lot of time and a specific kind of expertise to diagnose, restructure, and manage these campaigns effectively.

That's where a professional consultancy like us can make a huge difference. With years of experience and having turned around dozens of accounts in situations just like yours, we can help you identify the best strategies and take over the implementation for you. I can offer a free initial consultation where we can go through your account and strategy together. It's usually super helpful for business owners and gives you a taste of the expertise we can bring to your project. If that's something you'd be interested in, just let me know.

Either way, I hope this detailed breakdown gives you a clear path forward and helps you get things back on track.

Regards,

Team @ Lukas Holschuh