Solved: Sudden Ad Spend Spike and Bot-like Behavior

Hi there,

Thanks for reaching out about the strange behaviour you're seeing with your Meta ads campaign. That sounds like a proper frustrating situation. It's actually more common than you'd think, and it's one of those weird platform quirks that can make you want to throw your laptop out the window. Happy to give you some initial thoughts and guidance on what likely happened and how you can prevent it from wrecking your budget again.

The short answer is that while it feels like a bot attack, it's probably the Meta algorithm doing exactly what it's designed to do, just in a way that's completely useless for you as an advertiser. It found the cheapest, least engaged audience possible and dumped your budget on them in a few hours. Let's get into why that happens and how to build a campaign structure that forces the platform to find actual buyers instead of just clicks and impressions.

So, what on earth actually happened to your ad spend?

Alright, let's break down that bizarre chain of events. You switch on a campaign that's worked before, and within hours it's burned through €17.50, clocked up over 32,000 impressions on a single ad, and initially reported zero landing page views. Your first instinct is to blame bots, and while there's always a background level of invalid traffic, it's rarely the culprit for something this specific. This smells like an algorithmic problem.

Think of the Meta algorithm like a hyper-efficient but very literal employee. You've given it a budget and an audience, and its primary job is to spend that budget reaching people in that audience. When you switch a campaign on, especially one that's been dormant, it enters a frantic "learning" phase. Sometimes, it stumbles upon a massive, readily available segment of users within your targeting parameters that are incredibly cheap to show ads to. These are the users who endlessly scroll but never click, never engage, and certainly never buy. Their 'attention' is worthless, so it's sold for pennies. Your campaign hit a goldmine of duds.

The fact your CPM was €0.54 is the smoking gun. In any decent European market, a CPM under €5 is suspicious; under €1 is practically unheard of for any audience with commercial value. This tells us the algorithm found the absolute bottom of the barrel. The accompanying 0.5% CTR then confirms that this audience was completely unresponsive. They saw the ad, but it meant nothing to them.

What about the landing page views that appeared later? That’s just standard reporting lag. The Ads Manager dashboard is notoriously slow to update attribution data. Clicks are registered almost instantly, but the confirmation that a user landed on your page and the pixel fired correctly can take hours to filter through. So while it looked like 32,000 impressions yielded 0 clicks, the reality was just a reporting delay. The 98 views that eventually showed up against the 32k impressions just confirms how poor the audience quality was.

The most telling part is that you still got some Amazon sales. This proves the campaign wasn't completely broken, but it was phenomenally inefficient. It's like trying to catch fish by throwing dynamite in the ocean. You might get a few, but you'll waste a huge amount of energy and cause a lot of collateral damage. Your €17.50 found one or two actual potential customers amongst a sea of 32,000 zombies. We need to stop fishing with dynamite and start using the right bait in the right spot.

You're asking Facebook to find non-customers

Here’s the uncomfortable truth that most people don't want to hear. When you set your objective to something like 'Reach' or 'Brand Awareness', you're explicitly telling Meta: "Find me the most eyeballs for the least money." The algorithm obediently goes off and finds users whose attention is cheap because they are not in demand. They don't click, they don't buy, they just scroll.

But here's the kicker: even when you select a 'Conversions' or 'Sales' objective, if your audience targeting is too broad, stale, or poorly defined, the algorithm can easily revert to this default behaviour. It needs data to optimise for conversions. If it doesn't get conversion signals quickly, it will fall back on what it knows how to do best: spend your money to get impressions. Your campaign objective becomes a suggestion, not a command. You were trying to find buyers, but you ended up paying the world's most sophisticated advertising machine to find the absolute worst audience for your product. Awareness is a byproduct of making sales to the right people, not a pre-requisite for it.

We'll need to look at your customer's nightmare, not their demographic

So, how do we fix this? We start by tearing up the traditional idea of an Ideal Customer Profile (ICP). Forget the sterile, demographic-based profile like "Females, aged 25-40, interested in Amazon and online shopping." This tells you nothing of value and leads to generic ads that speak to millions but resonate with no one. This is exactly the kind of broad targeting that leads to the impression dump you experienced. To stop burning cash, you must define your customer by their pain.

You need to become an expert in their specific, urgent, expensive nightmare. What is the deep-seated problem that your product solves? Your customer isn't just a job title or an age bracket; they are in a 'problem state'.

Let's imagine you sell handcrafted, ethically sourced jewellery on Amazon.

• The demographic is: "Woman, 30-50, high household income, likes luxury brands". It's useless.
• The nightmare is: "I have a close friend's 40th birthday next week. She's impossible to buy for because she has everything and hates mass-produced, soulless tat from big brands. I'm terrified of giving her another generic gift that ends up in a charity shop. I need to find something unique and meaningful that shows I put real thought into it, and I need it delivered fast."

See the difference? The nightmare is an emotional scenario. It's a story. And it's packed with targeting clues. The person in this nightmare isn't just 'interested in jewellery'. They are likely following independent artists on Instagram, reading blogs about sustainable living, shopping on sites like Etsy or Not On The High Street, and maybe even members of Facebook groups dedicated to supporting small businesses.

Your job isn't just to sell a product; it's to sell relief from that nightmare. Once you've isolated that pain point, you can build your entire targeting strategy around it. This intelligence is the blueprint that stops you from wasting a single extra pound on ads. Do this work first, or you have no business running another campaign.

I'd say you need to structure your campaigns for profit, not impressions

Once you know who you’re talking to, you need to structure your campaigns in a way that guides them from discovering you to buying from you. Just running one ad set with seven ads is a recipe for the kind of chaos you just saw. A proper funnel structure is non-negotiable.

We use a simple but powerful ToFu/MoFu/BoFu (Top of Funnel, Middle of Funnel, Bottom of Funnel) approach. This lets us segment our audiences and messaging logically, ensuring we're not asking for a sale from someone who has never heard of us before.

Here’s how I would prioritise the audiences for a business like yours, assuming you're driving traffic from Meta to your Amazon listings. This is how you build a system that forces the algorithm to behave.

META ADS AUDIENCE PRIORITISATION FOR ECOMMERCE

ToFu (Top of Funnel - Cold Audiences): Finding New People

This is where you find people who fit your 'Nightmare ICP' but have never heard of you.

• -> 1. Detailed Targeting: This is your starting point. Based on your ICP research, you'd target interests like 'Etsy', 'Not On The High Street', specific ethical fashion magazines, influential designers in your niche, etc. Group them into themed ad sets (e.g., one for competitor brands, one for related publications).
• -> 2. Lookalike Audiences: Once you have enough data, these are gold. You can create lookalikes based on:
  • - Your Amazon customer list (if you can export it).
  • - People who have engaged with your Facebook or Instagram page.
  • - People who have watched 50% or more of your video ads.
  Start with a 1% lookalike in your primary countries, as it's the most precise.
• -> 3. Broad Targeting: Only test this once your Meta Pixel has thousands of conversion events. It can work, but you need to give the algorithm a lot of data to find the right people on its own. It's not for new or recovering accounts.

MoFu (Middle of Funnel - Warm Audiences): Re-engaging Interest

These are people who have shown some interest but haven't gone to your Amazon page yet. You want to build trust and remind them you exist.

• -> People who have engaged with any of your posts or ads in the last 30-90 days.
• -> People who have visited your Facebook or Instagram profile.
• -> People who have watched a significant portion (e.g., 10 seconds or more) of your video ads.

The messaging here is different. It's less about "buy now" and more about "here's what makes our products special" or showing customer testimonials.

BoFu (Bottom of Funnel - Hot Audiences): Closing the Sale

This is your highest-intent audience. They know who you are and have been close to buying. Tbh, this is difficult with Amazon sales as you don't have website pixel data like 'Add to Cart' or 'Initiate Checkout'. However, we can create a proxy for it:

• -> People who have clicked the link in your ads in the last 7-14 days but haven't been recorded as a purchaser (attribution is tricky here, but it's the best we can do). This is your most important retargeting audience. Your message to them should be direct, maybe with an offer of scarcity ("Limited stock!") to encourage immediate action.

By seperating your campaigns like this, you control the budget and messaging for each stage. You stop Meta from taking the easy way out and dumping all your money into a cheap, useless ToFu segment. You force it to spend money on the audiences most likely to convert.

You probably should calculate how much a customer is actually worth

This brings me to the most important mindset shift in all of paid advertising. The question isn't "How low can my CPM go?" but "How high a cost-per-purchase can I afford to acquire a great customer?" To answer that, you need to know your Customer Lifetime Value (LTV).

Obsessing over cheap metrics like CPM and CPC is what gets people into the mess you just experienced. It encourages the algorithm to find you rubbish traffic. When you know your LTV, you know what a customer is actually worth, which tells you how much you can sanely spend to get one. For an eCommerce business selling on Amazon, a simple LTV calculation would be:

LTV = (Average Order Value * Average Number of Repeat Purchases) * Gross Margin %

Let's say your average product sells for €50. Your gross margin after Amazon fees, cost of goods, etc., is 60%. And you find that, on average, a customer buys from you 1.5 times over their lifetime.

LTV = (€50 * 1.5) * 0.60
LTV = €75 * 0.60 = €45

So, each new customer you acquire is worth €45 in gross profit to your business. A standard rule of thumb is to aim for a 3:1 LTV to Customer Acquisition Cost (CAC) ratio. This means you can afford to spend up to €15 (€45 / 3) to acquire a single new customer and still run a very healthy, profitable business.

Suddenly, that €0.54 CPM doesn't look so attractive, does it? It looks like a trap. You'd be far better off with a €15 CPM if it meant you were reaching an audience that converted and gave you a €14 cost-per-purchase. This is the maths that unlocks intelligent, scalable growth. Use the calculator below to get a feel for your own numbers.

You'll need a message they can't ignore

Finally, having the right audience and structure means nothing if your ad creative is boring. Your ads need to speak directly to the 'nightmare' we identified earlier. You don't sell products; you sell solutions to problems. The best way to do this is with proven copywriting frameworks.

For a product-based business like yours, the Before-After-Bridge framework is incredibly effective.

• Before: Describe their current world, filled with the 'nightmare'. Paint a picture of the frustration.
• After: Describe the world after using your product. The problem is solved, and they feel relief and joy.
• Bridge: Position your product as the bridge that gets them from Before to After.

Let's write an ad for our hypothetical handcrafted jewellery seller using this framework:

Ad Copy Example:

Headline: Stop Giving Gifts They'll Forget.

Body Text:
(Before) That feeling of dread as you scroll endlessly through Amazon, knowing whatever you pick will be another mass-produced item that ends up in the back of a drawer. You want to give something special, something with a story, but you're out of time and ideas.

(After) Imagine their face lighting up as they open a gift that's truly unique. A piece that doesn't just look beautiful, but tells a story of craftsmanship and care. You're not just giving a necklace; you're giving a piece of art that makes them feel seen and cherished.

(Bridge) Our collection of handcrafted silver jewellery, made by artisans in the Scottish Highlands, is the bridge to that moment. Each piece is unique, just like the person you're giving it to. Click to discover the collection and give a gift that truly means something. Shipped via Amazon for fast, reliable delivery.

This ad doesn't talk about the type of silver or the dimensions of the pendant. It talks about the emotional journey of the buyer. It connects with their nightmare and sells them the dream. This is what gets clicks from the right people and drives your CTR up, which in turn tells the algorithm to find more people like them.

I've detailed my main recommendations for you below:

As you can see, the problem wasn't just a simple glitch. It was a symptom of a deeper strategic issue with the campaign's foundation. It's not just about pushing the right buttons in Ads Manager; it's about understanding the psychology of your customer and the mechanics of the platform, and then building a system where those two things work in harmony.

Getting this right can be tricky, and it involves a lot of testing and refinement. That's often where professional help can make a huge difference. An experienced eye can spot these foundational issues quickly and implement a robust strategy that not only prevents disasters like the one you had but also builds a reliable, scalable engine for growth for your business.

If you'd like to go over your account and strategy in more detail, we offer a free, no-obligation initial consultation where we can look at your specific setup and give you some more tailored advise. It's a great way to get a second opinion and a clear action plan.

Hope this helps!

Regards,

Team @ Lukas Holschuh