Solved: Inconsistent Ad Performance Across Meta and Google

Hi there,

Thanks for reaching out!

I had a look at the situation you described, and that feeling of constantly reacting to your ad performance is something I see a lot. It’s a frustrating cycle to be stuck in. You're essentially playing a game of whack-a-mole with your budget, and it's an exhausting way to try and grow a business. I'm happy to give you some initial thoughts and guidance on how to break out of that loop. The solution isn't about finding the perfect budget split, it's about building a proper system that can weather the natural ups and downs of these platforms.

We'll need to look at why you're stuck in this reactive loop...

Right now, you're treating Meta and Google Ads like two seperate thermostats you're constantly adjusting to keep the temperature 'just right'. One month Meta is hot, so you turn up the heat there and turn down Google. The next month, it's the other way around. Tbh, this is a flawed approach because you're misunderstanding what these platforms fundamentally are. They're not the same kind of tool, and they will never behave in the same way.

Google Search is a demand harvesting platform. You are catching people who are already actively looking for a solution. They've raised their hand and typed "electrician near me" or "best accounting software for small business" into the search bar. The performance here is limited by the number of people searching for those things each month, which naturally fluctuates. So if search volume for your key terms dips one month, your performance will dip too. It’s that simple.

Meta (Facebook & Instagram) is a demand generation platform. You are interrupting people while they are scrolling through photos of their friends' holidays. You have to grab their attention and convince them they have a problem they might not have even been thinking about two seconds ago. It's a completely different ball game. The algorithm is designed to find pockets of people who respond to your ad. It will hammer that audience until it's saturated, ad fatigue sets in, and performance plummets. This is not a bug; it's a feature of the platform. That's why your results are so volatile. The system is literally designed to work that way.

The constant budget shifting between them is like trying to fix a boat with two different leaks by bailing water from one side to the other. You're busy, but you're not actually fixing the underlying holes. The real problem is you lack a strategic system for each platform that accounts for their unique behaviours.

A big part of this is likely your campaign objectives. I see this all the time. People run "Reach" or "Brand Awareness" campaigns thinking they're building a brand. Here’s the uncomfortable truth: you are paying the algorithm to find you the worst possible audience. You've given it a command to "find me the cheapest eyeballs possible", and it does exactly that. It finds people who are least likely to click, engage, or ever buy anything, because their attention is not in demand and therefore cheap. The best awareness you can get is a sale. You should almost always be running conversion-focused campaigns, optimising for leads, signups, or sales. It forces the algorithm to find people who actually take action, which is a far better use of your money.

I'd say you need to stop thinking about platforms and start thinking about people...

Before you even think about audiences or ad copy again, you need to throw out your current idea of a customer profile. I’m willing to bet it looks something like "Companies in the finance sector with 50-200 employees" or "Women aged 25-45 who like yoga". This kind of demographic data is almost useless. It leads to generic ads that speak to everyone and therefore resonate with no one.

To stop burning cash, you have to define your customer by their pain. Their specific, urgent, expensive, career-threatening nightmare. Your customer isn't a job title; they are a person in a problem state. What keeps them awake at 3 AM? What's the one thing that, if it went wrong, could get them fired?

Let's take an example. Say you sell a project management SaaS tool. Your old ICP was "CTOs at tech startups". Your new ICP is "The CTO who is terrified of her best developers quitting out of frustration with a broken, chaotic workflow." See the difference? One is a job title. The other is a nightmare.

Once you've isolated that nightmare, your entire strategy changes. You're no longer targeting "CTOs". You're targeting the specific digital places where that nightmare is discussed. You find the niche podcasts they listen to on their commute, the industry newsletters they actually open, the specific influencers they follow on LinkedIn, the private Slack communities they're a part of. This isn't just data; it’s the blueprint for your entire targeting strategy. You have to do this work first, or you have no business spending another pound on ads.

You probably should rebuild your offer before you spend another pound...

Now we get to what is probably the single biggest reason for inconsistent campaign performance: the offer. If your offer is weak, even the best targeting in the world won't save you. When the algorithm hits a rough patch or audience fatigue sets in, a powerful offer can still pull conversions through. A weak one will fall flat on its face.

The most common offender I see, especially in B2B, is the "Request a Demo" button. This is perhaps the most arrogant Call to Action ever invented. It assumes your prospect, who is likely a busy, important person, has nothing better to do than schedule a meeting to be sold to. It's high-friction, low-value, and immediately signals that you're just another vendor in a sea of sameness. You are asking them for their time before you've given them any value. It's no wonder your conversion rates are inconsistent.

Your offer’s only job is to deliver an "aha!" moment. A moment of undeniable value that makes the prospect sell themselves on your solution. You must solve a small, real problem for them for free to earn the right to solve their entire problem for a price.

What does this look like in practice?

• For a SaaS Company: A free trial or a freemium plan, no credit card required. Let them use the actual product and feel the transformation. This generates Product Qualified Leads (PQLs) who are already convinced, not Marketing Qualified Leads (MQLs) for a sales team to chase. I remember one B2B SaaS client who implemented this approach with Meta ads and generated over 1,500 trials.
• For an Agency/Consultancy: A free, automated audit tool. For example, a marketing agency could offer a free SEO audit that shows a prospect their top 3 keyword opportunities. For us, it's a free 20-minute strategy session where we audit failing ad campaigns.
• For a High-Ticket Service: A valuable asset. A corporate training company could offer a free 15-minute interactive video module on 'Handling Difficult Conversations'. It gives a taste of their expertise and solves a genuine problem for a new manager.

Look at your own offer right now. Is it asking or giving? If it’s asking for a demo, a call, or a meeting, you need to change it. Build a value-first offer, and you'll build a much more resilient foundation for your advertising.

You'll need to understand the numbers that actually matter...

The cycle you're in—tweaking budgets based on last month's CPL (Cost Per Lead) or CPA (Cost Per Acquisition)—is a direct result of focusing on the wrong metrics. Chasing the lowest possible CPL is a race to the bottom. It forces you to use cheap awareness objectives and target low-quality audiences, which ultimately yields poor results and puts you right back in that reactive loop.

The real question isn't "How low can my CPL go?" but "How high a CPL can I afford to acquire a great customer?" The answer to that question lies in its counterpart: Customer Lifetime Value (LTV).

Most businesses I talk to either don't know their LTV or they've calculated it incorrectly. But it's the master metric that should guide your entire advertising spend. Let's break it down simply:

1. Average Revenue Per Account (ARPA): How much revenue, on average, does a single customer bring in per month?
2. Gross Margin %: What is your profit margin on that revenue? (Revenue - Cost of Goods Sold) / Revenue.
3. Monthly Churn Rate %: What percentage of your customers do you lose each month?

The calculation is straightforward:

LTV = (ARPA * Gross Margin %) / Monthly Churn Rate

For example, if your ARPA is £500, your Gross Margin is 80%, and your Monthly Churn is 4%, your LTV is (£500 * 0.80) / 0.04, which equals £10,000. Each customer is worth £10,000 in gross margin to your business over their lifetime.

This number changes everything. A healthy LTV to Customer Acquisition Cost (CAC) ratio is typically 3:1. This means for a £10,000 LTV, you can afford to spend up to £3,333 to acquire a single customer and still run a very profitable business. If your sales process converts 1 in 10 qualified leads into a customer, you can afford to pay up to £333 per qualified lead. Suddenly that £50 lead from Google or that £75 lead from Meta doesn't look so expensive, does it? It looks like a bargain.

This is the math that unlocks intelligent, aggressive growth. It gives you a clear ceiling for your acquisition costs and frees you from the tyranny of cheap leads. Use the calculator below to get a sense of your own LTV.

We'll need to look at building a proper Meta Ads system...

Okay, with the foundations of ICP, Offer, and LTV in place, we can now build a system for Meta that embraces its volatility instead of fighting it. As I mentioned, the algorithm will find a winning audience, exhaust it, and then performance will crash. Your job isn't to prevent the crash; it's to have the next test ready to go. You need a structured, systematic approach to testing audiences and creative.

This is how I prioritise audiences for any account, from eCommerce to B2B SaaS. It moves from the coldest audiences at the Top of the Funnel (ToFu) to your hottest prospects at the Bottom of the Funnel (BoFu).

META ADS AUDIENCE PRIORITISATION:

• ToFu (Top of Funnel - Cold Audiences): This is where you find new people.
  1. Detailed Targeting: Start here for new accounts. Target interests, behaviours, and demographics based on your "Nightmare ICP" research. Think about what software they use, what publications they read, what events they attend.
  2. Lookalike Audiences: Once you have data (at least 100 conversions, but ideally more), this becomes your most powerful tool. Build lookalikes in this order of priority: previous customers, initiated checkouts/completed lead forms, add to carts, website visitors. A 1% lookalike of your best customers is gold dust.
  3. Broad Targeting: Only test this once your pixel has thousands of conversion events. You target a country and maybe an age/gender, and let the algorithm do the work. It can perform incredibly well, but only on mature accounts.
• MoFu (Middle of Funnel - Warm Audiences): People who have engaged but not converted yet.
  • -> Retarget all website visitors from the last 30-90 days (exclude converters).
  • -> Retarget people who watched 50% of your video ads.
  • -> Retarget your social media page engagers.
• BoFu (Bottom of Funnel - Hot Audiences): People who were on the verge of converting.
  • -> Retarget people who added a product to their cart in the last 7-14 days (for eCommerce).
  • -> Retarget people who visited your checkout or pricing page.
  • -> This is where you show your most direct, high-intent ads. Maybe a testimonial or a special offer.

You should have seperate, always-on campaigns for each stage of this funnel. Your job then becomes a rythm of testing. Every week or two, you launch a new ad set in your ToFu campaign testing a new lookalike or interest group. You launch a new ad creative against your best-performing audiences. You monitor the results, turn off the losers that spend more than your target CAC without converting, and scale the winners. This creates a predictable system for growth out of an unpredictable platform.

Now, let's build a stable Google Ads foundation...

With Google, stability comes from owning the right auctions. The volatility you're seeing likely comes from two places: fluctuating search volume for your main keywords, or you're bidding on keywords with the wrong intent.

Just because a keyword gets a lot of searches doesn't mean it's valuable. You need to focus on keywords that signal high commercial intent. Someone searching "what is CRM software" is just doing research. Someone searching "best crm for sales teams uk" is looking to buy. The second keyword is far more valuable, even if it has lower search volume.

You need to structure your campaigns around this idea of intent. Here’s a simple breakdown:

Your search campaigns should be tightly themed around these high-intent keyword groups. You should also be using negative keywords aggressively to filter out irrelevant, low-intent searches. Someone searching for "plumber jobs" shouldn't see your ad for plumbing services. This alone can save a huge amount of wasted spend and stablise performance.

By focusing your budget on the keywords that are most likely to convert, you become less susceptible to broad fluctuations in overall search volume. You're fishing in a smaller, richer pond. This, combined with a clear understanding of your LTV:CAC target, allows you to confidently set your bids to win the auctions that truly matter, creating a much more predictable return.

This is the main advice I have for you:

Switching from a reactive to a systematic approach is a big change, but it's the only way to get off the performance rollercoaster. I've detailed my main recommendations for you below in a step-by-step action plan.

As you can see, this is a significant shift in thinking and requires a fair bit of work to set up correctly. This is where expert help can make a huge difference. Our job isn't just to 'run ads'; it's to design and install this entire growth system into a client's business so they have a predictable, scalable way to acquire customers.

If you'd like to go over how this system could be specifically applied to your business, I'd be happy to schedule a complimentary, no-obligation strategy session with you. We can take a deeper look at your accounts and map out a more detailed plan.

Hope that helps!

Regards,

Team @ Lukas Holschuh