Hi there,
Thanks for reaching out!
Happy to give you some of my initial thoughts on your situation. It's a really common problem, especially for small businesses, so you're not alone. The usual advice to just 'test more interests' is pretty useless when there are no obvious interests to test in the first place, or when your starting with no data.
The issue isn't really about finding some magic, hidden interest in Meta's system. It's about fundamentally changing how you think about your audience and what you're actually offering them. It’s less about the technical buttons you press and more about the strategy behind it. Below I’ve outlined how I’d approach this from the ground up.
TLDR;
- Your Ideal Customer Profile (ICP) isn't a demographic; it's a specific, urgent, and expensive problem state or 'nightmare'. Stop targeting job titles and start targeting the pain.
- Your offer is your most powerful targeting tool. A strong, low-friction offer built to solve a small part of that nightmare will attract the right people and repel the wrong ones, even in a broad audience.
- When you have no direct interests, you must use 'proxy' interests—the tools, publications, influencers, and competitors your ideal customer already engages with.
- Your first campaigns are not for ROI; they are to intelligently buy data. You must run conversion-optimised campaigns from day one to feed the pixel with high-intent signals.
- This letter includes an interactive calculator to help you figure out your Customer Lifetime Value (LTV), which is essential for knowing how much you can afford to spend to acquire a customer.
We'll need to look at your ICP... it's a Nightmare, Not a Demographic
Right, let's get this out of the way first. Forget the sterile, demographic-based profile your last marketing hire made. "Small business owners in the UK" or "Companies in the finance sector with 10-50 employees" tells you absolutely nothing of value. It's why your ads feel generic and get ignored. To stop burning cash, you have to define your customer by their pain.
You need to become an expert in their specific, urgent, expensive, career-threatening nightmare. Your customer isn't just a job title; they're a person staring at a problem that's keeping them up at night. Your ICP isn't a person; it's a problem state.
Let's make this real. Imagine you're a marketing agency trying to sell services to other small businesses. Targeting the "Small Business Owners" interest on Meta is a complete waste of time. It's a bucket filled with everyone from a teenager selling handmade bracelets on Etsy to a local plumber to a venture-backed startup founder. Their problems are worlds apart. Your message will be so diluted it speaks to no one.
Instead, you need to isolate the nightmare. What is it, *specifically*?
- -> For a local restaurant owner, the nightmare isn't 'needing marketing'. It's 'seeing empty tables on a Friday night while the place across the street is packed, and worrying they won't be able to make payroll next month'.
- -> For a B2B SaaS founder, the nightmare isn't 'needing more leads'. It's 'watching their churn rate creep up every month, terrified that their product isn't as sticky as they thought and that their investors are getting nervous'.
- -> For a financial advisor, the nightmare isn't 'people needing financial advice'. It's 'a 55-year-old professional realizing they haven't saved nearly enough for retirement and are panicking that they'll have to work until they're 80'.
See the difference? The nightmare is emotional. It's specific. It's urgent. When you understand the nightmare, you can craft a message that feels like you're reading their mind. This work is the absolute foundation. If you don't do this first, you have no business spending a single pound on ads, because you're just guessing.
I'd say you need to build an offer they can't ignore
Once you've defined the nightmare, the next step is to create an offer that acts as the immediate painkiller. This is the second reason most small business campaigns fail: the offer is rubbish. The number one mistake I see is a high-friction Call to Action like "Contact Us" or, the worst offender of all, "Request a Demo".
The "Request a Demo" button is perhaps the most arrogant CTA ever conceived. It presumes your prospect, who is a busy business owner dealing with their nightmare, has nothing better to do than book a meeting to be sold to. It's high-friction, low-value, and instantly positions you as a commoditised vendor. You must do better.
Your offer’s only job is to deliver a moment of undeniable value—an "aha!" moment that makes the prospect sell themselves on your solution. You must solve a small, real problem for free to earn the right to solve the whole thing. For small businesses with no data, this low-friction offer is also your primary tool for *generating* that data. People will happily give you their contact details in exchange for genuine value.
What does this look like in practice?
- -> For a marketing agency targeting restaurants: Don't say "Get a free marketing consultation". Instead, offer a "Free Local Competitor Analysis". The ad could say: "Wondering why 'The Gastropub' is always full? Get a free 1-page report showing you their top 3 online strategies you can steal for your own restaurant this week." This is specific, valuable, and directly addresses their pain.
- -> For a B2B SaaS company: A free trial (no credit card required) is the gold standard. Let them use the actual product and feel the transformation. If that's not possible, offer a free tool. For a FinOps platform, it could be a 'Free AWS Bill Health Check' that finds their top 3 areas of wasted spend.
- -> For a financial advisor: Don't say "Book a free call". Offer a "Free 5-Minute Retirement Readiness Calculator". The ad copy hits the pain: "Worried you're behind on your retirement savings? Use our free calculator to see your projected shortfall and the #1 action you can take today to start closing the gap."
The offer and the ad copy work together as your first and most important targeting filter. Even if you target a slightly broader audience, the right message will call out to the people experiencing the nightmare and cause everyone else to scroll past. This is how you pre-qualify your audience before they even click.
1. Identify the Nightmare
e.g., A local tradesman is missing calls (and jobs) while on-site.
2. Craft a Painkiller Offer
e.g., A free guide: "The 5-Minute Setup to Never Miss a Customer Call Again (Even With Your Hands Full)".
3. Write "Mind-Reading" Copy
e.g., "Tired of your phone ringing off the hook while you're up a ladder? Stop losing £1000s in jobs to your competitors."
You probably should rethink your targeting strategy from the ground up
Okay, so you've defined the nightmare and created a killer offer. Now, how do we find these people on Meta when there are no direct interests to target? The answer is to use proxy interests.
You need to think like a detective. What are the other things your ideal customer is interested in that are unique to them? What tools do they use? What publications do they read? What influencers or brands do they follow? What software do they pay for? These are your proxies. They are clues that indicate someone is highly likely to be your ideal customer, even if they haven't explicitly listed "small business owner" in their profile.
Let's go back to our examples:
- -> Targeting Local Restaurants: You can't target "Restaurant Owner" directly. But you can target people who are admins of a Facebook Business Page AND are interested in 'Michelin Guide', 'OpenTable', 'Toast POS', or 'Restaurant Business Magazine'. Someone who likes all four is almost certainly involved in the restaurant industry.
- -> Targeting B2B SaaS Founders: You can't target "SaaS Founder". But you can target people interested in 'SaaStr', 'Jason Lemkin', 'Stratechery by Ben Thompson', 'HubSpot', and 'Salesforce'. You can even layer that with a behaviour like "Business Page Admins". This combination filters out most of the general population.
- -> Targeting Tradespeople (e.g., electricians, plumbers): This is trickier, but possible. You could target interests in specific tool brands they use exclusively (e.g., 'Milwaukee Tool', 'DeWalt'), trade publications ('Professional Electrician & Installer' magazine), or supplier brands like 'Screwfix' or 'Toolstation', layered with a location and age demographic.
The key here is to build a hypothesis for each audience. Your hypothesis isn't "I think small business owners will like this". It's "I believe that people who are admins of a business page AND are interested in OpenTable are highly likely to be restaurant decision-makers who are struggling to get more bookings, and therefore will respond to my offer for a free competitor analysis."
You then create separate ad sets to test each of these hypotheses. Don't lump all your proxy interests into one giant ad set. You need to isolate them to see which 'clue' is the strongest indicator of your ideal customer. Maybe the 'Tools' audience works brilliantly, but the 'Publications' audience is a dud. You won't know unless you test them seperately.
You'll need to buy data, not just impressions
This brings us to the most important mindset shift for anyone starting with a new ad account. Your initial budget—the first £500, £1000, even £2000—is not for generating a positive ROI. The goal of that initial spend is to buy data.
You are paying Meta to help you identify who, within your proxy interest audiences, actually responds to your offer. Every click, every landing page view, and especially every lead form submission is a precious piece of data that feeds your Meta Pixel.
This is why you must, must, must avoid the trap of running "Reach" or "Brand Awareness" campaigns. This is an uncomfortable truth, but when you set your campaign objective to "Awareness," you are giving the algorithm a very specific command: "Find me the largest number of people for the lowest possible price."
The algorithm does exactly what you asked. It seeks out the users inside your targeting who are least likely to click, least likely to engage, and absolutely, positively least likely to ever buy anything. Why? Because those users are not in demand. Their attention is cheap. You are actively paying the world's most powerful advertising machine to find you the worst possible audience for your product. It's a total waste of money.
From day one, every single campaign you run should be optimised for a conversion event. For a new business, this probably means a 'Lead' objective, tied to your high-value free offer. You want to tell Meta, "Go find me people within this 'OpenTable' interest group who are most similar to the people who have already downloaded my free guide." The algorithm is incredibly good at this, but it needs those initial conversion signals to learn.
This is also where understanding your numbers becomes critical. How much can you afford to spend to get that initial data? That depends entirely on what a customer is worth to you over their lifetime. This is called Customer Lifetime Value (LTV). Most small businesses have no idea what their LTV is, so they can't make intelligent decisions about their ad spend. They just focus on getting the cheapest lead possible, even if those leads are low quality and never convert.
Let's do some quick maths. The real question isn't "How low can my CPL go?" but "How high a CPL can I afford to acquire a truly great customer?"
Interactive LTV & Affordable CPA Calculator
Suddenly, paying £20, £30, or even £50 for a lead doesn't seem so expensive when you know a customer is worth £10,000 to your business, does it? It looks like a bargain. This is the maths that unlocks aggressive, intelligent growth and frees you from the tyranny of cheap leads.
You'll need a proper structure to turn data into results
So, you're armed with your customer nightmare, a great offer, some proxy interests, and an understanding of what you can afford to pay for data. How do you structure your campaigns to actually execute this? I would use a phased approach.
Phase 1: Data Acquisition (First 2-4 weeks, or until ~100-200 conversions)
The goal here is purely to gather data and identify which of your proxy interest hypotheses is correct.
- -> Campaign Setup: Create one Campaign Budget Optimisation (CBO) campaign. Set the objective to 'Leads' or whatever your primary conversion event is.
- -> Ad Set Setup: Inside this campaign, create 3-5 different ad sets. Each ad set will target one of your proxy interest 'themes'. For example:
- Ad Set 1: Tools Audience (e.g., interests in OpenTable, Toast POS)
- Ad Set 2: Publications Audience (e.g., interests in Restaurant Business Magazine)
- Ad Set 3: Competitor Audience (e.g., people who like pages of famous local restaurants)
- -> Creative: Use 2-3 of your best, most pain-focused ads in each ad set. Let Meta's CBO allocate the budget to the ad sets and ads that start getting the cheapest conversions.
- -> Analysis: After a week or so, you'll start to see which ad set is performing best. This tells you which 'proxy' is the strongest signal for your ICP. You'll turn off the losers and potentially reallocate budget to the winner.
Phase 2: Scaling with Lookalikes and Retargeting (After 100+ conversions)
Once your pixel has seasoned with at least 100 high-quality conversions (people who downloaded your guide, filled out your form etc.), you unlock Meta's most powerful tool: Lookalike Audiences.
- -> Create Custom Audiences: First, you need to create custom audiences from the data you've just bought. The most important one is "Audience of all Leads from the last 90 days". Others include "Website Visitors 30 days", "Viewed 50% of Video Ad", etc.
- -> Build Lookalike Audiences: Now, you tell Meta to create a Lookalike Audience based on your "Leads" custom audience. Start with a 1% Lookalike in your target country. This tells Meta: "Go and find me the 1% of users in the UK who are most statistically similar to the people who have already become a lead." This is almost always going to outperform any interest-based targeting you can think of.
- -> Launch New Campaigns: Now you can start building out your structure.
- ToFu (Top of Funnel) Campaign: This is your prospecting campaign. You can now create a new ad set targeting your new 1% Lookalike audience. You'll likely find your Cost Per Lead drops significantly.
- MoFu/BoFu (Middle/Bottom of Funnel) Campaign: This is a separate campaign for retargeting. Your audience here would be your Custom Audience of "Website Visitors 30 days" (excluding anyone who already became a lead). The ads here are different. They might show a customer testimonial, handle a common objection, or present the main offer now that the person is familiar with you.
This phased approach takes you systematically from having no data and no idea who to target, to having a predictable system for attracting your ideal customer. It takes a bit of patience and a willingness to invest in data upfront, but its a far more reliable path to success than just randomly testing interests and hoping for the best.
This is the main advice I have for you:
I've detailed my main recommendations for you below in a table to give you a clear, actionable path to follow. This is the process we'd use for a client in your situation.
| Step | Action to Take | Why This is Important |
|---|---|---|
| 1. Define ICP | Stop using vague demographics. Define your ideal customer by their most urgent, expensive 'nightmare'. Write it down in one sentence. | This forces you to create messaging that is hyper-relevant and emotionally resonant, which is your best targeting filter. |
| 2. Craft Offer | Create a high-value, low-friction 'painkiller' offer (e.g., a free tool, guide, checklist, or audit) that solves a small part of their nightmare for free. | This builds trust, provides value upfront, and gives you a compelling reason for them to give you their contact info, feeding your pixel with data. |
| 3. Brainstorm Proxies | List 10-15 'proxy' interests: tools, publications, influencers, and competitors that your nightmare-experiencing ICP is likely to engage with. | These are your starting audiences when direct interests aren't available. They are educated guesses that you will validate with data. |
| 4. Launch Data Campaign | Set up a single CBO campaign with a 'Lead' objective. Create 3-5 ad sets, each testing one of your proxy interest themes. Your goal is 100+ conversions. | This is how you intelligently buy data. You're not aiming for ROI yet; you're paying Meta to identify the winning audience segment for you. |
| 5. Analyse & Learn | After spending enough to get meaningful data, identify the winning ad set(s). Turn off the poor performers. | This validates your audience hypothesis and tells you where to focus your future prospecting budget for the best results. |
| 6. Build & Scale | Once you have 100+ conversions, create a Custom Audience of those leads. From that, create a 1% Lookalike Audience and launch a new prospecting ad set targeting them. | This is how you scale. Lookalikes of high-intent converters are Meta's most powerful targeting tool and will almost always outperform interest targeting. |
| 7. Implement Retargeting | Create a separate campaign to retarget website visitors who did not convert. Use different ads that build more trust or present a more direct offer. | It often takes multiple touchpoints to convert a lead. Retargeting recovers potentially lost sales and increases your overall campaign efficiency. |
As you can probably tell, this is a strategic shift, not just a tactical tweak. It requires a deep dive into your customer's psychology, crafting a compelling offer, and a disciplined, structured approach to testing and data analysis. It's a lot more involved than just boosting a post or chucking a few interests into an ad set.
Executing this properly—analysing the data correctly, knowing when to scale and when to kill an ad set, writing copy that truly connects—takes time and experience. Many business owners find they get stuck on one of the steps or misinterpret the data, which can end up being quite costly.
We do this all day, every day for our clients. For one of our B2B software clients, we generated 4,622 registrations at just $2.38 each. For another client, a medical job matching platform, we were able to bring their Cost Per Acquisition down from £100 to just £7.
If you'd like to walk through your specific business and how this framework could be applied, we offer a free, no-obligation initial consultation. We can have a look at what you've tried so far and give you some concrete advice on the best way forward. Let me know if that's something you'd be interested in.
Hope this helps!
Regards,
Team @ Lukas Holschuh
