The short version
- Search advertising catches demand that already exists. Facebook and Instagram create it. Those are different jobs and they need different expectations.
- The genuine strengths are reach at low cost, retargeting, and getting in front of people who do not yet know they need you.
- The genuine weakness for B2B is intent. Nobody opens Instagram intending to procure a supplier.
- Creative now does more work than targeting. The platform finds the audience; your job is to give it something worth showing.
The case for Facebook advertising used to be made with audience-size statistics, as though scale alone were the argument. It never really was, and for a business selling a considered purchase it is close to irrelevant. Being able to reach a very large number of people is only useful if some of them are going to buy from you.
So here is the more useful version of the question. Meta’s platforms do certain things better than anything else available, and certain things badly. Knowing which is which is the whole of the decision.
The fundamental difference from search
Someone typing “commercial epoxy flooring contractor” into Google has a problem and is looking for a supplier. You are answering a question that has already been asked. That is why search converts well and why it is competitive and expensive.
Someone scrolling Instagram at nine in the evening is not looking for you. They have no active requirement, and your ad is an interruption. That is not a criticism, it is simply the mechanism, and it explains almost everything about how Meta advertising behaves.
The practical consequence: judged on the same conversion metrics as search, paid social will nearly always look worse. Judged on what it is actually for, it can be the cheapest way to build the demand that search later harvests.
What it is genuinely good at
Reaching people before they are in the market
Most of your future customers are not searching for you today. Search cannot reach them at all. Meta can, cheaply, and repeatedly. For businesses with a long buying cycle, being familiar before the requirement exists is worth a great deal, because you enter the shortlist rather than having to displace someone on it.
Retargeting
The most reliably effective use, and the one we would set up first. Somebody read three pages of your site and left without enquiring. Reaching that person again for a modest cost is straightforward on Meta and much harder anywhere else. It works because you are not interrupting a stranger, you are following up on genuine interest.
Geographic and demographic reach
Where the audience is defined by where they are rather than what they searched, Meta is very strong. A development launching in a particular town, an event, a new site opening, a recruitment drive. The audience exists and is reachable, but it is not searching for anything.
Building a register of interest
Related, and underused. Collecting names before you have anything to sell is cheap on Meta because you are not competing with anyone for a search term that does not yet exist. This is why it works well ahead of a housing development launch, and the same logic applies to any launch with a fixed date.
Lookalike audiences from real customers
If you can upload a list of actual customers rather than all enquiries, the platform can find people who resemble them. The quality of this depends entirely on the quality of the list, which is a data problem rather than an advertising one.
Retargeting first, because it is the most reliable and it validates your tracking. Then audiences built from your own customer data. Only then broad prospecting, which is the hardest to make work and the easiest to waste money on.
Where it falls short
Intent, and therefore lead quality
Interrupting someone produces enquiries from people who are curious rather than committed. Lead forms make this worse: they are frictionless by design, which means a lot of submissions from people who barely registered what they were signing up to. If you use them, add a qualifying question and accept fewer, better leads.
B2B targeting is weaker than people expect
Meta knows a great deal about people’s interests and very little that is reliable about their job. Professional targeting options have been reduced over the years, and what remains is often inferred rather than declared. For narrow B2B audiences, this is a real constraint, and it is why broad B2B prospecting on Meta so often disappoints.
Attribution is genuinely harder
Because the click and the purchase are far apart in time and intent, Meta’s own reporting will usually claim more credit than a last-click model gives it, and the truth is somewhere between. This is not the platform being dishonest, it is two measurement models disagreeing about a real ambiguity. It does mean you should not take either number at face value, which is covered in our guide to tracking which marketing generates revenue.
Creative fatigue is fast
Search ads can run for months. Social creative burns out in weeks, because the same people see it repeatedly. Budget for a steady flow of new creative or accept declining performance. This is the cost people forget when comparing the two channels.
Not sure whether paid social is worth it for your business? We will tell you honestly. Sometimes the answer is that your budget belongs in search or in fixing your tracking first, and we would rather say so than take the work.
Creative does the heavy lifting now
Targeting used to be the craft. Automated audience selection has largely absorbed that, and the platform is now better than most humans at finding who to show an ad to, provided you give it a sensible objective and enough signal.
What is left, and what actually differentiates campaigns, is the creative.
- The first second decides everything. People scroll fast. If the opening frame does not stop them, nothing else in the ad matters.
- Make it look native, not like an advert. Polished corporate video routinely underperforms something that looks like it was filmed on site on a phone.
- Show the work. For trades, manufacturing, construction and property, footage of the actual thing being made or built outperforms any amount of brand messaging.
- Test few things, properly. Several genuinely different concepts beats twelve variations of the same headline. If the budget will not support that, run fewer.
- Write for sound off. Most of it is watched silently. Caption everything.
How to judge whether it is working
The mistake is to judge paid social by the same standard as search, conclude it is expensive, and switch it off, having paid for the awareness and never collected on it.
More useful measures:
- Retargeting separately from prospecting. They do different jobs. Blending them produces a number that describes neither.
- Watch branded search volume. If prospecting is working, more people should be searching your name over time. This is one of the clearest signals that awareness spend is landing.
- Ask new enquiries how they heard of you. Crude, and it will not reconcile with any platform, but it catches influence that no tracking model sees.
- Look at blended cost per customer. Total marketing cost over total new customers. If that improves while paid social runs and worsens when it stops, you have your answer regardless of what any dashboard claims.
Sensible expectations, in short: retargeting should pay for itself fairly quickly. Prospecting should be judged over quarters, not weeks, and mainly on whether the rest of your marketing is getting easier.
Common questions
Facebook or Instagram?
They are bought together and the platform will place your ads where they perform. Which one dominates depends on your audience rather than your preference, and it is not usually worth agonising over at the outset.
Is Facebook advertising still worth it for B2B?
For retargeting and for building familiarity in a defined geography, yes. For finding a narrow professional audience cold, it is a poor fit and search or LinkedIn will usually serve you better. Be honest about which of those you are trying to do.
How much do I need to spend to test it?
Enough to reach the same audience several times over a few weeks, which is a different sum for every business. Thin budgets spread across multiple audiences produce data you cannot learn from. The reasoning is set out in our piece on setting a paid media budget.
Should I boost posts instead?
Boosting is convenient and gives you almost no control over objective, placement or measurement. It is fine for local visibility. It is not a substitute for a campaign, and it is not where managed paid media starts.
