If people already type what you sell into Google, start with Google Ads. You pay to appear at the moment someone is looking, and that intent is why search ads are usually the quicker of the two to pay for themselves. If hardly anyone searches for what you sell, but the right person would want it as soon as they saw it, start with Facebook and Instagram ads, which put your offer in front of people who were not looking yet. This article explains how to tell which group your business is in, and what changes once you can fund both.
The real difference: searching versus scrolling
Google Ads answers a request. Someone types “accountant for Self Assessment” in early January, with the 31 January deadline close, and your ad appears above the results while they are deciding whom to ring. The need already exists; you are competing to meet it.
Facebook and Instagram ads interrupt. The person is reading a friend’s post or watching Reels, and your ad appears between them. It has to earn attention within a second or two and then create the want. That is slower and less certain, but it reaches people Google never will, because they had not thought to search.
Neither is better in general. The costly mistake is asking one to do the other’s job: feed ads for an emergency repair service, or search ads for a product nobody knows to look for.
How Google Ads decides who sees your ad
You choose the searches you want to appear for, write ads for them and point each ad at a page on your site. For each search, Google runs an auction between the advertisers whose keywords match, weighing each bid against how relevant the ad and page look. On search campaigns you normally pay only for clicks.
The settings that matter most in a first campaign are the keywords and how loosely Google may match them. Keyword match types decide whether “kitchen fitter Leeds” can also trigger your ad for “kitchen fitter jobs” or “how to fit a kitchen yourself”, and a loose setting with no excluded terms is where many first budgets drain away. Check location too: by default Google can show ads to people who have shown interest in an area as well as those in it, which matters to a firm that only serves Greater London.
Google also sells YouTube, Display and Performance Max campaigns, some of which behave more like social ads. For a first campaign I would almost always start with Google search campaigns built around what your customers type, because that is where the buying intent sits and the reports are easiest to read.
How Facebook and Instagram ads decide who sees your ad
Facebook and Instagram are sold through one Meta ad account, and a single campaign can run on both. Instead of keywords, you give Meta a goal, such as purchases or enquiries, plus a location, a budget and your creative. Meta shows the ads to the people its system predicts are most likely to complete that goal, and adjusts as it sees who responds.
Two things follow. First, the creative does most of the targeting: the image, the video and the opening line decide who stops scrolling, and so who Meta looks for next. A business that cannot produce fresh creative every few weeks will find Meta harder going than Google.
Second, Meta has to see results to find more people like those who convert. That data comes from the Meta Pixel on your site, the Conversions API, or forms completed inside Meta’s apps. Without it, Meta optimises for what it can see, usually clicks or engagement, and the people who click and comment are often not the people who buy.
For service businesses, Facebook lead ads with an instant form are often the simplest start, because the enquiry happens inside the app. A form that easy to complete is also easy to complete half-heartedly, so leads need a fast reply and a qualifying question or two.
Side by side
| Question | Google Ads (search) | Facebook and Instagram ads |
|---|---|---|
| What triggers the ad | The words someone types | Who the person is and how they behave on Meta’s apps and on sites that share data with Meta |
| The viewer’s frame of mind | Looking for a solution now | Doing something else; your ad interrupts |
| What does most of the work | Keyword choice, ad text and the landing page | The image or video and the first line of copy |
| What you usually pay for | Clicks | Impressions, priced per thousand |
| Strongest at | Catching demand that already exists | Creating demand and reminding past visitors |
| Struggles when | Few people search for what you sell | The purchase starts with an urgent search, or the offer is hard to show |
| What you must keep supplying | New search terms to exclude and pages that match the searches | New creative, as audiences tire of the same ads |
When Google Ads is the better first channel
Start with Google if most of these describe your business:
- People search for what you sell, by name. “Emergency locksmith”, “MOT near me”, “employment solicitor”, “industrial shelving supplier”. If you can write down the searches your customers make, Google can put you in front of them.
- The need is urgent or tied to a date. A leaking roof, a tax return, a planning application, a replacement part. People in that position search; they do not wait for an ad to appear in their feed.
- The service is hard to show in a picture. Bookkeeping, IT support and conveyancing make dull Instagram posts, but people search for them every day.
- You serve a defined area. Search ads with tight location settings reach people in the postcodes you cover at the moment they need you.
- A customer is worth enough to pay for a serious buyer’s click. Clicks in competitive UK sectors can be expensive, but a click from someone ready to buy can still be the cheapest route to a customer.
When Facebook and Instagram ads are the better first channel
Start with Meta if these sound more like you:
- People do not know to look for it. A refill subscription for cleaning products, a new kind of fitness class, a board game you designed. Few people search for an unfamiliar idea, so you have to take it to them.
- It sells on sight. Clothing, food, homeware, a wedding venue, a treatment with a visible result. A good photograph or short video does most of the persuading.
- The decision is quick and the price modest. Someone can see it, want it and buy it from their phone in minutes.
- Your buyers share a place, stage of life or taste rather than a search term. Meta can find more of them once it learns from the first ones who convert.
- You are filling dates. Events, quiet weeknights at a restaurant and courses with a start date suit ads that reach many local people quickly.
Some businesses sit in both camps. A private dental practice gets searches for “emergency dentist” and also sells cosmetic treatments people weigh up over weeks. There I would start with whichever side has the better margin and clearer demand, and add the other later.
Which is cheaper
Comparing the price of a click across the two platforms tells you very little. Google search clicks often cost more, because the person clicking has asked for what you sell and every advertiser in the auction knows it. Meta usually charges per thousand impressions, so its cost per click is a by-product of how well the ad catches attention. A cheap click from someone idly scrolling has further to go before it becomes a customer, and often never gets there.
The number to compare is the cost of a sale or a qualified enquiry, measured against what a customer is worth to you after costs. If a new client brings £600 of profit over their first year, you can afford far more per enquiry than if they bring £40 once. A break-even ROAS calculator turns that into a target before you spend anything, and the UK cost per click and cost per lead benchmarks help you judge whether the target is realistic in your sector.
Both platforms also need data before they settle. Meta’s own guidance is that an ad set stays in its learning phase, with less stable results, until it records roughly 50 optimisation events in a week, and Google’s automated bidding learns from conversions in a similar way. A budget too small to produce a steady flow of conversions makes either channel look worse than it is, which is the best argument for funding one channel properly rather than half of each.
What UK privacy law changes on both platforms
Under PECR and UK GDPR, advertising tags such as the Meta Pixel and Google’s conversion tag need the visitor’s consent before they set cookies, so a UK site should load them only after someone accepts the cookie banner. Every visitor who declines is one whose enquiry the platform cannot link back to an ad through those cookies. This tends to hurt Meta more than Google search: Meta relies on conversion data to decide whom to show your ads to, while a search ad still reaches whoever typed the right words.
Two set-ups narrow the gap without cutting corners on consent. Google’s consent mode adjusts Google’s tags to the visitor’s choice and fills part of the gap with modelled conversions. Meta’s Conversions API sends events from your server rather than the browser, and should still respect that choice. Before you judge either channel, check your conversion tracking records real enquiries and sales. A channel can look like a failure when only the measurement has failed.
The rules on what an ad may claim are the same on both. The CAP Code, enforced by the Advertising Standards Authority, covers paid ads on Google and Meta alike, so a claim you could not print in a newspaper advertisement is no more acceptable in a Facebook ad.
Running both, in the right order
Once a business can fund both properly, they usually do more together than apart. For most service businesses, this is the order I recommend:
- Start where demand already exists. Usually Google search. Get the tracking right, cut the searches that waste money and find the cost per enquiry the account settles at.
- Remind the visitors who did not enquire. Most people who click a search ad leave without getting in touch. Facebook and Instagram retargeting shows them a review or an answer to a common worry over the following days. Under UK consent rules this audience is smaller than your visitor numbers, because only people who accepted cookies can be added.
- Then build new demand on Meta. With tracking proven and creative tested on people who already know you, campaigns for new audiences have something to learn from. Part of their effect often shows up as more searches for your business name on Google, so judge them across both channels, not by Meta’s report alone.
For a business selling something visual and quick to buy, the order flips: Meta does the main work, and a small search campaign catches people who saw the ad and later searched for the name. Either way, add the second channel once the first is tracked, profitable and near the most it can deliver, not as a hedge in month one.
Mistakes that waste a first ad budget
- Pressing “Boost” and calling it a test. Boosting a post buys reach and engagement with far fewer controls than a campaign built in Ads Manager. A boosted post that did nothing says little about what a properly built campaign would do.
- Sending paid clicks to the home page. A search for a specific service should land on a page about that service, with the form or phone number visible on a phone screen. The same goes for a Meta ad promising a particular offer.
- Judging each platform on the wrong number. Click-through rate, cost per click and reach are diagnostics. The verdict is cost per enquiry or sale against what a customer is worth.
- Stopping after a fortnight. The first weeks on either channel are partly spent finding which searches, audiences and ads to cut. Decide in advance how much spend or how many conversions you will see before judging it.
Where to go from here
If your customers search for what you sell, see how I run Google Ads for UK businesses. If they need to see it before they want it, the page on Facebook and Instagram ads services sets out the options. If you are still unsure, book a consultation and tell me what you sell, where, and what a customer is worth to you. I will tell you which channel I would start with, and why, before you put money into either.