This is for the person who owns the business and also, at half past ten at night, runs the ads. A salon in Sheffield, a physiotherapy practice in Bristol, a two-van maintenance company covering three counties. No marketing team, no appetite for a retainer, and quite possibly one bad experience already with someone who went quiet after month two.
What a tool genuinely takes off you
Audience construction
Twenty minutes for an experienced buyer, two hours of second-guessing for you. A good tool reads the platform's live targeting catalogue and builds layered segments rather than one broad guess.
Keyword research
Real search volumes rather than terms that sound right. This is where self-run search campaigns usually fail, because the failure is silent.
Creative production
An ad image with your offer as readable text, or a short video, without a designer and without fighting Canva at midnight.
Campaign structure
Objectives, budgets, placements, ad set mechanics. Where most self-run campaigns quietly go wrong, because the errors produce no warning.
What it cannot do
It does not know why your customers choose you. It does not know that people come to you because you do evening appointments, or because you actually answer the phone, or because your quote includes the parts. Those are the things that make an ad work and they have to come from you.
💡 The single highest-leverage thing you will do is write the business description properly. Two honest sentences about who your customer is and why they pick you produce a better campaign than any amount of tool-switching. Most people type something generic, get a generic campaign, and blame the software.
It also does not know what a customer is worth to you — and that is the number that makes every other number meaningful. A £90 cost per lead is excellent if a client is worth £3,000 and catastrophic if an average sale is £40. No tool can supply that. You can, in about five minutes, and it will change every decision you make afterwards.
The floor you will hit first
Meta enforces a minimum daily budget per ad set, tiered by objective — a conversion objective sits at roughly three times the impression-objective floor, in your account's own currency. Set a budget below it and the campaign is refused with an error that does not explain the rule.
The practical consequence: testing two audiences against each other on a conversion objective needs double the per-ad-set floor. If your budget will not carry two properly funded ad sets, run one properly funded ad set. Two underfunded ones is the most common self-inflicted wound in small-budget advertising.
Search or social, for you specifically
People search for it
Repairs, emergencies, professional services, anything urgent. Google first. Intent does the work and small budgets go further.
People discover it
Restaurants, retail, beauty, fitness, events. Meta first. Nobody searches for a new nail salon; they see one.
Both
Most businesses eventually. Start with whichever matches above, add the other once the first is producing.
The compliance thing nobody warns you about
AI writes the exact copy patterns the CAP Code restricts — superlatives, invented statistics, outcome guarantees — because that is what persuasive advertising looks like in its training data. The ASA holds you responsible, publishes rulings with company names attached, and does not accept "the AI wrote it".
💡 One rule removes nearly all of this: no claim, figure or superlative goes into an ad unless you could produce the evidence within an hour. As a bonus, specific claims outperform superlatives anyway — "same-day appointments, 200 metres from the station" beats "the best in town" both legally and commercially.
Which is also why you want a tool that shows you everything before it publishes rather than launching automatically. Auto-publishing removes the only review step you had.
A sensible first three months
Month one
One platform, one campaign, one properly funded ad set above the floor. Write the description honestly. Do not touch it for two weeks.
Month two
Add a second audience or a second keyword theme so you have something to compare. Now you are learning which customer is cheaper to reach.
Month three
Add the other platform. By now you know your real cost per enquiry and can judge whether the second channel beats it.
Two timing facts so you do not panic in week one: a new click or conversion ad set typically takes four to five hours before its first impression, and three days of data means nothing. Reacting to noise makes it worse, because every edit restarts the platform's evaluation of that ad set.
Two questions before you pay anyone
- 1Does it launch into my own ad account? If the tool runs ads from an account it controls, your pixel history and audiences are not yours, and leaving costs you everything you built. Get the answer in writing.
- 2What happens when I stop paying? Campaigns in your own account keep running and the data stays. Campaigns living inside someone else's product do not.
Campaigns built from a short description of your business
Adyft builds and launches Google and Meta campaigns from a short description of your business, into your own ad account, and shows you everything first. Fourteen-day free trial.
Start free — no card needed