Meta Ads for Startups: What to Do in Your First 90 Days
Meta Ads for startups work best when you treat the first 90 days as a planned test rather than a launch. Here's how I'd spend that time and budget.
A common pattern with new businesses goes like this. The founder boosts a few posts, sees some likes and decides to run proper ads. They set £10 a day, optimise for purchases and launch. Within a week Ads Manager shows “Learning limited”, sales haven’t moved and the campaign gets rebuilt from scratch. Then the same thing happens again.
Nothing was wrong with Meta. The setup asked the platform to learn from data that didn’t exist yet. The first 90 days should be about creating that data in a planned way, so that by month three you know whether the channel works for your business.
Why Meta Ads work differently for a new business
An established business brings years of customers, website visitors and conversion data to Meta. A startup brings almost none, and the platform now leans heavily on that kind of signal.
Meta’s ad delivery system has changed a lot. Search Engine Land’s analysis of Andromeda explains how it now reads the creative itself to decide who should see each ad. In practice your ads do much of the targeting work that interest settings used to do.
The other thing to understand is the learning phase. Meta’s guidance is that an ad set needs roughly 50 optimisation events within seven days of its last significant edit to settle. If it can’t get there, Ads Manager labels it Learning limited. That isn’t a penalty. It means Meta has less data to work with, and it’s the most common reason startups decide Meta Ads have failed.
Your 90-day Meta Ads plan at a glance
Days 1 to 14: foundations
Set up ownership, tracking and consent properly, and choose an optimisation event you can actually reach.
Days 15 to 45: creative testing
Run broad targeting with several genuinely different ad angles and let each one gather data.
Days 46 to 90: decide and scale
Judge results on real customers, cut what isn't working and grow the budget in steps.
Days 1 to 14: get the foundations right
Own your accounts
Set up a Meta business portfolio in the company’s name and make sure you’re an admin of the ad account, the Facebook Page and the Instagram profile. Startups often lose access when the freelancer or friend who set things up moves on. It’s a ten-minute job now and a painful one later.
Pixel, Conversions API and consent
Install the Meta Pixel and connect the Conversions API, which sends events from your website’s server as well as the browser. Many website platforms and plugins handle this for you.
Check your cookie banner too. UK rules expect consent before tracking cookies are set, so make sure the banner works properly and your events still record once people accept.
Choose an event you can actually reach
Optimising for purchases makes sense when you’ll get plenty of them. For most startups you won’t, at first. Do the maths before you launch.
Say your budget is £20 a day and a lead costs £15. That’s roughly nine leads a week, well short of 50, so the ad set will sit in Learning limited. You could optimise for a more frequent step, such as landing page views or add to basket, and move to leads once volume builds.
These figures are made up to show the method, so use your own numbers.
If you’re new to Ads Manager itself, my getting-started guide to Meta Ads for small businesses covers the basics of campaign setup.
Days 15 to 45: test creative, not audiences
For most startups, Meta’s targeting works best kept broad. Leave Advantage+ audience switched on, set your location, and put your effort into the ads themselves.
Creative diversity means genuinely different ideas, not five versions of the same image in different colours. Search Engine Land makes the case that the variety of your messages matters more than sheer volume. Four distinct angles is a realistic start for a startup:
Leave it alone for a week
Every significant edit sends an ad set back into learning, including targeting changes, adding a new ad and large budget moves. Make changes in planned batches once a week rather than every time you open Ads Manager. Two bad days tell you very little.
Days 46 to 90: decide what's working and scale carefully
Judge on real customers
Ads Manager reports what Meta can see. Your business sees what actually happened. Keep a simple record of the leads or orders that came from Meta, which of them turned into paying customers and what each was worth. A campaign with a higher cost per lead can still be the better one if those leads buy.
Scale in steps
When something works, raise the budget gradually rather than doubling it overnight, because large budget changes also reset learning. Add new creative alongside the winners so you aren’t relying on a single ad when it starts to tire.
Add retargeting
By now you’ll have website visitors, video viewers and people who’ve engaged with your page. A small retargeting campaign aimed at them, with a different message from your cold ads, often brings some of the cheapest results in the account.
I can run your Meta Ads on a fixed monthly fee, or train you to manage them yourself on your own account.
When Meta Ads aren't the right first channel for a startup
Meta isn’t always where I’d start. If people already search for what you sell, such as a locksmith in Colchester or an accountant in Chelmsford, Google Ads will usually bring enquiries faster because the demand already exists. I’ve compared the two in Google Ads vs Meta Ads.
If you sell high-value services to other businesses, LinkedIn or Microsoft Ads may suit you better, which I cover in are LinkedIn Ads worth it for small businesses. And if your website can’t take an enquiry or payment smoothly yet, fix that before paying for traffic.
For a broader view of where to put a small budget, see my guide to the best digital marketing channels for startups.
Meta Ads for startups: FAQs
There’s no fixed minimum that guarantees results, but very small budgets struggle to give Meta enough data. Work backwards from what a lead or sale is likely to cost and aim for enough results each week to learn from. For many UK startups that means at least a few hundred pounds a month, kept in one campaign rather than spread thin.
Expect the first couple of weeks to be uneven while Meta learns who responds. Most startups can tell within six to eight weeks whether an offer and a few creative angles have potential. A full 90 days gives you a much fairer picture, especially if you’re optimising for leads or sales rather than clicks.
Often, yes. Advantage+ settings use broad targeting and automation, which suits a new business with no audience data of its own. The trade-off is less control, so the quality of your creative and your conversion tracking matters more. Check the breakdown of where your budget is going every week, and switch off anything that clearly isn’t working.
Pick the most valuable event you can realistically reach around 50 times a week, since Meta’s learning phase works best at that level. If purchases are too rare, start with leads, add to basket or another meaningful step, then move closer to the sale as volume builds. Avoid optimising for clicks or reach, which bring cheap traffic that rarely buys.
It depends on whether people already search for what you sell. Google Ads captures existing demand, so it suits services people look for, like an emergency plumber. Meta Ads creates demand by putting your product in front of people who aren’t searching yet, which suits new products, visual brands and local offers. Many startups use both once the budget allows.
Start your Meta Ads on the right footing
Book a call and we'll work out a realistic budget and creative plan for your first three months.
