Running your first ads on a small budget
A small budget can’t absorb guesswork, but it can buy clear answers when each dollar tests one idea and every result is tracked. Follow these steps in order, from setting the goal to deciding what deserves more money.
Set one goal and picture one buyer
Pick a single result for the campaign: online sales, phone calls, form leads, store visits or app installs. Ad platforms optimize toward the goal you select, and a campaign chasing awareness and sales at once tends to do neither well.
Describe the person most likely to buy: where they live, the problem they’re solving and the words they would search for. Then work out the most you can pay to win a customer. If an average order earns $60 in gross profit, a cost per acquisition above $60 loses money unless customers come back to buy again.
Match channels to existing demand
Search ads capture demand that already exists: someone types a need into Google or Bing, and your ad answers it. Social ads create demand by showing something new to people who weren’t looking, which suits products that are easy to show in a photo or a short video. Start with one or two channels so each gets enough budget to learn from. Other channels worth knowing:
- Display and retargeting: banner ads across websites and apps, plus follow-up ads for people who visited your site, which need a tracking tag your privacy policy should disclose.
- Online video and connected TV: ads on video sites and streaming services, better for building familiarity than for immediate clicks.
- Local print, radio and out-of-home: newspapers, local stations, billboards and transit posters, best tracked with a unique offer code or short web address.
- Marketplace ads: sponsored listings on sites such as Amazon or Etsy, which reach shoppers already searching inside the store.
Build the creative around one offer
Each ad should make one offer and ask for one action, such as shopping a sale, requesting a sample or calling for an estimate. As a rule of thumb, the first two or three seconds of a video decide whether people keep watching, so open with the product, the problem or the result rather than a logo. Add captions, because many people scroll with the sound off.
Produce each concept in the shapes the placements use, instead of letting a platform crop it badly:
- 1:1 square, which fits most feeds.
- 4:5 vertical, which fills more of a phone screen in a feed.
- 9:16 full-screen vertical, for stories, reels and short-form video.
- 16:9 horizontal, for standard video players and connected TV.
Send clicks to a page that keeps the promise
The landing page should repeat the ad’s offer in its headline, show the same product and price, and make the next step obvious. Sending paid clicks to a general home page makes visitors hunt for what they were promised, and many won’t bother.
Speed matters as much as the message. Much of your traffic will arrive on phones, so test the page on a phone over a cellular connection, compress large images and cut anything that delays the first screen. Keep forms short, since every extra field costs some responses.
Check the rules before anything runs
In the United States, the FTC requires a clear disclosure whenever an endorser or influencer has a material connection to the brand, such as payment, free products, a discount or a family or business tie. The disclosure must be easy to notice in the post or video itself, and the FTC advises against relying on a platform’s built-in label alone. Fake reviews, and reviews paid for on condition that they’re positive, can also break FTC rules.
Objective claims, such as “lasts twice as long” or “clinically proven,” need evidence in hand before the ad runs, and health claims call for competent and reliable scientific evidence. Platforms add their own policies on top: they restrict categories such as alcohol, supplements and financial products, and they limit targeting for housing, employment and credit ads.
Launch small and test one variable at a time
Before spending anything, tag every ad link with UTM parameters, the short labels such as utm_source, utm_medium and utm_campaign that tell your analytics tool where each visit came from. Set up conversion tracking too, so the platform learns which clicks turn into sales or leads.
Then run a modest test of two versions that differ in one thing only: the headline, the image, the audience or the offer. Change two things at once and you won’t know which one made the difference. As a rule of thumb, let a test run at least a week so weekday and weekend behavior both show up.
Measure honestly, then scale what works
Four numbers tell most of the story. Click-through rate is clicks divided by impressions; conversion rate is conversions divided by clicks; cost per acquisition is spend divided by conversions; return on ad spend is ad-driven revenue divided by spend. For example, $300 that buys 1,500 clicks and 30 sales means a 2 percent conversion rate and a $10 cost per acquisition, and $900 in resulting sales means a return on ad spend of 3.
Treat platform dashboards as estimates. Each platform credits itself under its own attribution rules, two platforms can claim the same sale, and privacy settings hide some conversions, so check the numbers against your actual orders. When a version wins, raise its budget in steps rather than all at once, watch whether cost per acquisition climbs as you reach less eager buyers, and refresh the creative when results fade.