Ad Automation Rules Every Store Should Set

Ad accounts do not sleep, but the people running them do. A campaign can overspend overnight, an ad can stop converting on a Saturday, or a product can sell out while ads keep sending traffic to it. Automation rules are simple if-this-then-that instructions that watch your campaigns and act, or alert you, when conditions are met. Set up well, they protect your budget and save hours of checking. Set up badly, they can switch off good campaigns or scale bad ones. This guide covers the rules most stores find useful, how to choose sensible thresholds and the mistakes to avoid.

Every threshold in this article is an example chosen for illustration. The right numbers for your store depend on your margins, prices, conversion rates and budget, and you should set them yourself.

How automation rules work

A rule has four parts:

• Scope: which campaigns, ad sets or ads it applies to.
• Condition: the metric and threshold that triggers it, such as cost per purchase above a value.
• Time window: the period of data it looks at, such as the last three days.
• Action: what happens when the condition is true, such as pause, change budget or send an alert.

The time window is the part people most often get wrong. Rules that look at very short periods react to random swings. Rules that look at long periods react slowly. Most stores use a mix.

Protective rules: stop losses early

Pause ads that spend without converting

Example rule you might choose: "If an ad has spent more than two times my target cost per purchase and has zero purchases, pause it." This catches ads that are clearly not working without waiting days. Base the multiplier on your own target, which in turn should come from your break-even ROAS.

Flag underperformers over a longer window

Example rule: "If an ad set's ROAS over the last seven days is below my break-even ROAS and spend is above a minimum I set, reduce its budget by 20 percent and send me an alert." Pairing an action with an alert means you see what changed and can step in.

Daily spend guardrails

Example rule: "If total daily spend across the account exceeds a cap I set, notify me immediately." Spend caps on the platforms themselves are also worth using, since they act at the source.

Growth rules: scale what works, gradually

Increase budget on proven winners

Example rule: "If ROAS over the last three days is above my target and the ad set has at least a minimum number of purchases, increase budget by 15 percent, no more than once every two days." The minimum purchase count prevents scaling something that got lucky with two orders. The frequency limit prevents compounding increases that can disrupt delivery. Large, sudden budget changes can push some campaign types back into a learning phase, so many advertisers prefer modest steps.

Set a ceiling

Example rule: "Never raise an ad set's daily budget above a maximum I set without my approval." Growth rules should always have an upper limit.

Creative and fatigue rules

Watch frequency

Example rule: "If an ad's frequency over seven days goes above a level I choose and its click-through rate drops by a set share compared with its first week, alert me to refresh the creative." Fatigue tends to build gradually, so alerts here are often more useful than automatic pauses.

Rotate in backups

If you keep approved backup creatives ready, a rule can switch them on when a lead ad is paused. This keeps the ad set running while you prepare new tests.

Operational rules that are easy to forget

• Out-of-stock protection. If your setup allows it, pause ads for products that have sold out, or at least get an alert when inventory drops below a level you choose. Paying for clicks to an unavailable product wastes money and frustrates customers.
• Tracking checks. An alert when purchases reported by the platform suddenly fall to zero while spend continues can point to a broken pixel or checkout rather than a bad ad.
• Disapproval alerts. Get notified when an ad is rejected so you can review it against the platform's current policy, rather than finding out days later.
• Weekly summaries. An automated report of what each rule did during the week helps you check that the rules behave as intended.

Mistakes to avoid

• Thresholds without data behind them. Copying someone else's numbers produces rules that do not fit your margins. Build them from your own break-even and target figures.
• Windows that are too short. A rule that reacts to a single bad hour will pause good ads. Use enough data for the decision.
• Conflicting rules. A rule that raises budget and another that cuts it can trigger in turn. Review rules together, not one at a time.
• No human review. Rules carry out instructions. They do not understand a holiday sale, a supplier delay or a tracking outage. Check what they did regularly.
• Forgetting attribution delays. Some conversions are reported hours or days after the click. Very recent data may look worse than it will end up, so be cautious with same-day rules.

Running rules with Novaris

Novaris is an AI growth platform for e-commerce brands, built with our partner Atiko. Its automations run around the clock on the Growth plan, with custom automations on Scale: pausing underperformers, scaling winners within the caps you set, rotating creative and sending alerts. You choose every threshold, and the ad platforms still control delivery and review. Results depend on your products, market, budget and ad accounts. Related controls for spreading spend are described on our budget optimization page.