Most accounts I audit set an ad schedule so the campaign only runs 9 to 9, because night CPA looked ugly on a two-day chart. Delivery dies at 9pm. Learning never settles. Morning spend is a restart, not a continuation. They call it protecting budget. It is a daily off switch that never lets the ad set exit.

Do this:

1. Prospecting stays on 24 hours. Do not restrict hours to weekdays, evenings, or 9 to 9 to save money.

2. Read cost per purchase on a 7-day window, not 8pm vs 10pm. An hourly CPA is noise.

3. If nights look worse after a real sample, keep the ad on and kill on the full-day number. Do not schedule the winner off to hide eight hours.

4. If a schedule is already on, turn it off. Duplicate if the live ad set is still in learning. Judge the always-on version after a real sample.

Grab the free checklist: https://whop.com/checkout/ch_ZRoQxx2ylGqCmBW/

Most accounts I audit leave Advantage+ Shopping on existing customers because blended ROAS looks beautiful. Meta spends on people who already bought. New-customer CPA is empty. They scale the campaign because Ads Manager is green. Cash in the store is the same buyers. They call it a shopping campaign. It is a repurchase machine wearing a prospecting label.

Do this:

1. Set Advantage+ Shopping to new customers, or put a hard cap so existing-customer spend cannot take the budget. Do not leave it at 100% existing.

2. Exclude purchasers for 180 days on any row you call prospecting. Existing buyers belong in a separate retargeting budget.

3. Kill and scale on new-customer cost per purchase after a real sample. Ignore blended ROAS if the new-customer row is empty.

4. If the campaign is already dumping spend on existing customers, duplicate it with new-customer targeting. Leave the original alone until you have a real sample, then keep the version that buys new customers.

Grab the free checklist: https://whop.com/checkout/ch_ZRoQxx2ylGqCmBW/

Most accounts I audit stack 14 interests, then turn on detailed targeting expansion so delivery does not die. Meta spends outside the list. They think they tested dropshipping. They tested broad with extra clicks. CPA looks random. They add another interest. The expansion already threw the cage away.

Do this:

1. If the ad set has interests, expansion stays off. Do not unlock a stack you claim is the targeting.

2. If delivery is thin, kill the interests and go broad. Expansion is not a fix. It is broad wearing an interest label.

3. Kill on cost per purchase after a real sample. Do not keep expansion on because the graph spent.

4. If expansion is already on, duplicate the ad set with it off, or rebuild as broad with purchaser exclusion. Leave the original alone until you have a real sample, then keep the version that buys.

Grab the free checklist: https://whop.com/checkout/ch_ZRoQxx2ylGqCmBW/

Most accounts I audit leave iOS and Android in the same ad set because the platform mix looks simpler. Android inventory is cheaper. Meta dumps spend there. CPA looks fine on the blended row. iOS sits at a few dollars and never hits a kill. They call it one audience. It is an unread Apple auction hiding under cheap Android clicks.

Do this:

1. Split iOS and Android into two ad sets with their own budgets. Do not let one ad set pick the cheaper OS.

2. Each OS needs a clean $20 per ad. A $6 Android row is unread, not a winner. A starved iOS row is unread, not a failed platform.

3. Kill on cost per purchase after a real sample on that OS. Do not keep the mix because blended CPA looks cheaper.

4. If iOS and Android are already mixed, duplicate into two platform ad sets. Leave the original alone until you have a real sample, then keep the OS that still buys.

Grab the free checklist: https://whop.com/checkout/ch_ZRoQxx2ylGqCmBW/

Most accounts I audit send every ad to a clean product URL with no UTMs because the page looks cleaner. Ads Manager shows 40 purchases. Shopify shows 22. They cannot see which hook paid. They scale the campaign that looks green. Cash leaves. They call it a pixel issue. It is an untagged URL that cannot be reconciled.

Do this:

1. Every ad URL gets utmsource=facebook, utmmedium=paid, utmcampaign as the campaign name, and utmcontent as the hook name. No naked URLs.

2. Read Shopify orders by UTM, not only Ads Manager purchases. If Ads Manager is up and the UTM row is empty, that spend is unreadable.

3. Kill and scale on the Shopify UTM row after a real sample. Do not scale a campaign you cannot find in the store.

4. If live ads have no UTMs, duplicate them onto tagged URLs. Leave the original alone until you have a real sample, then keep the version you can actually attribute.

Grab the free checklist: https://whop.com/checkout/ch_ZRoQxx2ylGqCmBW/

Most accounts I audit switch a winning daily budget to lifetime with a Friday end date so Meta paces the leftover money. Meta dumps spend on day one. CPA spikes. Learning never settles. Friday the ad set dies even if it was buying. They call it controlling spend. It is a scheduled reset that turns a winner into a sprint.

Do this:

1. Winners stay on a daily budget with no end date. Do not switch them to lifetime to spend leftover money.

2. If you need a cap, lower the daily budget. Do not add an end date that turns the ad off.

3. Kill on cost per purchase after a real sample. A calendar date is not a kill rule.

4. If a lifetime end date is already on, duplicate the winner onto a daily budget with no end date. Leave the original alone until you have a real sample, then keep the version that still buys.

Grab the free checklist: https://whop.com/checkout/ch_ZRoQxx2ylGqCmBW/