Ad strategy blueprint

The full ad blueprint: Meta, TikTok, Google and $0

Campaign structures, day-by-day budgets, the exact numbers at which you kill a creative, and copy-paste script templates. Written for a $50/day budget, because that is what most people actually have.

Prices verified August 202622 min readNo email required

TL;DR — Quick verdict

Updated August 202622 min read

Run one Meta Advantage+ Shopping campaign at $50/day with 4–6 creatives against a single product, and do not touch it for 72 hours. Judge on cost per add-to-cart on day 3 and on gross profit per order on day 5 — never on ROAS in isolation. If $200 of spend produces no add-to-carts, the product is the problem and no targeting change will fix it. TikTok is second, Google Shopping only works for products people already search by name, and the $0 route costs 15–20 hours a week for 4–10 weeks.

Traffic channels compared for a new dropshipping store. CPM ranges are US, outside Q4 — expect $30–$45 CPMs on Meta in November.
Channel	Realistic entry budget	Cost	Time to signal	Difficulty	Best for
Meta (FB + IG)	$50/day × 14 days	$12–$25 (US)	3–5 days	Medium	Impulse products under $80 that demo well on video.
TikTok Ads	$30–$50/day	$6–$12 (US)	3–7 days	Medium — creative-hungry	Visual, novel products for an under-35 audience.
Google Shopping	$20–$40/day	CPC $0.30–$1.20	7–14 days	High setup, low babysitting	Products people already search for by name.
Organic short-form	$0 + 15–20 hrs/week	Free	3–6 weeks	Hard to sustain	No ad budget, or building a brand alongside ads.

Traffic channels compared for a new dropshipping store. CPM ranges are US, outside Q4 — expect $30–$45 CPMs on Meta in November.

Meta Ads: the Advantage+ testing framework

Interest targeting stopped being the lever years ago. Meta's algorithm finds the buyer faster than your audience research does, so the whole game is now creative volume and a structure that does not fight the learning phase.

The structure is deliberately boring: one campaign, one ad set, broad targeting, several creatives. Complexity splits your budget across audiences that each get too little data to exit learning, which is how people burn $500 and learn nothing.

Every toggle, including the existing-customer budget cap people leave wrong, is listed inAdvantage+ Shopping settings. No budget at all?The organic plan is $150 and five videos a day, and on TikTok the format choice isSpark Ads against dark posts. On Google there is no ad copy at all —the feed title is the ad.

Campaign

Advantage+ Shopping Campaign (ASC)
Objective: Sales, optimise for Purchase
Budget at campaign level: $50/day
One product per campaign, always

Targeting

Broad. Country only, age 18–65+
No interests, no lookalikes under 1,000 buyers
Existing-customer budget cap: 0–10%
Advantage+ placements on, all of them

Creative

4–6 creatives, genuinely different hooks
Vertical 9:16 video, 15–30 seconds
Primary text under 125 characters
One clear CTA, no emoji walls

Where the $700 goes, day by day

This is the 14-day cycle a $750–$2,500 budget buys. Every row has a decision attached, because a budget plan without kill rules is just a spending schedule.

A 14-day, $700 Meta testing cycle at $50/day. Spend column is cumulative for that block.
When	Spend	What you do	What you judge on
Days 1–3	$150	Learning phase. Change nothing. Zero purchases here is normal.	Only check that delivery and pixel events fire.
Day 3	—	First real read: cost per add-to-cart and outbound CTR.	Kill creatives under 0.8% CTR. Keep the campaign running.
Days 4–5	$100	Let the winning creatives spend. Add 2 new hooks against the same product.	Cost per purchase vs break-even CPA.
Day 5	—	Decision point.	No add-to-carts after $150–$200 means the product or offer is wrong. Stop.
Days 6–10	$250	Winner keeps running. Test a second angle, not a second product.	Break-even or better on gross profit, not ROAS alone.
Days 11–14	$200	Duplicate the winning campaign at $100/day. Leave the original alone.	5 straight days above break-even before scaling further.

A 14-day, $700 Meta testing cycle at $50/day. Spend column is cumulative for that block.

Benchmarks and kill rules

Numbers for a US audience, e-commerce, outside Q4. They shift by country and season, but the decision logic does not: diagnose from the top of the funnel down, and stop at the first metric that fails.

The one people misread most is click-through rate, because Meta shows the flattering version by default — what CTR is good on Facebook ads covers outbound CTR and the 0.8% kill line.

Meta benchmarks for dropshipping, US, outside Q4. Read them top to bottom — the first failing row is the problem to fix.
Metric	Good	Acceptable	Kill it	What it actually tells you
CPM (US, outside Q4)	$12–$18	$18–$25	Above $30	Above $30 usually means a narrow audience or a creative nobody watches.
Outbound CTR	Above 1.5%	1.0–1.5%	Below 0.8%	A CTR problem is always a creative problem. Never fix it with targeting.
Cost per outbound click	Under $1.00	$1.00–$1.80	Above $2.50	Follows CTR. Fix the hook, not the bid.
Add-to-cart rate (of sessions)	Above 8%	5–8%	Below 3%	Low ATC with good CTR means the product page is lying about something.
Cost per add-to-cart	Under $2.50	$2.50–$5.00	Above $8	The first metric worth judging on day 3, before purchases exist.
Checkout completion	Above 45%	30–45%	Below 25%	Below 25% is shipping cost shock or a broken mobile checkout.
Store conversion rate	2.5–4%	1.5–2.5%	Below 1.2%	Under 1.2% on 500+ sessions: rebuild the page before you spend more.

Meta benchmarks for dropshipping, US, outside Q4. Read them top to bottom — the first failing row is the problem to fix.

Creative formulas that still work

Creative is 80% of performance now. The first three seconds decide whether the other 25 get watched, so write the hook first and the rest afterwards.

Four to six per product is the right count, andthe arithmetic behind that numberis simply your test budget divided by your break-even CPA. Film the first round yourself; bought UGC runs $80–$250 a clip.

The five hooks to test against every product

Problem hook

"If your back hurts by 3pm every day, this is why."

Names the pain in the viewer’s own words. Highest CTR on health and utility products.

Result hook

"This is what 30 days of wearing it looks like."

Shows the after-state immediately. Works when the change is visible on camera.

Unboxing hook

"Ordered this at 2am. Here is what showed up."

Native to the feed, low production cost, strong on novelty products.

Comparison hook

"$300 version on the left. $40 version on the right."

Justifies price without saying "cheap". Strong on tools and gadgets.

Objection hook

"I thought these were a scam too."

Opens with the doubt the viewer already has. Best on saturated products.

Problem–Agitate–Solution, timed for short-form

30-second UGC script template

0:00–0:03Problem"[SPECIFIC SITUATION] and nothing worked." Show the problem on camera. Face to camera, no logo, no music intro.

0:03–0:08Agitate"I tried [ALTERNATIVE 1] and [ALTERNATIVE 2]. [WHY THEY FAILED]." Keep it concrete — money wasted, time wasted, still in pain.

0:08–0:15Solution"Then I found [PRODUCT]." First clear shot of the product in use. This is the frame that has to look real, not staged.

0:15–0:24ProofDemonstrate the mechanism. "It works because [ONE SENTENCE]." Show the result, not a claim about the result.

0:24–0:30CTA"[OFFER] at the link." One offer, one link. No discount stacking, no countdown timer.

TikTok Ads: Spark Ads vs dark posts

TikTok CPMs run $6–$12 in the US, roughly half of Meta's, and the traffic converts worse. Net, it is usually a wash on profit and a win on volume — which makes it the right second channel once Meta is profitable, not the first channel.

Spark Ads versus standard in-feed ads on TikTok.
Format	What it is	Advantage	Cost	When to use
Spark Ads	Boosts a real organic post from your account or a creator’s, keeping likes, comments and the profile link.	Social proof carries over, comments build trust, cheaper CPMs.	Needs a creator authorisation code, and comments can turn against you.	Default choice. Use for anything with existing engagement.
Dark posts (non-Spark)	An ad-only video that never appears on a public profile.	Full control, no comment risk, test 8 variants without spamming a profile.	No social proof, and it looks like an ad within 1 second.	Volume creative testing, or claims you do not want publicly attached.

Spark Ads versus standard in-feed ads on TikTok.

Three UGC scripts that convert on TikTok

The POV

1POV: you finally fixed [PROBLEM] for $[PRICE].
2Hold the product up, unbroken shot, 2 seconds.
3Show it working. No cuts — cuts read as an ad.
4"Link in bio, they sold out last time." Only say that if it is true.

The three-reason

1"Three reasons I stopped buying [EXPENSIVE ALTERNATIVE]."
2Reason 1: price, on screen as text.
3Reason 2: the thing the expensive one does badly.
4Reason 3: the demo. End on the demo, not on a CTA card.

The reply

1Reply to a real comment with the comment on screen.
2"Someone asked if it works on [SPECIFIC CASE]."
3Answer honestly, including where it does not work.
4The honesty is the conversion mechanism. Do not fake the comment.

Google Shopping: harvesting demand that already exists

Meta and TikTok create demand. Google captures it. If nobody is typing a name for your product, Shopping has nothing to match and will spend your budget on adjacent queries.

When the search volume does exist, the economics are usually better than social: CPCs of $0.30–$1.20, conversion rates of 2–4% because intent is high, and campaigns that need far less daily attention than Meta.

Feed optimisation is the whole job

There is no ad copy in Shopping. Your product feed is the ad, and the title is 80% of it.

Product feed rules for Google Shopping. Get the title and category right before touching bids.
Feed field	Rule	Example	Why it matters
Product title	Brand + product type + key attribute + size/colour	Aurora Posture Corrector — Adjustable Back Brace, Unisex, Black, M	Google matches search terms against the title first. 70 characters, front-loaded.
Product type	Your own taxonomy, 3 levels deep	Health > Back Support > Posture Correctors	Drives bid grouping and gives Performance Max better context.
Google product category	The official taxonomy value, exact	Health & Beauty > Health Care > Supports & Braces	Wrong category is the most common reason for disapproved or throttled feeds.
Description	First 160 characters carry the keywords	Adjustable posture corrector for men and women. Discreet under clothing…	Truncated in most placements. Everything after 160 characters is for humans.
Image	White background, product fills 75–90% of frame	1200×1200 PNG, no text, no watermark, no lifestyle overlay	Text overlays get feed items disapproved. Lifestyle shots belong in the ad copy.

Product feed rules for Google Shopping. Get the title and category right before touching bids.

Standard Shopping first, Performance Max later

Start on a Standard Shopping campaign at $20–$40/day so you can see search terms and add negatives. Performance Max hides most of that, and handing a black box to a feed you have not cleaned means paying to learn the same lesson twice. Move to PMax after roughly 30 conversions, when there is enough data for target ROAS bidding to mean anything.

Negative keywords to add on day one

free
diy
how to make
used
second hand
repair
wholesale
near me
jobs
review reddit

Then check the search terms report every 3 days for the first month. Half your wasted spend hides in queries you would never have predicted.

The zero-budget organic plan

This is not the cheap version of ads. It is the slow version, and it costs 15–20 hours a week instead of $50 a day. Done properly it takes 4–10 weeks to the first sale, and the accounts that get there almost all look identical in week two: 200 views a post and no comments.

Week 1

Volume over quality

5 videos a day, one product, one account. You are training yourself, not the algorithm. Expect 150–400 views.

Week 2

Find the format

Keep the 2 formats with the best 3-second retention. Drop everything else. Still 5 a day.

Week 3

Iterate the winner

One format, new hooks daily. First video above 5,000 views usually lands here. Link in bio goes live.

Week 4+

Repeat and post more

A single video above 100k does more than a month of 500-view posts. Volume is what buys that lottery ticket.

Scaling without killing the winner

You have five consecutive days above break-even. Now is where most people destroy a working campaign by raising the budget 300% overnight and resetting the learning phase. The full sequence is in when to scale a winning product.

Duplicate, do not edit. Copy the campaign at double the budget and leave the original running untouched. If the duplicate underperforms, you still have the original, which is the entire point.

Expect ROAS to fall. Moving from $50 to $200 a day means buying more expensive audiences. A drop from 2.4 to 1.9 while daily profit rises is a successful scale. Watch profit per day, not the ratio.

Protect the account. At $200/day a restriction costs you $200 a day. Have a second Business Manager, a second payment method and a backup ad account warmed up before you need them. Set up the fallback while things are going well — nobody builds it during an outage.

Scaling checklist

5 consecutive days above break-even ROAS
Gross profit per order confirmed against real supplier invoices, not estimates
Supplier can handle 3x current volume without a stockout
Second Business Manager and payment method live
Duplicate at 2x budget, original untouched
Do not raise an existing budget more than 20% in a day

Ad questions, answered with numbers

How much should I spend per day testing a dropshipping product?

$50/day for 5 days is the standard test, so $250 to a first verdict. At a $18 US CPM that buys roughly 2,800 impressions a day, which is enough to read cost per add-to-cart by day 3. Below $25/day the data is too thin to distinguish a bad product from an unlucky week. If your total budget is under $400, run organic instead — an underfunded paid test tells you nothing you can act on.

What is a good ROAS for dropshipping?

The only ROAS that matters is your break-even ROAS, which is selling price divided by gross profit per unit. On a $49.95 product with $35.20 of gross profit that is 1.42, so anything above 1.42 makes money and anything below loses it. A "2.0 ROAS" is excellent on a $30 product and unprofitable on a $150 product with thin margins. Calculate yours before you launch.

How many creatives do I need to test one product?

Four to six for the first test, all against the same product with different hooks. Not five variations of the same hook — five genuinely different opening 3 seconds: problem, result, unboxing, comparison, objection. Meta needs variety to find the audience segment that responds, and creative is the only lever with real leverage left in the auction.

Spark Ads or dark posts on TikTok?

Spark Ads by default. They run on a real post, so likes, comments and shares carry into the ad and the CPM comes in lower because the platform treats it as native content. Use dark posts when you want to test eight creative variants quickly without flooding a public profile, or when the comments on a claim would be a liability. Spark needs a creator authorisation code, which takes about two minutes to generate.

Does Google Shopping work for dropshipping?

Only for products people already search for by name. Google Shopping captures existing demand — it does not create it, which is exactly the opposite of Meta and TikTok. If your product is a novelty nobody has a word for, Shopping will spend money on irrelevant queries. If it is a known category (posture corrector, dog nail grinder, car phone mount), CPCs of $0.30–$1.20 with 2–4% conversion often beat Meta on profit per dollar.

Can you dropship with no ad budget?

Yes, and it takes 15–20 hours a week for 4–10 weeks. The plan is 3 TikToks and 2 Reels a day on one product, one hook format per week, filmed on a phone. Judge on 3-second retention and average watch time, not on views. The failure mode is quitting in week 3 when the account is still doing 200 views a post, which is precisely when the algorithm has enough data to start testing you on bigger audiences.

My ad account got restricted. What now?

Request a review through Business Manager first — automated restrictions are reversed about half the time within 48 hours. While you wait, do not create a new account on the same payment method or device; that is what turns a temporary restriction into a permanent one. The structural fix is prevention: a second Business Manager, a second payment method, and no health, weight-loss or income claims in your copy, which is what triggers most restrictions in this niche.

How do I scale a winning product without killing it?

Duplicate the campaign at 2x budget rather than raising the original by more than 20% a day. A budget change resets the learning phase and you lose 2–3 days of performance. Scale only after 5 consecutive days above break-even, and expect ROAS to drop 15–30% as you move past the cheapest audience. If profit per day still climbs while ROAS falls, keep going — profit is the number that pays you, not the ratio.