How to Set an Ad Budget for Dropshipping in 2026 (Testing vs Scaling)

How much should you spend testing a dropshipping product?

This is the question every new dropshipper asks and most get wrong. Spend too little and your test data is statistical noise — you kill products that would have been winners given two more days. Spend too much and you burn cash on losers before you have enough data to know they are losers. The right testing budget sits in a narrow band that depends on your product margin, your conversion rate, and the platform you are advertising on.

This guide walks through the exact formula for setting a testing budget, the daily minimums for Facebook/TikTok/Google in 2026, the difference between testing budgets and scaling budgets, and how to know when to kill or scale a product. By the end, you will never again wonder "should I spend more on this campaign?" — you will know.

The testing budget formula

The right testing budget for a dropshipping product depends on three numbers: your max CPA (the most you can spend to acquire one order and still break even), your conversion rate, and the number of orders you need to make a confident decision. The formula:

Total Test Budget = Max CPA × Target Orders (10)
Daily Budget = Total Test Budget ÷ Testing Days (5-7)
Min Daily Budget = $20 (Facebook/Google) or $30 (TikTok)

Why 10 orders? Because below 10 orders, your data is dominated by random variance. A 0% CVR over 5 orders might mean your product is terrible — or it might mean you got unlucky with placement. At 10+ orders, you have enough signal to make a confident kill-or-scale decision.

Worked example: $39.99 product

Let's calculate the testing budget for our $39.99 LED lamp. You sell for $39.99 with $14.91 in non-ad costs (product + shipping + fees + other), so your max CPA is $25.08. You expect a 2% conversion rate. Testing period: 5 days. Math:

Max CPA: $25.08
Target Orders: 10
Total Test Budget: $25.08 × 10 = $250.80
Daily Budget: $250.80 ÷ 5 days = $50.16/day

So you should spend about $50/day for 5 days to get 10 orders of data. At $50/day and a 2% CVR, you would expect 2 orders/day — which gives you 10 orders in 5 days. The math is consistent. If your CVR assumption was 1% (half), you would need 10 days at $50/day to get 10 orders. If your CVR was 4% (double), you would get 10 orders in just 2.5 days.

Use our Ad Budget Calculator to run this math on your actual product in 30 seconds.

Minimum daily budgets by platform (2026)

Each ad platform has a "learning phase" — the period when its algorithm is figuring out who converts. If your daily budget is too low, the algorithm never exits the learning phase and your data is unreliable. Here are the practical minimums:

If your calculated daily budget from the formula is below the platform minimum, you have two options: extend your testing period (so total budget is the same but daily budget is lower) or raise your daily budget to the minimum and accept that you will get your 10 orders faster (which is usually the better choice).

Testing vs scaling budgets

Testing budget is what you spend to find out if a product works. Scaling budget is what you spend once you know it works. They are calculated differently:

Testing budget = Max CPA × 10 orders, spread over 5-7 days. Goal: get enough data to decide kill or scale.

Scaling budget = Start at the daily budget that produced your winning ROAS, increase by 20% every 2-3 days. Goal: maximize profitable volume without breaking the campaign.

The scaling rule is critical: never increase budget by more than 20% at a time, and never more frequently than every 48 hours. Larger or faster increases trigger the algorithm to re-enter the learning phase, which often kills the campaign's performance. We have seen 7-figure campaigns die because someone doubled the budget in one edit.

How to know when to kill or scale

After 5-7 days of testing, you should have 10+ orders of data. Here is the decision framework, using your BEROAS as the threshold:

• Actual ROAS ≥ BEROAS × 1.5: Strong winner. Scale immediately. Increase budget 20% every 2-3 days.
• BEROAS × 1.25 ≤ Actual ROAS < BEROAS × 1.5: Workable winner. Scale cautiously. Watch CVR and CPMs closely for 7 more days before aggressive scaling.
• BEROAS ≤ Actual ROAS < BEROAS × 1.25: Marginal. Test new creative before scaling. If new creative lifts ROAS above 1.5x BEROAS, scale. If not, kill.
• Actual ROAS < BEROAS: Loser. Kill the campaign. Do not "give it more time" — the math will not improve.

The biggest mistake new dropshippers make is the "give it more time" trap. If a campaign is below BEROAS after 5-7 days and 10+ orders, more time will not save it. The algorithm has already optimized for who it thinks will convert, and that audience is not converting profitably. Kill it and move on.

The 5-day vs 7-day vs 14-day test

How long should you test? It depends on your daily budget and conversion rate:

• 5-day test: Use when daily budget is $50+ and expected CVR is 2%+. You will get 10+ orders fast and can decide quickly.
• 7-day test: Use when daily budget is $30-$50 or CVR is 1-2%. Allows for weekday/weekend variance.
• 14-day test: Use when daily budget is under $30 or CVR is under 1%. You need more time to accumulate 10 orders. Be honest with yourself — if you cannot afford $30/day, you probably cannot afford to scale this product anyway.

Never run a test shorter than 5 days. Ad platforms have weekday/weekend variance that distorts 2-3 day data. A product might convert terribly Monday-Wednesday and great Friday-Sunday. You need a full week to see the real pattern.

What to do when you cannot afford the testing budget

If the calculated testing budget is more than you can afford, you have three options, in order of preference:

1. Pick a lower-CPC niche or platform. Pinterest and TikTok organic have lower CPMs than Facebook. Some niches (home decor, beauty) have lower CPCs than others (finance, B2B).
2. Start with organic traffic. Build a TikTok/Instagram presence first. Once you have organic traction, paid ads become much cheaper because retargeting converts at 2-3x cold traffic.
3. Save up and test properly. Testing a product with $5/day for 30 days gives you worse data than $50/day for 3 days. Save up, test properly, get clean data.

What you should not do is run ads with a budget so low that you never exit the learning phase. You will burn $150 over a month and have no usable data — that is the worst outcome.

Use the Ad Budget Calculator

Our free Ad Budget Calculator runs this math in 30 seconds. Enter your selling price, non-ad costs, expected CVR, and testing period — it returns your max CPA, total test budget, and recommended daily budget. Pair it with the Break-Even ROAS Calculator to know your kill/scale threshold, and the ROAS Calculator to track your actual campaign ROAS.