Google Ads Bidding Strategies for Dropshipping: The Full Breakdown

Dropshipping stores blow their bidding one of two ways. They slap a tight ROAS target on a day-old campaign and watch it starve. Or they never set a target at all, spend goes flat, and profit leaks out the bottom. Both come from treating bidding as a switch you flip once, instead of a strategy that changes as your data grows.

The short version that fixes it: start every new campaign on Maximize Conversions with no target. Let it run to roughly 30 conversions. Then switch to target ROAS, or target CPA if your margins are flat across products. Raise budget in steps of 20 to 30%, never bigger, or the campaign drops back into learning. Everything below is the reasoning, so you know why it works and when to break the rule.

The two phases every campaign goes through

Every Google Ads campaign moves through two phases, whether you plan for them or not.

Phase one: data gathering. The account has little or no conversion history. Google's algorithm knows almost nothing about who buys from you. Your only job here is volume: get conversions flowing so the system has something to learn from.

Phase two: profit. Once the algorithm has enough real conversions to work with, your job flips. Now you protect margin. You tell Google exactly what a good result costs, and it optimizes toward that number instead of just chasing raw volume.

Most bidding mistakes come from running phase-two tactics during phase one, or never leaving phase one at all. The full dropshipping Google Ads guide covers the account structure underneath this. This post is just about the bid strategy layer sitting on top of it.

Maximize Conversions vs Maximize Conversion Value

These are the two no-target bidding strategies, and they optimize for different things.

Maximize Conversions tells Google: get me as many conversions as possible with this budget. It treats every sale the same. A $20 order and a $200 order count equally.

Maximize Conversion Value tells Google: get me the most total revenue possible with this budget. A $200 order is worth 10x a $20 order, and the algorithm bids accordingly.

For a dropshipping store selling a mix of cheap and expensive products, Maximize Conversion Value usually wins once you are out of the data-gathering phase, because it naturally leans spend toward your higher-ticket, higher-margin winners. But early on, when volume is what you need, plain Maximize Conversions often gets you to 30 conversions faster, because it is not filtering as hard on value.

Target ROAS and target CPA: adding a number to the bid

Target CPA (tCPA) is the same idea, but for cost instead of revenue. You tell Google the exact dollar amount you are willing to pay for one conversion, and it bids to hit that average.

The choice between the two comes down to your catalog. If your products sell at wildly different prices with different margins, tROAS is the better fit, because it scales the target with order value. If every product in the campaign sells at roughly the same price and margin, tCPA is simpler and does the same job.

The four core strategies stack up like this.

	Max Conversions	Max Conv Value	Target CPA	Target ROAS
Best for	New campaigns, building data fast	Mixed-price catalogs once data exists	Flat-margin, same-price catalogs	Mixed-margin catalogs protecting profit
Needs a target?	No	No	Yes, a dollar cost per order	Yes, a revenue multiple
Conversion data needed	Works from day one	Some history helps	~30 conversions	~30 conversions
Optimizes for	Conversion count	Total revenue	Cost control	Margin control
When to avoid	Once you know your break-even number	Before you have any conversion history	Wide range of product prices	Under 30 conversions in the window

Break-even ROAS decides which target you set

Whichever target you pick, it has to sit on top of one number: your break-even ROAS.

Break-even ROAS = 1 divided by your margin.

Say your margin, after product cost, shipping, and fees, is 25%. Break-even ROAS is 1 / 0.25 = 4. That means you need $4 back for every $1 spent just to cover costs. A 3x ROAS on that same 25% margin loses money, even though 3x sounds like a good number on a dashboard. This is why raw ROAS means nothing without knowing the margin behind it.

Once you know break-even, set your target ROAS a bit above it, not right on the line. A target sitting exactly at break-even aims you at zero profit by design. Most stores set targets 15 to 25% above break-even to bank real money on top. Track this alongside the core metrics every dropshipper should watch so a good ROAS number and a good profit number never quietly drift apart.

When to start with no target, and when to add one

Start every new campaign with no target. That means Maximize Conversions or Maximize Conversion Value, target field left blank.

Add a target once two things are both true:

1. You have got roughly 30 conversions in the trailing 30 days.
2. You know your break-even ROAS or CPA, so the number you type in actually means something.

If you add a target before either condition is met, you are asking Smart Bidding to hit a number it does not have enough data to reach. It usually responds by cutting spend hard, because the safest way to hit an unrealistic target is to barely spend at all. The single most common self-inflicted wound we see on new accounts is a day-one ROAS target that never let the campaign breathe.

The Google Ads learning phase, explained

The learning phase is the period right after you launch, or meaningfully change, a Smart Bidding campaign. During this window, the algorithm is testing auctions, gathering signal, and cost per result can jump around a lot before it settles.

The learning phase is mostly about time and data. It usually runs one to two weeks, and Google's own guidance is that Smart Bidding wants somewhere around 30 to 50 conversions a month to bid well. For dropshipping accounts, treat 30 conversions in a rolling 30-day window as your real threshold before you trust a target's performance. Judging a campaign on day 3 of learning is like judging a new hire on their first hour.

What resets or restarts the learning phase:

• Changing the budget by more than roughly 30% in one move
• Switching the bid strategy, or editing an existing target
• Editing ad copy or creative inside a live ad group
• Adding or removing a large share of keywords or products at once
• Pausing a campaign and reactivating it later

Any one of these can send the campaign back into a mini learning window. Stack two or three of them in the same week, and you can watch a stable campaign go noisy again for no obvious reason. That is usually not the campaign failing, it is your own last edit still working its way through.

Portfolio bidding across campaigns

Portfolio bidding groups multiple campaigns under one shared bid strategy instead of giving each campaign its own target. Google pools the conversion data across all of them and optimizes toward one combined ROAS or CPA goal.

This works well once you are running three or more campaigns that share the same margin profile, so the break-even math is the same across all of them. Three is roughly where pooling gives the algorithm meaningfully more data than each campaign holds alone. A smaller campaign that has not hit 30 conversions on its own can then lean on the learning happening in a bigger sibling.

Portfolio bidding works badly when campaigns have different margins. A 60%-margin bestseller campaign and a 20%-margin clearance campaign under one portfolio target will fight each other, because one needs a low ROAS target to spend well and the other needs a high one. Split those into separate portfolios, or keep them on individual strategies, and never blend them into one number.

If you are still running everything manually, manual Shopping campaigns are worth reading before you jump to portfolio bidding, since the manual-to-Smart-Bidding switch is its own decision with its own timing.

The budget ladder: raising spend without restarting learning

Once a campaign is stable and hitting target, you will want to spend more. Do it in steps, never in one jump.

1

Step 1: Confirm the campaign is out of learning

Check that you have had at least 30 conversions in the trailing 30 days and the target ROAS or CPA has held steady for at least a week. Do not touch budget on a campaign still finding its feet.

2

Step 2: Raise budget by 20-30%, not more

A 20-30% increase is small enough that Google can absorb it without treating it as a new campaign. Go bigger, and you risk a full or partial learning reset, right when things were finally working.

3

Step 3: Hold for 7-14 days before the next raise

Give the new budget level enough time to prove itself. A week of stable ROAS at the new spend is your green light. A dip that does not recover in two weeks is your signal to hold or step back down.

4

Step 4: Repeat, and watch break-even the whole way up

Keep laddering up 20-30% at a time as long as ROAS stays above your break-even-plus-margin target. The moment ROAS creeps toward break-even as spend rises, that is the ceiling for this campaign at its current structure, not a signal to push harder.

This ladder is also exactly how we recommend planning a dropshipping ad budget from the start, so the two guides work as one system: budget in steps, target set once the data earns it.

The switch sequence in one place

Put it all together, and the order of operations for any new dropshipping campaign looks like this:

1. Launch on Maximize Conversions, or Maximize Conversion Value if you already have store-wide conversion history, no target.
2. Track conversions daily. Do not touch bids while you wait.
3. At roughly 30 conversions in a rolling 30-day window, calculate break-even ROAS as 1 / margin.
4. Add a target ROAS 15-25% above break-even, or a target CPA under your per-order profit ceiling.
5. Hold for a full learning cycle (7-14 days) before judging the target.
6. Raise budget 20-30% at a time once performance is stable, never more.
7. Once you run 3+ campaigns with matching margins and enough shared volume, consider portfolio bidding to pool their data.

Get your exact break-even number before you set anything. Run it through the ROAS calculator and set your target from real math, not a guess.

Frequently asked questions

What is the best Google Ads bidding strategy for dropshipping?

There is no single best one. Start new campaigns on Maximize Conversions with no target so the algorithm can gather data fast. Once you pass 30 conversions and know your break-even ROAS, switch to target ROAS or target CPA to protect margin. The right strategy changes as your data grows.

How many conversions do you need to exit the Google Ads learning phase?

There is no fixed count. The learning phase is mostly time-based, usually one to two weeks, and Google's guidance is that Smart Bidding wants around 30 to 50 conversions a month to work well. Treat 30 in a rolling 30-day window as your practical threshold. Below that, judge the campaign on trend, not any single day's number.

Should I use target ROAS or target CPA for a dropshipping store?

Use target ROAS when your products have different prices or margins, because ROAS scales with order value. Use target CPA when every product sells at close to the same price and margin, so a flat cost-per-order target makes sense. Most dropshipping catalogs are mixed-margin, so target ROAS wins more often.

What resets the Google Ads learning phase?

Five things reset it: a budget change over about 30%, editing the bid strategy or its target, new ad copy or creative in an ad group, adding or removing a big chunk of keywords or products, and pausing then reactivating the campaign.

Why does a new dropshipping campaign fail when I add a tight ROAS target on day one?

Smart Bidding needs failure data to learn, not just wins. A tight target from hour one tells the algorithm to only chase the safest, narrowest slice of traffic. With zero conversion history, it has nothing to calibrate against, so it either spends almost nothing or misses your real audience entirely.