How to increase ROAS: fix the page before the ad, and model the campaign before you spend

How to increase ROAS: raise the revenue each click earns (a better page and offer) or lower what each click costs (a better ad), in that order, because the page multiplies everything the ad does. ROAS, return on ad spend, is revenue from ads divided by ad spend. Before you change anything, find your break-even ROAS, because that number tells you whether a campaign is working at all.

Watch: Let's make your Shopify store unbeatable

The video above shows the page side of the story: a real Shopify store with weak ad returns, set beside a leader in its niche. Same kind of ads, different destination. That is the idea of this guide: the stores that win on ROAS are rarely the ones with cleverer ads, they are the ones whose page turns a click into an order more often than the competition does.

Want to see where your store stands before you spend? See your store's modeled 7-day ROAS range →

What is ROAS, and what ROAS do you need?

ROAS is revenue from ads divided by ad spend: as an invented example, $1,000 of ad spend that brings $3,000 of revenue is a ROAS of 3. The number alone says nothing about profit. What matters is your break-even ROAS. Per Shopify's break-even ROAS guide, gross profit margin is revenue minus variable costs (product, shipping, payment processing, pick and pack) divided by revenue, and break-even ROAS is 1 divided by that margin. Shopify's example: a $100 product with $60 of variable costs has a 40% margin, so break-even ROAS is 2.5: you need $2.50 of revenue per $1 of ads just to cover costs.

That is why there is no universal "good ROAS". A store with a 60% margin breaks even at about 1.7x; a store with a 20% margin needs 5x. Write your own number down before you read any benchmark, and measure every campaign against it.

How do you raise ROAS? Split it into three levers

Revenue per click is the conversion rate times the average order value; spend per click is the cost per click. So ROAS = conversion rate × order value ÷ cost per click. Three levers, and they are not equal in how much control you have.

An illustrative example (invented numbers, to show the arithmetic): with an average order value of $50 and a click cost of $1.00, a page that converts 1% of clicks returns $0.50 per $1 spent, a ROAS of 0.5. At 2% it is 1.0 and at 3% it is 1.5. Same ad, same budget, three times the return, because the page changed. Cutting the click cost from $1.00 to $0.75 at 1% would lift the ROAS from 0.5 to 0.67. Both help; the page lever is often the larger one and, unlike the auction, entirely in your hands.

Lever 1: the conversion rate (the page and the offer)

In our experience this is where many stores lose the most money. A stranger from an ad judges the page in seconds, and Nielsen Norman Group's dwell-time research found that users often leave pages within 10 to 20 seconds. Work through the checks in our 6-point readiness check: a clear offer, one action, proof near the buy button, imagery that fits the niche, an easy path to purchase and mobile speed. Speed has a clear target: Google's web.dev guidance treats an LCP of 2.5 seconds or less as good. Shipping surprises matter too: in Baymard's survey of why shoppers abandon, 40% cited extra costs such as shipping, tax or fees. For the full stage-by-stage plan see how to increase your Shopify conversion rate.

Lever 2: the average order value

Raising the basket raises revenue per click without changing the click. The common tools are bundles, a free-shipping threshold a little above your average order and complementary products near the cart. Test them one at a time and watch the conversion rate: in our reasoning, a threshold set too high can lower the rate by more than it raises the basket.

Lever 3: the cost per click (the ad)

Creative and audience set the click-through rate and the price of impressions. Test one thing at a time and give each test enough conversions to mean something. Meta documents a learning phase in its help center, so read it before you judge a campaign's first days or change it every time a number dips. If clicks are cheap but orders are rare, the problem is lever 1, not this one; our guide to why Meta ads aren't converting shows how to tell.

Lever 4: measurement you can trust

A ROAS you cannot trust is a guess. Install the Meta Pixel following Meta's setup guide and consider the Conversions API, which Meta describes as a connection between your marketing data and Meta that can run alongside the Pixel. Then compare the purchases Meta reports with your Shopify orders over the same period.

How do you model ROAS before you spend?

You can do a simple version by hand. Pick a daily budget, a click cost and a conversion rate for your page, and compute spend, clicks, orders, revenue and ROAS. Then repeat with a worse and a better conversion rate, and compare each to your break-even. If even the good case is below break-even, more budget will only lose money faster. It takes ten minutes and prevents the most expensive mistake in paid ads: funding a page that cannot yet convert.

A worked example with invented numbers. Budget $50 a day, click cost $1.00, average order value $40, gross margin 40%, so break-even ROAS is 2.5. That budget buys about 50 clicks a day.

Case	Conversion rate	Orders a day	Revenue	ROAS	Vs. break-even 2.5
Worse	1%	0.5	$20	0.4x	Loses money
Expected	2%	1	$40	0.8x	Loses money
Better	4%	2	$80	1.6x	Still below

Even the better case doesn't reach 2.5. At these inputs it takes a conversion rate of about 6.25% (2.5 × $1.00 click cost ÷ $40 order value). The model tells you before you spend that the problem isn't the budget: the conversion rate has to climb a long way, the order value has to rise, the click has to get cheaper or the margin has to improve. Changing one input at a time shows which lever gets you across the line most cheaply.

Why repeat purchases change the number you need

Break-even ROAS above is for the first order. If customers buy again, the lifetime profit per customer is higher than the first order's, and you can accept a lower first-order ROAS. That only holds if repeat purchases are real and measured: use the repeat rate you see in your own orders, not a hoped-for one, and keep the first-order break-even as your safety line until you have the data.

The hard part is making the three cases realistic: they depend on your niche, your country, the competition and, above all, how good your page really is. That is what EliteVault's modeler does from an audit of your store, and the stored example below shows what it returns.

The hard part is making the three cases realistic: they depend on your niche, your country, the competition and, above all, how good your page really is. That is what EliteVault's modeler does from an audit of your store, and the stored example below shows what it returns.

EliteVault in action

A projection that said: do not spend yet

One of the owner's own stores (home decor, name hidden) was audited on 2026-10-21: the verdict was not ready for paid traffic, with two blockers, a fake-looking countdown timer and no authentic social proof. Then we ran EliteVault's Meta Campaign Scenario Modeler with an average order value of $35 and a budget of $100 a day, for a physical product sold in the US with medium competition.

Scenario	Spend	Revenue	Sales	7-day ROAS (range)
Conservative	$420	$253	7	0.60x (0.48–0.72x)
Balanced	$700	$445	13	0.64x (0.48–0.80x)
Aggressive	$980	$635	18	0.65x (0.42–0.87x)

All three scenarios projected a loss: ROAS stayed below 1.0x, so every dollar of ads returned less than a dollar of revenue before any product cost. Spending more did not help: the aggressive scenario spends $980 to return $635. The conservative scenario's recommendation was to fix the trust problems and the product photography before raising the budget, which matches what the audit had already flagged. The balanced scenario's advice was different: it focused on testing ad creative. That mix is a reminder that the text around the numbers is guidance to check against your own data, not a verdict.

These numbers are modeled, not results. The model assumes days 1 to 3 convert worse (Meta's learning phase), that the conservative, balanced and aggressive scenarios spend 60%, 100% and 140% of the stated budget, and that a weak audit score caps the ROAS a store can reach. They are EliteVault's own modeling assumptions, so treat the output as a way to compare scenarios and see the break-even gap, not as a forecast.

The modeled range appears with every audit: See your store's modeled 7-day ROAS range →

What this means for you: the numbers and the audit's blockers pointed the same way, at the page. The honest next step after a free audit is to fix what the audit flags; if you want the three-scenario projection for that audit, a one-time $7 unlock includes one projection, and Pro ($19 a month) includes one projection every month. Scale ($29 a month) removes the limit and adds the Ads Optimizer, which proposes CPC, CPM, CTR and ROAS targets for your niche. The numbers in all of these are modeled estimates, not guarantees.

Where does ROAS leak between click and purchase?

What you see	Where the leak is
Low click-through rate, expensive clicks	The ad: creative or audience
Many link clicks, far fewer landing page views	Speed or tracking: people click and never see the page. Meta introduced landing page views to count visitors who actually arrive after a click
Landing page views, few add-to-carts	Offer, trust or product page
Add-to-carts, few checkouts	Surprise costs or a confusing cart
Checkouts, few purchases	Checkout friction or trust

When should you scale spend?

Scale what is above break-even, and only that. Scaling a campaign that sits below your break-even number scales the loss. Raise the budget in steps, watch whether the ROAS holds as spend grows, and keep fixing the page while you do: every point of conversion you add makes every future dollar of ads worth more. If you want to compare tools for the stack around this, see the best Shopify tools to increase ROAS.

Common mistakes when trying to raise ROAS

• Chasing a benchmark instead of your break-even. A 3x ROAS can lose money at a thin margin.
• Testing creative on a page that doesn't convert. You learn that nothing works.
• Judging a campaign in its first days. Read what Meta says about the learning phase first.
• Trusting an unchecked pixel. Reconcile reported purchases with your real orders.
• Raising budget before fixing the leak. Money into a leak leaves faster.

Make your page the one that earns more per click

The advertiser with the better page can bid more for the same click and still profit, which is the real advantage. Check how your store compares with the best one in your niche, fix the biggest gap first and model before you fund. Our Shopify store audit (ours — full disclosure) shows what a cold click sees and ranks what to fix; the Meta Ads scenario modeler page explains the forecast.

Run the audit on your store →