Shopify Marketing Budget: Where to Spend First

Quick answer: Work the budget out from your margin, not from a percentage of revenue. Take what one order leaves after product, shipping, fees and returns, subtract your fixed costs and the profit you want to keep, and what remains is the most your store can afford to spend on marketing at today’s revenue. Then spend in this order: tracking you can trust, conversion on the store, retention and organic, and paid ads sized to what Meta and Google need to give you readable results. The percentages published online disagree with each other, so I would not use any of them as your target.

Search for how much to spend on marketing for a Shopify store and the answers are percentages of revenue. Which percentage depends on the page. Even Shopify’s own pages give different figures, and the 2026 Gartner survey behind one of them is mostly companies far bigger than yours.

This guide starts from the money each order leaves behind, sets a ceiling, checks it against how long you can wait to be paid back, and then divides the result between five jobs in a sensible order. It is written for a founder with a store turning over about $150,000 to $1 million a year, the range I work with.

Prism3 does not buy media or run email. We do Shopify SEO, GEO and CRO. We do not run ads, email or SMS, and nothing here is a pitch for ad management. Three of the five jobs below (tracking, conversion and organic) overlap with what we do, so read my ordering with that in mind and check the arithmetic yourself.

Every dollar figure for the example store is illustrative. The store does not exist, and the numbers are not benchmarks, targets or results from any store. Figures from a published page are named with their source, and I read each of those pages on 31 October 2026. Money is in Australian dollars, with GST taken out.

Why the percentages you find online do not agree

The usual advice is a share of revenue. Here is what I found when I opened the sources myself.

Shopify attributes two of those three figures to Gartner. I opened the 2026 Gartner release: it describes a survey where the vast majority of respondents reported annual revenue over $1 billion, so a store turning over A$400,000 is very unlikely to be represented. I did not open the 2022 release. The 2% to 8% range has no named source on the page. None of them is a target for you. At most they are a rough sanity check after you have done your own sums.

Splits by channel are less reliable still. Hostinger’s Australian guide to ecommerce marketing budgets, last modified on 10 March 2026, is one example. It splits a $4,000 social media ad budget across five platforms at 19%, 18%, 11%, 10% and 7%. Those add to 65%, or $2,600 of the $4,000. The page says the split is based on HubSpot research, which I did not open, so I cannot say where the missing 35% went. A split that does not add up is not one to copy.

A percentage of revenue also ignores margin, and margin decides what you can afford.

A margin-first method in five steps

The example store sells in Australia, turns over A$400,000 a year net of GST, and takes 4,000 orders at an average of A$100 each. All of it is made up.

Step 1: Find what one order leaves

Use the layer I call CM2: net order value, minus product cost, minus what you pay to ship and pack it, minus payment fees, minus an allowance for returns. It is measured before any ad cost. The contribution margin worked example takes one order through every line, including GST and Shopify’s card fees.

That is 54% of net order value. If you cannot fill this table in from your own records, do that first. Everything below depends on it.

Step 2: Set the margin ceiling

The ceiling is the most marketing your store can carry at today’s revenue, after the other bills and your profit.

• CM2 across the year: 4,000 orders × A$54 = A$216,000
• Less fixed costs that are not marketing (wages, rent, software, platform): A$128,000
• Less the profit you want to keep: A$40,000
• Room for all marketing: A$48,000 a year, or A$4,000 a month

That A$4,000 is 12% of revenue, and it has to cover every marketing cost: ad spend, creative, email and analytics tools, freelancers and agency fees. Now change only the margin.

The same A$400,000 of sales gives a ceiling from below zero to 23% of revenue. No single percentage can hold that spread, which is why I do not give you one.

The ceiling is for today’s revenue. You can sensibly spend above it while marketing grows revenue, but that is a decision made with measured payback, not a default. And if your number is negative, as in the first row, the fix is not a smaller ad budget. It is price, shipping, discounting, product cost or fixed costs.

Step 3: Set the payback rule

The ceiling says how much the year can carry. Payback says how long you are willing to wait for each customer to repay what you paid to win them. That is a cash decision, and only you can make it.

On the first order alone, the example store can pay up to A$54 for a customer and break even on that order. Now suppose you will wait 90 days, and you assume (illustratively) that 20% of new customers order again inside that window at the same A$54 of contribution. Each new customer is then worth A$54 + (0.2 × A$54) = A$64.80 by day 90. That is your maximum cost per new customer under that rule.

Two ways to use it:

• Break-even ROAS on the first order is 1 ÷ 0.54 = 1.85 when ROAS is measured on revenue with GST taken out. If your ad platform records GST-inclusive order values, the figure differs. The break-even ROAS guide shows how to match the two.
• Customer acquisition cost (CAC) above A$64.80 means the store is lending money to customers for longer than you decided. The CAC payback guide covers measuring payback from your own cohorts instead of an assumed 20%.

Step 4: Find the floor

A budget can also be too small to read, because platforms need results to learn from and you need results to judge.

Meta’s help centre says an ad set usually leaves the learning phase after about 50 results in the week after its last significant edit, that ad sets in learning are less stable and usually have a higher cost per result, and that you should use realistic budgets. On minimum budgets, Meta says that with the cost per result goal bid strategy your daily budget should be at least five times the goal, and that minimums may vary by country and objective. For Google, the Target ROAS page says Search and Shopping campaigns need at least 15 conversions in the past 30 days to use that strategy.

Put illustrative numbers on that. Say a purchase costs A$45 on Meta, and you set a cost per result goal of A$45:

• 50 purchases a week costs A$2,250 a week, about A$321 a day, or about A$9,770 across 30.4 days.
• Five times the goal is A$225 a day, about A$6,840 a month.
• 15 purchases at A$45 is A$675 a month, just to be eligible for Target ROAS on Google.

Compare that with the ceiling. The whole marketing room is A$4,000 a month, or about A$132 a day, so one Meta ad set optimised for purchases at those numbers would need more than the entire budget. That is the honest result for a store this size, and it is why “put most of it into ads” is not automatically right.

If the floor is higher than the money you can put into paid, you have three options:

1. Run fewer ad sets and make fewer edits. Meta says combining similar ad sets also combines their learning, and that frequent budget changes can send an ad set back into learning. It gives no percentage for how big a change counts, so plan changes instead of nudging the budget daily.
2. Optimise for an earlier event. Meta’s minimum-budget page says that if you do not want to raise the budget, you can try an event that is easier and cheaper to achieve than a purchase. My caution: you may get people who add to cart and never buy, so judge it on purchases in Shopify.
3. Accept a longer, noisier test. Judge paid on your blended numbers over a longer window, not on one ad set’s daily reading. ROAS, MER and blended CAC explains why.

Step 5: Choose the number

Your monthly budget sits between the floor for what you want to learn and the ceiling the margin allows. If the floor is above the ceiling, you have two honest choices: lift the ceiling by fixing margin or conversion, or run paid as a small, capped test and spend the rest elsewhere. Shopify’s help centre tells merchants to set a budget before choosing channels; the method above is one way to set it.

Where to spend first: five jobs, in order

Now divide the A$4,000. The shares come out of this store’s numbers and are not a recommendation for yours.

Check: 5 + 20 + 15 + 20 + 40 = 100, and 200 + 800 + 600 + 800 + 1,600 = 4,000. Add up any split you copy.

Treat the table as steady state. In month one, tracking and conversion fixes are lumpy, so pay for them out of the paid line, not on top of the budget.

1. Tracking and measurement: first, and usually the cheapest

Every other line is read through this one. If Shopify, GA4 and Meta disagree and you cannot say why, you cannot tell whether paid is working or a campaign is double counting. The growth order post sets out the reconciliation test I use.

One dated change to know about. Shopify says its Analytics session measurement rollout ran from 21 to 23 September 2026: identified bot sessions are now filtered out by default, some sessions without a page view now count, and sessions no longer end at midnight UTC. Orders, sales and customer counts are not affected, but conversion rate is calculated from sessions, so it can move even when orders do not. Shopify says to treat the update as a new baseline. A conversion rate from August and one from October are two different measurements, so use orders and sales as the check.

Decision rule: spend until you can explain every gap between Shopify’s orders and what GA4 and Meta report. Then stop, and re-test whenever you change theme, apps or checkout.

2. Conversion: the multiplier on everything else

Your cost per customer is partly a conversion question. If a paid visit costs A$1.20 (illustrative), then at a 2.0% conversion rate each order costs A$60 in traffic, and at 2.5% it costs A$48. Same traffic, same price per visit.

The store-wide version, again illustrative. About 333 orders a month at 2.0% means about 16,670 sessions. At 2.4% on the same sessions, orders would be about 400. Those 67 extra orders at A$54 of CM2 are about A$3,600 a month, close to the whole marketing room. That is arithmetic, not a forecast. We do not promise conversion results, and some stores have no leak worth fixing.

Conversion is a line Prism3 works on, so here is the test I would apply with no agency involved. Find the biggest leak with session recordings and your own funnel numbers, then ask whether fixing it costs less than the extra traffic you would buy for the same orders. I have no sourced benchmark for a “good” conversion rate, so compare against your own history.

Decision rule: fix the largest known leak before you raise paid spend, as long as the fix costs less than the traffic needed for the same orders.

3. Retention: the cheapest orders you will get

A repeat order that needs no ad leaves its full A$54 before retention costs (platform fees, any incentive). A first order won at a A$60 customer cost leaves A$54 − A$60 = −A$6. That gap is why retention sits above paid in the order. How email lifts customer lifetime value goes into the mechanics.

We do not run email or SMS, so this is the line I know least about operationally. In any store I would check that welcome, abandoned-cart and post-purchase flows exist and work before spending on campaigns.

Decision rule: fund the flows that reach every new customer first. Add campaigns once those run and you can see repeat orders in Shopify.

4. Organic: the slow line you fund before you need it

Organic search has no per-click charge, but it is not free: it takes content, technical work and time, and pays back slowest of the five. How organic traffic lowers blended CAC explains how it pulls the average cost per customer down. Nobody can promise a timeline, including us, so do not judge it after one month.

Decision rule: keep it as a steady line. If cash is short and your payback tolerance is days, not months, trim organic before conversion or retention.

5. Paid acquisition: sized by the floor and your maximum CAC

Paid is the fastest line to turn on and the easiest to overspend. Three tests, using the earlier numbers:

• CAC at or below first-order CM2 (A$54): paid pays for itself on the first order. Scale in planned steps and check blended results each time.
• CAC between A$54 and A$64.80: fine if your cash covers the wait. Watch repeat orders in your cohorts, not platform dashboards.
• CAC above A$64.80: do not raise the budget. Look at conversion, the offer and margin first.

At A$1,600 a month, about A$53 a day, the example store would see roughly 36 purchases a month at an A$45 cost per purchase. That is about 8 a week, well under Meta’s 50. That is below the A$225 a day that Meta’s five-times guidance implies for a A$45 goal, and the ad set is likely to show Learning limited. Keep the structure simple and read results over weeks.

How budget settings behave on Meta and Google

One more dated fact, as at 31 October 2026. Google’s Target ROAS page says that from 17 August 2026 it is updating its bidding systems for campaigns limited by budget, and that Target CPA and Target ROAS campaigns may see temporary performance and traffic fluctuations. If your recent numbers look odd, check that page before you rewrite your budget.

If you are reading this in Black Friday week

Black Friday is Friday 27 November 2026 and Cyber Monday is Monday 30 November. Do not set a year’s budget from a sale weekend, because discounts lower CM2. Give the sale its own capped budget using the same sums, with the discount already taken out of net order value, and avoid last-minute edits to running campaigns, since Meta says editing an ad set during its learning phase resets learning. The Black Friday checklist covers the store side.

When to re-run the numbers

Recalculate monthly, and again when product cost, shipping or payment fees move, when you change discounting, when measured CAC passes your maximum CAC, or when a tracking fix or platform measurement change (like Shopify’s in September) resets your baseline.

Key takeaways

• Published percentages conflict. Shopify’s pages give 2% to 8%, just under 10% and 7.8%, and the vast majority of respondents in the 2026 Gartner survey behind one of them reported revenue over $1 billion.
• Margin sets the budget. In the illustrative example, marketing room swings from −2% to 23% of revenue as CM2 moves from 40% to 65%.
• Set the ceiling from CM2, fixed costs and the profit you want, then a payback rule to get your maximum CAC.
• Check the floor. Meta says about 50 results in the week after a significant edit; at an illustrative A$45 per purchase that is far above a small store’s whole marketing room.
• Spend in order: tracking, conversion, retention and organic, then paid sized to the floor and your maximum CAC.
• Prism3 does not buy media or run email. The sums work whoever manages your ads.

FAQs

How much should I spend on marketing for my Shopify store?

There is no sourced single figure. Work it from margin: what one order leaves, less fixed costs and the profit you want. The remainder is your ceiling at today’s revenue. In the illustrative example it was A$4,000 a month, and a different margin moved it from below zero to 23% of revenue.

What percentage of revenue should a Shopify store spend on marketing?

Shopify’s Australian small-business guide says 2% to 8% of gross annual revenue without naming a source. Its other guides attribute just under 10% (2022) and 7.8% (2026) to Gartner, and the vast majority of respondents in Gartner’s 2026 survey reported revenue over $1 billion. Treat them as a sanity check at most.

How much should a small Shopify store spend on Meta or Google ads?

Enough to read the results, and no more than margin allows. Meta says ad sets usually leave learning after about 50 results in the week after the last significant edit, and Google’s Target ROAS needs at least 15 conversions in 30 days on Search and Shopping. Run that arithmetic with your own cost per purchase; if it exceeds your ceiling, run a small capped test.

Should I spend on SEO or ads first?

Fix tracking and the biggest conversion leak before either. After that, ads switch on fastest but you pay for traffic as it arrives, while SEO has no per-click charge and takes longest to pay back. If you can wait months, fund both, with ads capped by your maximum CAC. If cash is tight, ads are easier to switch off.

What if my margin leaves no room for marketing?

Then the problem is upstream of the budget. Look at price, discounting, shipping you absorb, product cost and fixed costs before you spend on ads.

Is a daily ad budget a hard cap?

No. For most campaigns, Google can spend up to two times the average daily budget on a day, within a monthly limit of 30.4 times it. Meta can spend up to 75% over on a day, within a weekly limit of seven times the daily budget. Plan by week and month.

Keep reading

• Shopify growth order: what to fix first before ads
• ROAS vs MER vs blended CAC for Shopify
• High CAC, low LTV on Shopify: what to do first

Not sure which line of your budget is leaking?

The first two lines of the budget order are tracking you can trust and conversion on the store; those are the two we work on. Book a free 30-minute diagnostic: we look at where your store loses buyers after the click, using your own numbers, and tell you what to fix first. We do not run ads or email. No pitch. No email gate.

If conversion and tracking are the lines you want to understand first, see how our Shopify CRO service checks tracking before any optimisation.