What is the minimum ad budget for an online store?

Why is there no single right number for an ad budget?

There is no single right number because the budget depends on what one order costs you to win and what it earns you. An online store selling 45 riyal phone cases and one selling 3,000 riyal watches need very different budgets on the same platform. A rule like "spend 10% of revenue" ignores both numbers, so it is a habit, not a method.

What you sell matters only through its numbers. Three of them decide the budget:

What is the most I can pay in ads for one order?

The most you can pay in ads for one order is your profit on that order before advertising. Take your average order value and subtract everything the order costs you: the product, delivery, payment fees, packaging, and an allowance for returns and refused cash-on-delivery orders. What is left is your break-even cost per order.

Here is the worksheet, with illustrative numbers for an online store in Qatar:

This online store can pay up to 120 riyals in ads for an order and still break even. If it wants 30 riyals of profit on each first order, its target is 90 riyals or less.

Many online stores in Qatar and the Gulf leave out the last cost line. A cash-on-delivery order refused at the door still costs delivery both ways. If some of your orders come back, include them, or your limit will look higher than it is.

You will also hear about break-even ROAS. ROAS (return on ad spend) is sales divided by ad spend. Your break-even ROAS is your order value divided by your profit per order: here 300 divided by 120, or 2.5x.

Count repeat orders only if you have real data on how often customers come back. Large brands often accept a loss on the first order because they know later orders pay it back. A young online store usually cannot afford that bet until its own numbers prove it.

What will an order probably cost me?

Your expected cost per order is what you pay for one visit from an ad, divided by the share of those visitors who buy. If a visit costs 1.6 riyals and 2% of those visitors place an order, each order costs 80 riyals in ads.

Use the best numbers you have, in this order:

1. Your own ad results from the last three months, if you have run ads before.

2. Your website's conversion rate from your analytics or your Shopify or WooCommerce reports. Visitors from social media ads often buy less readily than visitors who search for you, so treat this as the best case.

3. A small test, which Stage 1 below gives you anyway.

Be careful with average figures from blogs and calculators. Most come from other countries, years and products, and few say how they were measured.

Then compare the two numbers. If your expected cost per order is below your limit, you can test. If it is above, more budget only makes the loss bigger, so fix the offer first. Raise the order value with a bundle or a free-delivery threshold, improve the product pages, or lower your costs. Our guide on why an online store gets traffic but no orders shows where to look.

The conversion rate moves the budget more than most owners expect. One of our clients, a garden and home online store in Qatar, saw its conversion rate rise from 0.42% to 0.78% after we redesigned and restructured the online store. Six months after launch, gross sales were up 573% on the baseline before the work, with no increase in ad spend.

How many orders does the ad platform need to learn?

Meta, which runs ads on Facebook and Instagram, wants about 50 results a week for each ad set before it stops "learning". An ad set is the part of a campaign that holds the budget, the audience and the ads. Meta's Business Help Centre (November 2026) calls an ad set "Learning limited" when it is unlikely to reach about 50 results. The count runs for the week after its last significant edit.

Three things follow from Meta's rule:

• It counts per ad set, not per campaign. Split one budget across five ad sets and each one needs its own 50 orders.

• Edits restart the count. Meta lists changes to the audience, the ads or the goal, adding a new ad, and pausing for 7 days or longer. A small budget change usually does not matter; a very large jump may.

• Learning costs more, but it still sells. Meta says learning can bring less stable results and a higher cost per result. It does not say results are impossible.

Meta's Peak Season 2026 guidance (21 October 2026) adds a planning formula: your weekly budget divided by (cost per result × 50) is the most ad sets your budget can support. With an 80 riyal cost per order, 4,000 riyals a week supports one ad set.

Other platforms set their own rules, as of November 2026. Google Ads asks for at least 15 conversions in the past 30 days before Target ROAS bidding can be used on Search and Shopping campaigns. Target ROAS is the setting where Google adjusts your bids to reach a return on ad spend you choose. Snapchat gives a range close to Meta's 50: wait for 30 to 50 conversions before judging an ad set. TikTok publishes its own budget guidance, so check its help centre before you plan.

Should I trust the ad platform's budget advice?

Trust the platform's rules about how its system learns, but decide for yourself how much to spend. The advice is accurate, and it also comes from the company that is paid for every riyal you spend.

You can see this in the advice itself. When an ad set is Learning limited, one of the fixes Meta lists in 2026 is to raise your budget. Meta's Peak Season post (October 2026) says ad sets under about 50 conversions a week ran about 28% more expensive per purchase in Meta's own analysis. Meta does not say how it measured this. Google Ads can apply some recommendations automatically, including changes to your cost and return targets, and those changes move spend.

Experienced media buyers often push back on the idea of spending enough to exit learning from day one. Their view, which matches what we see with online stores, is this:

• Ad sets can sell while they are still learning. Learning limited is a warning, not a verdict.

• Raising the budget to escape learning does not fix an ad that loses money. It makes the loss bigger, faster.

• Meta's 50-a-week level is where you arrive once the ads are profitable, not the price of entry.

What should I check before I spend anything?

Check that every order is recorded correctly, or every number in this guide will be wrong. The Meta Pixel is a small piece of code on your website that tells Meta when someone buys. The Conversions API sends the same order from your online store's server, so fewer orders go missing. Make sure both record purchases with their value. Then place a test order and confirm it appears in your online store and once, not twice, in the ad platform's report. Our Shopify store audit checklist covers these checks step by step.

If you take cash on delivery, remember that the ad platform counts an order when it is placed, even if it is refused at the door. Judge your cost per order on delivered orders, not on the platform's count.

How do I set the budget in three stages?

Set the budget in three stages. A proof budget shows whether the ads can sell at a price you can afford. A learning budget gives the platform enough orders, and a ceiling tells you when to stop. Each stage is earned by the results of the one before it.

Stage 1: the proof budget. Spend enough to buy about 20 to 30 orders at your expected cost per order, and never more than you could lose without harm. With fewer orders, a good or bad week can be luck. Use one campaign, one ad set, one offer (usually your best seller) and three to five different ads. Launch all the ads on the first day, because adding an ad later restarts learning.

Make no changes for 7 to 10 days, and count orders in your online store, not clicks. The stage is passed when your cost per order sits at or below your limit.

Stage 2: the learning budget. Once Stage 1 passes, raise the budget in steps towards Meta's level of about 50 orders a week for that ad set. A common practice is to raise it by about 20% every few days. Check the cost per order after each step. If it rises above your limit, step back and fix the ads or the offer.

Stage 3: the ceiling. Keep raising the budget while each step still brings orders below your limit. Each extra riyal reaches people a little less likely to buy, so the cost per order rises as the budget grows.

To see where to stop, compare the extra orders with the extra spend after each step. Say the budget rises from 4,000 to 4,800 riyals a week and weekly orders rise from Meta's level of 50 to 56. The extra 800 riyals bought 6 extra orders, about 133 riyals each. That is above a 120 riyal limit, so go back to the last step. Your average cost per order can still look fine while those extra orders lose money.

What does this look like for a real budget?

Here is the staged method with the illustrative numbers above: a 120 riyal limit per order and an expected cost per order of 80 riyals.

The usual formula would ask this online store for 4,000 riyals a week from the first day. The staged method risks about 2,000 riyals to find out first whether the ads can sell at a price it can afford.

For a monthly figure, multiply the weekly budget by about 4.35, the average number of weeks in a month. In this example that is about 17,400 riyals a month at the learning stage.

What if I cannot afford even the proof budget?

If you cannot afford the proof budget, shrink the test, not the logic. A small budget spread thinly gives you several ad sets that learn nothing.

Try these in order:

1. One campaign and one ad set. Meta itself lists combining ad sets and campaigns as a fix for Learning limited.

2. One product or one offer. Choose the one that sells best and earns the most per order.

3. One channel. Do not add Google or Snapchat until the first channel works.

4. A more frequent goal for now. Meta suggests moving from purchases to add to cart when purchases are too rare. If, say, 10% of ad visitors add a product to their cart, each add to cart costs 16 riyals, and 50 a week cost 800 riyals. The catch: Meta will find people who add to cart, and some of them never buy, so keep counting real orders. Switch back to purchases once orders come in steadily.

5. A higher order value. A bundle or a free-delivery threshold can lift profit per order enough to make the numbers work.

If none of these makes the test affordable, wait. Fix the product pages and checkout, grow your audience without ads, and come back when the numbers work. Ads below a readable level usually waste the money and teach you nothing.

How do I add Google, TikTok or Snapchat?

Add channels one at a time, and give each one its own proof budget. Each platform learns separately, so a budget split across three platforms from day one often leaves all three stuck learning.

When you run more than one channel, judge them together. Divide your total online store sales by your total ad spend, and track how many orders come from new customers. Each platform counts the sales it touched, so adding up their reports often counts one order twice. At larger budgets, pause one channel for a period and see whether total sales really drop.

What do online stores in Qatar get wrong with their first ad budget?

The most common mistake is treating the budget as the problem when the online store, the offer or the tracking is the real one. These are the mistakes we see most often in our work with online stores in Qatar:

1. Advertising into a broken checkout. One client, a watch brand's online store in Doha, saw shoppers add $382K of watches to their carts in a two-week test while only $8.45K was bought. A fault in the card payment step was blocking orders. Cart value is the value of products added to a cart, not revenue, and no budget would have fixed a payment step that did not work.

2. Judging the ads on the platform's numbers. Clicks, cart value and the platform's own sales count all look like progress. Only orders paid in your online store, from new and returning customers, tell you whether the budget works.

3. Paying for the wrong goal. Part of the budget often goes to traffic, engagement or profile-visit campaigns because they look cheap. They bring visitors and likes, not buyers. For an online store, the first budget belongs in a sales campaign.

Where to go from here

Fill in the worksheet with your own numbers today: order value, product cost, delivery, fees and returns. If your expected cost per order is below your limit, plan a proof budget for 20 to 30 orders in one campaign and one ad set. If it is above, fix the offer before you spend. If you would like a second pair of eyes on your online store, tell us about it.