Google Ads Metrics for eCommerce: The 8 That Actually Matter (2026)

The Google Ads metrics that matter for eCommerce are the ones tied to profit, not clicks. Track these 8: ROAS, POAS (profit on ad spend), conversion rate, cost per acquisition (CPA), average order value (AOV), customer lifetime value (LTV), conversion value, and impression share. Ignore the pretty ones like raw impressions and clicks. The single most important number is POAS, because a store can post a 5x ROAS and still lose money after costs. This guide breaks down each metric in plain English, with 2026 benchmarks. V8 Media has driven R2+ billion in client sales since 2018.

Why most eCommerce stores track the wrong metrics

Open most Google Ads accounts and you see the same thing. The owner is staring at clicks and impressions.

Big numbers. Green arrows. Feels like winning.

It is not.

A click is not a sale. An impression is not a Rand. You can have a million impressions and a bank account as empty as a fridge in stage 6 load-shedding. We have audited hundreds of stores running exactly like this, scaling spend on numbers that mean nothing.

One rule fixes most accounts. Track what connects a click to money in the bank. Everything else is noise Google is happy to charge you for.

The goal of an eCommerce Google Ads account is not traffic. It is profit. So your metrics need to measure the journey from a click to money in the bank, not the noise in between.

The 8 Google Ads metrics every eCommerce store should track

Google gives you 50-odd metrics. Most are useless for a store. These 8 are the ones that move profit.

1. ROAS (return on ad spend)

ROAS is revenue divided by ad spend. Spend R10,000, make R50,000 in sales, that is a 5x ROAS, or 500%.

It is the metric everyone quotes. It is also the one that lies to you the most.

Why? ROAS counts revenue, not profit. It ignores your product cost, shipping, payment fees, returns, and VAT. A 5x ROAS on a product with a thin margin can still be a loss. More on this in the next section, because it is the trap that sinks most stores.

2. POAS (profit on ad spend)

POAS is the one that tells the truth. It is profit divided by ad spend, not revenue divided by ad spend.

This is the metric we build client accounts around. Revenue is vanity. Profit is sanity. A store chasing ROAS scales the products that sell the most. A store chasing POAS scales the products that make the most money. Those are rarely the same list.

If you only ever fix one thing about your reporting, switch your headline number from ROAS to POAS.

3. Conversion rate

Conversion rate is the share of ad clicks that turn into orders. 100 clicks, 3 sales, a 3% conversion rate.

This number tells you whether your ads and your store are pulling in the same direction. A great ad sending traffic to a slow, confusing product page burns money on every click.

For eCommerce Search ads, the average conversion rate is 2.81%, according to WordStream's benchmark data. Shopping ads sit lower, around 1.91%. If you are well under that, the leak is usually the landing page or the offer, not the ad.

4. CPA (cost per acquisition)

CPA is what you pay to land one order. Spend R5,000, get 25 orders, your CPA is R200.

The only question that matters: is your CPA below the profit on the order? If a sale makes you R300 profit and your CPA is R200, you keep R100. If your CPA climbs to R350, you are paying to lose money.

WordStream's benchmark data has historically put the average eCommerce Search CPA at about $45 (a global average). Useful as a sanity check, but your own margin is the number that decides if a CPA is good or fatal.

5. AOV (average order value)

AOV is your total revenue divided by number of orders. R200,000 in sales across 500 orders is an R400 AOV.

AOV is the quiet hero of a profitable account. Lift it and a CPA that looked scary becomes comfortable overnight.

Say your CPA is R250. At an R400 AOV with a 40% margin, you make R160 a sale and lose R90. Push the AOV to R700 with a bundle or an upsell, and that same order now makes R280. Same ads, same CPA, suddenly profitable. That is why we obsess over AOV before we touch the bidding.

6. LTV (customer lifetime value)

LTV is the total profit a customer brings you over their whole relationship, not just the first order.

This is the metric that lets you win. If a customer buys three times a year for two years, their first order is not the whole prize. You can afford to break even, or even bleed a little, on that first sale and still win the customer.

The store that knows its LTV can outbid the store that only looks at the first order. Every time. While your competitor panics at a R250 CPA, you happily pay it because you know that customer is worth R2,000 over two years.

7. Conversion value

Conversion value is the Rand figure you send back to Google when a sale happens. For a store it is simply the order value.

This one is not just a report. It is the fuel for Google's smart bidding. Feed it accurate Rand values and Target ROAS bidding can chase your most valuable orders. Feed it nothing, or wrong numbers, and the machine optimises for the wrong sales.

Accounts that bid on conversion value instead of conversion count consistently outperform on ROAS in eCommerce. Because Google is finally chasing money, not order count. Get your tracking right first, or this number is built on sand. Our guide on Google Ads conversion tracking walks through the setup.

8. Impression share

Impression share is the percentage of the available impressions you actually got. 100% means you showed every time you could have.

It answers a simple question: how much room is left to grow? A profitable campaign sitting at 40% impression share has more demand to capture. You are leaving sales on the table.

It also flags problems. If your "lost impression share (budget)" is high, your budget is capping a winner. If "lost impression share (rank)" is high, your ads or bids need work. Either way it tells you where to push.

ROAS vs POAS: the metric that lies to you

This is the most important section in this guide. Get it wrong and you will scale a store straight into a loss.

ROAS measures revenue. POAS measures profit. The gap between them is every cost ROAS ignores: product cost, shipping, transaction fees, returns, and VAT.

Here is the trap in Rands. Two products, both running at a 4x ROAS.

• Product A: a R1,000 supplement that costs you R200 to make. Big margin. A 4x ROAS here is a goldmine.
• Product B: a R1,000 electronics item that costs you R850 to source. Thin margin. That same 4x ROAS is a loss after shipping and fees.

Same ROAS. Opposite outcome. If you only watch ROAS, you treat these two products as equals and scale both. One prints money. One bleeds it.

POAS sees the difference instantly, because it works off profit, not revenue. This is why we moved every client account off ROAS as the headline number years ago. We break the full comparison down in our guide to ROAS vs POAS for eCommerce.

The fix is not complicated. Know your true margin per product, feed those numbers into your reporting, and judge campaigns on profit. The data only tells you the truth if you set it up to.

eCommerce Google Ads benchmarks for 2026

These are averages, not targets. Your margin and price point can push them in either direction. Here is where they sit so you can spot if something in your account is badly off.

Benchmark figures from WordStream, as compiled by Store Growers (updated January 2026). Dollar figures are global averages.

On ROAS, the average eCommerce account runs closer to 2.9x, with the median nearer 2x, according to Triple Whale's benchmark data. Strong performers push for 4x to 8x as a target. But any ROAS number is meaningless without your margin.

Here is the maths that actually matters. Your break-even ROAS is 1 divided by your margin. A store with 40% margins breaks even at 2.5x (250%). Thin 25% margins need 4x just to stand still. Fat 60%+ margins can profit below 1.7x. This is exactly why POAS beats a blanket ROAS target. The "good" number depends entirely on what you keep after costs.

One more South African note. These benchmarks are in dollars and skew to large global accounts. Local CPCs in many SA niches run cheaper, but so does buying power. Use them for direction, then build your own baseline from your real numbers.

The vanity metrics to ignore (or demote)

Some numbers look important and are mostly noise for a store owner. Do not bin them completely. Just stop making decisions on them.

• Impressions. How often your ad showed. Nice to know. Pays no invoices. Big impressions with no sales just means you are renting attention you cannot convert.
• Raw clicks. Clicks are traffic, not customers. A cheap click that never buys is more expensive than a pricey click that does.
• CTR on its own. A high click-through rate feels great. But a clickbait ad pulling clicks from people who never buy will wreck your conversion rate and your CPA. Watch CTR next to conversion rate, never alone.
• Quality Score obsession. Useful as a diagnostic. It is not a goal. Nobody ever banked a Quality Score of 10.

The pattern is simple. Top-of-funnel activity metrics are for diagnosing, not deciding. Profit metrics are for deciding.

How to actually use these metrics (a Rand example)

Metrics are useless until they change a decision. Here is how the right ones flip a call you would otherwise get wrong.

Two campaigns. Same R20,000 spend. Same store.

• Campaign A: 200 orders. CPA of R100. AOV of R350. Looks brilliant on cost per order. But the products are low margin, making R120 profit each. Profit: R24,000 on R20,000 spend. You cleared R4,000.
• Campaign B: 80 orders. CPA of R250. AOV of R900. Looks expensive. But these are high-margin bundles making R450 profit each. Profit: R36,000 on R20,000 spend. You cleared R16,000.

CPA says Campaign A wins. POAS says Campaign B wins by four times the profit.

If you only watch CPA and order count, you scale the loser and starve the winner. If you watch profit, AOV, and POAS, you do the opposite. Same data, opposite decision. That is the whole point of measuring the right things.

Tracking is what makes this possible. If your Google Ads account is not feeding accurate conversion values back to Google, none of these numbers can be trusted. Fix the plumbing first.

What South African store owners need to know

Two local things change how you read these metrics.

Track in Rands and track profit. Most SA stores log a "conversion" and stop. A conversion is not money in the bank. Tie every order to its real Rand value, subtract your true costs including the 15% VAT, and watch profit per order. A campaign with 200 cheap orders can lose to one with 60 good ones. The numbers only show you that if you set them up to.

Mind the margin squeeze. Local couriers, import costs, and a weak Rand on imported stock all eat margin that a US benchmark never sees. That makes POAS even more important here than overseas. A 4x ROAS that works for a US store can quietly lose money for an SA store importing its products.

Get the margin maths right and everything else follows. For the full picture, see our guide to healthy eCommerce profit margins.

Where these metrics fit in your bigger picture

Google Ads metrics are one slice of your store's scoreboard. They tell you if your paid traffic is profitable. They do not tell you if your whole business is healthy.

Roll these up into your monthly review alongside the rest. Start with our guide to the monthly eCommerce KPIs every store should track. Then, if you also run lead-gen campaigns, the metrics shift, so read our companion guide on the Google Ads metrics for lead generation.

Same rule every time. Measure profit, not noise. Then do what the numbers say, even when it costs you a campaign you liked.

Frequently asked questions

What is the most important Google Ads metric for eCommerce?

POAS, profit on ad spend, is the most important metric for an eCommerce store. It measures the actual profit you keep for every Rand spent on ads, after product cost, shipping, fees, and VAT. ROAS gets quoted more, but it counts revenue and can hide a loss.

What is a good ROAS for eCommerce Google Ads?

Many eCommerce brands aim for a 4x to 8x ROAS (400% to 800%) as a target, though the average account runs closer to 2.9x. A "good" ROAS depends on your margin. Your break-even ROAS is 1 divided by your margin: a 40% margin store breaks even at about 2.5x (250%), while a 60%+ margin store can profit below 1.7x (170%).

What is the difference between ROAS and POAS?

ROAS is revenue divided by ad spend. POAS is profit divided by ad spend. POAS subtracts your product cost, shipping, fees, returns, and VAT, so it shows the money you actually keep. Two products on the same ROAS can have wildly different POAS.

What is a good conversion rate for eCommerce Google Ads?

The average conversion rate for eCommerce Search ads is about 2.81%, and around 1.91% for Shopping ads, per WordStream benchmark data. Top performers reach much higher. If you sit well below average, the leak is usually your landing page or offer, not the ad.

Should I track CPA or ROAS for my online store?

Track both, but lead with profit. CPA tells you what one order costs to win and must sit below your profit per order. ROAS shows revenue efficiency. For eCommerce, ROAS (and ideally POAS) is the better headline metric because order values vary so much.

Why are clicks and impressions not enough?

Clicks and impressions measure attention, not money. You can have huge impressions and clicks and still make no profit. They are useful for diagnosing problems, but decisions should be made on profit metrics like POAS, CPA against margin, and conversion rate.