A £50,000 month in tracked revenue can still be a bad result if the discounts were too deep, fulfilment costs rose, and new customers never came back. That is why the best ecommerce ad metrics are not the ones that make a platform dashboard look impressive. They are the ones that tell you whether paid media is creating profitable, scalable demand.
For established eCommerce brands, clicks, impressions and even headline conversion volume are supporting data. They are not the commercial outcome. The job of Google Ads, Shopping, Performance Max and Meta is to acquire the right customers at a cost your margins can sustain, then give you enough confidence to spend more without eroding profit.
Start with the metric your business can actually afford
Before judging a campaign, establish your breakeven point. This is the foundation of sensible media buying and the point many accounts skip entirely.
Your breakeven cost per acquisition is not just your average order value divided by a target return on ad spend. It should account for product cost, payment fees, delivery subsidy, picking and packing, returns, discounting, VAT where relevant, and any variable costs that rise with each sale. If you sell products with materially different margins, a single account-wide target can be misleading.
A £40 acquisition cost may be excellent for a high-margin bundle and destructive for a low-margin hero product. The platform will happily optimise towards revenue if you let it. It has no natural interest in your contribution margin.
For that reason, the most commercially useful north-star metrics are contribution profit after ad spend and cost of sale against a defined profitability threshold. ROAS remains useful, particularly for day-to-day bidding and channel comparison, but it is a ratio, not a profit statement. A 5x ROAS can be weak if margins are thin. A 2.5x ROAS can be highly attractive for a brand with strong gross margin and repeat purchase behaviour.
The best ecommerce ad metrics, ranked by commercial value
1. Contribution profit after ad spend
This is the closest your advertising reporting gets to reality. It measures the money left after the variable costs of delivering an order and the advertising cost have been removed.
It stops teams celebrating revenue that does not contribute meaningfully to the business. It also changes how you assess sales promotions. A campaign may produce a sharp revenue spike during a discount event while reducing contribution profit because the lower selling price and higher media cost cancel out the gain.
You may not be able to feed every cost directly into Google or Meta. That is fine. The calculation can sit in your reporting layer, provided it uses clean source data and is reviewed consistently.
2. Blended MER and blended cost of sale
Marketing efficiency ratio, commonly called MER, compares total revenue with total advertising spend across channels. Blended cost of sale is the inverse: total ad spend divided by total revenue.
These figures matter because platform attribution is never the complete picture. Google may claim a conversion, Meta may claim the same conversion, and both can be technically correct within their attribution settings. Your finance figures do not double count the order.
Blended reporting gives decision-makers a harder-to-game view of performance. If individual channel ROAS is improving while total revenue stays flat and blended efficiency worsens, something is wrong. You may be paying more to capture demand that would have converted anyway, or simply shifting credit between platforms.
MER should not be used in isolation. Organic demand, email performance, seasonality, stock availability and pricing all affect it. But paired with channel-level data, it is one of the clearest checks on whether paid media is genuinely adding value.
3. New customer acquisition cost
A campaign that repeatedly converts existing customers can look very efficient, especially for brands with a loyal audience. That does not automatically make it a growth campaign.
New customer acquisition cost shows what it costs to bring a genuinely new buyer into the business. For brands with repeatable products, subscriptions or a strong replenishment cycle, this can justify a lower first-order ROAS than the standard dashboard target suggests.
The trade-off is cash flow. You cannot fund an aggressive acquisition strategy with future lifetime value if the business lacks the working capital to carry it. Treat projected lifetime value carefully too. Use cohort data from real customer behaviour, not an optimistic assumption that every first-time customer will return.
4. New customer revenue share
New customer acquisition cost tells you the price of growth. New customer revenue share shows whether your paid media mix is actually producing it.
If 80% of attributed paid revenue comes from returning customers, branded search or remarketing, your account may be harvesting an existing customer base rather than expanding it. There is a role for those campaigns. Brand protection and remarketing can be highly profitable. The issue is reporting them as if they are prospecting success.
Separate campaigns and reporting where possible. This is particularly relevant in Performance Max, where broad automation can blur the line between new demand generation and brand-led conversion capture.
5. Conversion rate and checkout completion rate
Conversion rate is not a vanity metric, but it is often misused. A higher conversion rate does not always mean better advertising. Branded traffic, returning visitors and aggressive remarketing will usually convert at a higher rate than cold prospecting.
Use conversion rate to diagnose the quality of traffic and the strength of the landing experience. Compare like with like: non-brand Search against non-brand Search, prospecting audiences against prospecting audiences, and product categories with comparable price points.
Checkout completion rate is often more revealing. If product page traffic is healthy but checkout completion falls, the problem may be delivery pricing, payment options, trust signals or a technical issue rather than campaign targeting. Do not keep changing bids to solve a website problem.
6. Average order value and margin-weighted revenue
Higher average order value can improve the economics of acquisition without demanding lower media costs. Bundles, multi-buy offers, sensible upsells and product-led landing pages can change what a viable CPA looks like.
However, average order value needs context. It can rise because a campaign is only reaching existing high-value customers, or because discounting has pushed customers to add products they would not otherwise buy. Track margin alongside it.
For multi-category retailers, margin-weighted revenue is more useful than plain conversion value. If your feed and campaign structure treat every pound of revenue as equal, automated bidding may prioritise products that turn over well but deliver little profit. Feed optimisation, product segmentation and clearer value signals help correct that bias.
Metrics that diagnose waste before it becomes expensive
Profit metrics tell you what happened. Diagnostic metrics help explain why it happened and where to act.
Search impression share can identify demand you are missing on profitable terms, but only after you know the terms themselves meet your cost-of-sale target. Cost per click can flag auction pressure or weak relevance, yet a cheaper click is worthless if it reduces purchase intent. Frequency on Meta can expose audience fatigue, especially where prospecting performance deteriorates as spend increases.
For Shopping and Performance Max, review product-level spend, revenue, cost of sale and conversion volume. A campaign can meet its overall target while wasting a meaningful portion of spend on products with poor availability, weak margins, unsuitable price points or persistent return issues. Excluding or restructuring around those products is often more valuable than another creative test.
Also watch the rate at which budget increases translate into incremental revenue. The first £5,000 of monthly spend may be highly efficient. The next £10,000 may require broader audiences, weaker search queries or more expensive auctions. Scaling is not simply increasing budgets. It is managing the point at which marginal efficiency starts to fall.
Build reporting around decisions, not platform screenshots
A useful weekly report should answer a small number of commercial questions: Are we above or below profitable cost of sale? Is total efficiency holding as spend rises? Are we acquiring enough new customers? Which products, campaigns or audiences are consuming spend without a credible path to profitability?
Avoid a reporting pack packed with unprioritised metrics. It creates activity without accountability. A founder should be able to see the relationship between spend, revenue, contribution, new customer growth and the next action within a few minutes.
At Oxedent, that discipline starts with knowing the brand’s real targets before campaign changes are made. No amount of platform expertise compensates for unclear margins, poor product data or a reporting model that rewards revenue at any cost.
The useful next step is not adding another dashboard widget. Take your last 90 days of spend, calculate the contribution from paid orders, separate new and returning customers, and identify the products carrying the cost. That exercise will usually show where profitable scale is being built and where budget is merely being spent.