A growing ad account can still be a bad business decision. If revenue is rising but contribution margin is shrinking, stock is tied up in slow-moving lines, or customer acquisition is eating the cash needed to fulfil orders, the campaign is not scaling properly. A serious ecommerce paid media strategy starts with that commercial reality, not with a platform recommendation or an agency dashboard full of green arrows.
For established retailers, paid media should be a controllable growth engine. That means knowing what a new customer is worth, where budget is being wasted, which products can absorb more demand, and when a healthy-looking ROAS is actually masking an unprofitable outcome.
Start with the number the business can afford
Before deciding how much to spend on Google Ads, Meta or Performance Max, define the maximum acquisition cost the business can sustain. This is not a generic industry benchmark. It is a calculation built from your own gross margin, delivery costs, payment fees, returns rate, discounting, overheads and repeat purchase behaviour.
A brand with a 70% gross margin can often tolerate a very different cost of sale from a retailer selling bulky, low-margin products with expensive fulfilment. The same applies to product categories. Your hero line may justify aggressive prospecting spend, while an accessory with a modest average order value may only work through remarketing or as part of a bundle.
There is also a choice to make between first-order profitability and lifetime value. Both can be valid. If repeat purchases are reliable, tracked properly and supported by cash flow, accepting a lower initial ROAS may be rational. If retention is weak or demand is largely one-off, spending against assumed lifetime value is a fast route to overstating performance. Make the decision deliberately, then use it to set channel targets.
Build the ecommerce paid media strategy around intent
Not all paid traffic has the same job. Treating every campaign as though it should deliver the same return creates poor decisions, especially when platforms overlap.
Google Shopping and search campaigns typically capture existing demand. Someone searching for a specific product, brand or category has already signalled intent, so these campaigns are often efficient at converting demand that exists now. The limitation is volume: you cannot scale indefinitely if search demand is finite, and bidding harder can quickly erode profitability.
Meta advertising is better placed to create and shape demand. It can introduce products to relevant audiences before they actively search, build consideration and feed future branded search. That does not give Meta a free pass on profitability. It does mean evaluating it solely against last-click reporting can understate its role, particularly for distinctive products, higher-consideration purchases or brands with a strong creative proposition.
Performance Max can be highly effective when product data, conversion tracking and asset quality are under control. It can also become a black box that leans heavily on branded demand and existing customers while claiming credit for revenue that would have arrived anyway. The answer is not to avoid automation. It is to structure measurement, brand exclusions where appropriate, product segmentation and reporting so that automation works within commercial boundaries.
Put product economics into campaign structure
A product feed is not administration. For many ecommerce accounts, it is one of the biggest levers available to improve Shopping and Performance Max performance.
Titles should reflect how customers search, without becoming unreadable keyword strings. Product type, brand, colour, size, material and key specifications can all matter when they clarify relevance. Correct GTINs, accurate availability, compelling images and clean categorisation give platforms better inputs and shoppers fewer reasons to hesitate.
More importantly, campaigns should reflect the commercial value of the catalogue. Do not give equal budget priority to every SKU simply because it exists in the feed. Segment products by margin, stock depth, price point, conversion rate and proven demand. Bestsellers with healthy margin may deserve their own budget and bidding approach. Clearance lines may need controlled support. Low-margin products that repeatedly consume spend without contributing profit should be excluded or positioned as cross-sell products rather than acquisition drivers.
This structure must stay fluid. Stock levels change, seasonality changes and product performance changes. A campaign structure that was sensible three months ago may now be directing budget towards products that cannot scale or should no longer be promoted.
Measure profit, not platform theatre
Clicks, impressions and even conversion volume are supporting metrics. They help diagnose performance, but none tells you whether advertising is making the business stronger.
The core reporting should connect ad spend to revenue, cost of sale, gross profit and contribution wherever the data allows. It should also separate new and returning customers. A campaign generating impressive ROAS from people who already know the brand is not equivalent to one profitably acquiring new customers at scale.
Platform attribution should be treated as directional rather than absolute. Google Ads and Meta each have an incentive to claim as much value as their models can justify. Compare their reports with ecommerce platform data, analytics trends, blended MER or marketing efficiency ratio, and branded search movement. No single view is perfect. Together, they reveal whether spend is creating incremental growth or simply harvesting demand generated elsewhere.
Tracking quality is non-negotiable. Verify purchase values, currencies, refunds where possible, consent mode, enhanced conversions and deduplication between browser and server-side events. Bad data does not merely make reporting inconvenient. It trains automated bidding towards the wrong outcomes.
Scale in controlled increments
The temptation after a strong week is to double budget. Usually, that is how efficient campaigns become expensive ones. Scaling changes the auction opportunities a platform can access. Once the most obvious converting users are exhausted, cost per acquisition often rises.
Increase spend in measured increments, then allow enough time for performance to settle before drawing conclusions. The appropriate pace depends on conversion volume, buying cycle and seasonality. A high-volume consumable brand can learn faster than a retailer selling premium furniture or specialist equipment.
Budget increases should be paired with a clear hypothesis. Perhaps a high-margin category has limited impression share. Perhaps Meta prospecting creative has held acquisition cost below target across several weeks. Perhaps a new product range has enough stock and a proven conversion rate to justify dedicated investment. Scale the reason, not the excitement.
Creative also becomes more important as spend rises. A single winning advert eventually reaches saturation. Build a testing process that examines different product angles, offers, formats and objections. New creative should not be produced for the sake of novelty. It should answer a commercial question: can this message attract a better customer, improve conversion quality or open a larger profitable audience?
Cut waste before asking for more budget
Most mature accounts have waste hiding in plain sight. It appears as irrelevant search queries, broad campaigns with weak product groups, poorly controlled brand traffic, unprofitable placements, duplicate targeting, out-of-stock products and spend on products with no viable margin.
Waste reduction is not about making an account look tidy. It releases budget for campaigns that can genuinely grow. Review search terms, product-level returns, audience overlap and spend concentration regularly. If 20% of spend is producing negligible profit, fixing that may improve results faster than launching another campaign type.
Be careful not to over-correct. Excluding too aggressively can limit discovery, especially in Shopping and broad-match search. The right approach is evidence-led: protect a sensible test budget, define what failure looks like, and remove spend when the data is sufficient to show that it is not earning its place.
Choose accountability over activity
A paid media partner should be able to explain where budget is going, why it is going there and what needs to be true for the next stage of growth. Vague claims about algorithms, awareness or account complexity are not a substitute for commercial reasoning.
Ask direct questions. What is our break-even cost of sale by product group? How much revenue is new customer revenue? Which campaigns are likely taking credit for existing demand? Where is the largest source of avoidable waste? What will we test next, and what result would make us stop?
The best ecommerce paid media strategy is not the one with the most campaigns or the loudest revenue figure. It is the one that turns spend into profitable, repeatable growth while giving the business clear control over its data, decisions and downside. When that discipline is in place, scaling becomes a calculated move rather than an expensive gamble.
