An ecommerce PPC agency runs paid ads that sell products, which is a different job from running ads that collect leads. The work is dominated by your product feed, by Shopping and Performance Max campaigns, and by return on ad spend against margin rather than cost per lead. Before you hire one, ask how they manage the feed. That answer tells you almost everything.
Most agencies say they do ecommerce. Far fewer are actually good at it, because the skills do not transfer cleanly. A lead gen specialist is excellent at keywords, negatives and form conversion rate. None of those are the main lever when you sell 3,000 products.
Here is what genuinely differs, what to ask on the call, and how to tell a real ecommerce team from a generalist who runs one Shopping campaign.
How is ecommerce PPC different from lead gen PPC?
The short version: in lead gen you buy keywords, in ecommerce you feed a catalog. Google decides which of your products to show for which searches, based mostly on the data you send it.
That single difference changes the metric you optimize, the campaign types you use, and what breaks when things go wrong.
| Lead gen PPC | Ecommerce PPC | |
|---|---|---|
| Goal metric | Cost per lead, then close rate | ROAS, then profit after cost of goods |
| Main campaign types | Search, Local Services, Meta lead forms | Shopping, Performance Max, Search, remarketing |
| What you control | Keywords and ad copy | Product data, prices, images, campaign segmentation |
| The daily work | Search terms, negatives, ad tests | Feed errors, disapprovals, SKU level performance, stock |
| Creative unit | Headline and description | Product title, image, price and reviews |
| What breaks first | Conversion tracking or the landing page | The feed. Disapprovals, missing GTINs, stale prices |
| Seasonality | Mild for most categories | Severe. Q4 can be several times normal volume |
What this means for you: if an agency’s ecommerce plan is mostly about keywords and ad copy, they are running a lead gen playbook on a store. Ask what they will do to your product data in the first two weeks.
Why does the product feed matter more than bidding?
Because the feed is what decides whether you show up at all. Google matches searches to your products using titles, descriptions, attributes and categories. Bad data means your product never enters the auction, and no bid adjustment fixes that.
Google lays out the requirements in its Shopping ads documentation, and the practical work looks like this:
- Titles rewritten to match how people search. “Model 4B Navy” is useless. “Navy Merino Wool Crew Neck Sweater, Men’s, Model 4B” wins searches.
- Required attributes filled in. GTIN, brand, condition, availability, size, color. Missing fields cause disapprovals and weak matching.
- Images that meet the rules and look good small. Most Shopping impressions are a thumbnail. Busy lifestyle shots lose to clean product shots.
- Prices and stock synced often. A price mismatch between feed and site gets items disapproved.
- Custom labels for segmentation. Margin band, season, bestseller, clearance. Without these you cannot bid differently on a 60% margin item and a 6% one.
That last point is the one most generalists skip. Custom labels are how an ecommerce team splits campaigns by profitability instead of treating every SKU the same. We cover the setup in our Google Shopping ads guide.
Feed quality beats bid tweaking. If you only get one thing right, get the feed right.
What should you ask an ecommerce PPC agency?
Five questions separate specialists from generalists fast. Listen for specifics, not enthusiasm.
How do you manage the product feed?
Good answer: they name a tool or a process, they talk about title structure, custom labels and a schedule for checking disapprovals. Weak answer: “we connect Shopify to Merchant Center.” That is the starting line, not the work.
What do you do with poor performing SKUs?
Good answer: they separate them rather than delete them. Products with clicks and no sales get pulled into their own campaign with a lower target, or excluded, or tested with a new title and image first. Weak answer: “we optimize them.” Ask what that means.
Do you optimize to revenue or to profit?
Good answer: profit, and they ask for your cost of goods so they can send margin data instead of revenue. This is the single biggest tell. Revenue ROAS looks great while you lose money on discounted stock. If margin data is new to you, start with what ROAS actually means and then set your floor with break even ROAS.
How do you handle new product launches?
Good answer: new items get their own campaign or asset group so they can gather data instead of being buried by proven bestsellers. Automated bidding always favors what already has history, so new SKUs need protection or they never get impressions.
What is your approach to Performance Max asset groups?
Good answer: multiple asset groups split by product category or margin, real creative in each, and listing group exclusions used to control what feeds where. Weak answer: one asset group holding the whole catalog. Google’s Performance Max documentation explains how asset groups and listing groups work, and our Performance Max guide covers how to keep control of a campaign type that hides a lot.
What does ecommerce PPC actually cost?
Ad costs vary by category, and there is less reliable public ecommerce benchmark data than there is for lead gen. What we can say is that the all industry Google Ads Search averages in 2026 were $5.42 per click and an 8.18% conversion rate, based on WordStream’s 2026 benchmarks across 13,474 US search campaigns.
Retail categories generally sit below that conversion rate, because buying a product outright is a bigger ask than filling in a form. In the same data set, furniture converted at 2.99%. Use that as a reminder that a 2% to 3% site conversion rate can still be a healthy ecommerce account, and that judging a store against a service business average will mislead you.
The number that matters to you is not cost per click. It is whether ROAS clears your break even point after cost of goods, shipping and returns.
How do you tell if an agency is really an ecommerce team?
Look at what they ask you for. An ecommerce specialist asks for your margin by product category, your average order value, your return rate and your stock feed before they talk about budget.
A few more signals worth checking:
- They ask about inventory. Ads that push out of stock items burn budget and annoy customers. Good teams pause on low stock automatically.
- They talk about Merchant Center, not just Google Ads. Half of ecommerce problems live in Merchant Center diagnostics.
- They plan around your calendar. Sales, launches and Q4 change strategy weekly, not quarterly.
- They mention returns. A category with a 30% return rate needs a higher ROAS target than one with 3%.
- They separate new customers from repeat buyers. Paying full price to reacquire someone who would have bought anyway is a quiet way to waste money.
If none of that comes up, you are talking to a search generalist. That may still be fine for a small catalog, but price it accordingly.
What should the first 90 days look like?
Month one is a feed and tracking clean up plus a rebuild of campaign structure by margin. Nobody should be chasing a ROAS target while half the catalog is disapproved.
Months two and three are where segmentation earns its keep: bestsellers separated from long tail, new products protected, poor performers isolated, and Performance Max asset groups split so you can see what is actually working. Expect ROAS to move in month two and to stabilize by month three.
Ongoing, the rhythm is weekly. Feed errors, stock changes, SKU level performance, and seasonal shifts do not wait for a monthly call. Our ecommerce PPC strategy guide goes through the campaign structure in more detail.
Common questions
Is Performance Max enough on its own for ecommerce?
Usually not. It works well once you have conversion history, but it hides a lot and it favors products that already sell. Most healthy accounts pair it with standard Shopping or a Search campaign so new and slow moving products still get traffic.
What is a good ROAS for ecommerce?
There is no universal number, because it depends entirely on your margin. A store with 70% margins can profit at a 2x ROAS. A store with 20% margins cannot. Work out your break even point first, then set a target above it.
How big does my catalog need to be for Shopping ads?
Shopping works with a handful of products. The difference is that small catalogs need tighter control, since a few SKUs will take almost all the spend. Large catalogs need segmentation so bestsellers do not starve everything else.
Should an ecommerce PPC agency also run Meta ads?
Often yes, because product ads on Meta share the same catalog feed. The advantage of one team is that the feed gets fixed once and both channels benefit. The risk is a team that is strong on one platform and thin on the other, so ask about both.
Who should own my Merchant Center and ad accounts?
You should. Agencies can be given access, but the accounts, the feed and the historical data need to stay in your name. If an agency will not agree to that, walk away.
How long does it take to fix a bad product feed?
Basic disapprovals and missing attributes can be cleared in a couple of weeks. Rewriting titles across a large catalog takes longer and is usually done in priority order, starting with the products that already earn the most revenue.
Want a straight read on your store’s ads?
If your ROAS looks fine but your bank account does not, the gap is usually margin data, feed quality or Performance Max eating credit for sales you would have made anyway. Our ecommerce advertising team can look at your feed, your campaign structure and your product level numbers and tell you which one is costing you most.
No pitch and no commitment. Get in touch and we will show you what we find.
