Performance Max for an Online Store: Prepare, Launch, and Track Sales

Launch Performance Max after checking three things: products are eligible to advertise in Google Merchant Center, your website reports purchase values correctly, and your store has calculated an affordable advertising spend. For a Ukrainian online store, the practical goal is to generate completed orders whose contribution covers advertising costs and leaves money for the business to operate.

Before launch, agree with your advertising specialist on the product list, delivery areas, budget, and rules for stopping spend. After launch, regularly reconcile ad data against payments, cancellations, and returns in the store’s records. Google automation uses the goals and data you provide, so errors in them need to be corrected immediately.

How Performance Max uses an online store’s products

Performance Max, or PMax, is a Google Ads campaign type that optimizes advertising for selected goals. A single campaign can access ad inventory across Search, YouTube, the Display Network, Discover, Gmail, and Google Maps. Access to these channels does not mean that every product or ad will necessarily appear on every platform.

For a store with product data in Merchant Center, that data forms the basis of its product ads. Text, images, logos, and videos add to the campaign’s options. Google combines the provided assets, settings, and automated bidding to increase conversions or their value. This is described in the Performance Max Help Center guide.

Google lets you create PMax campaigns for sales, leads, website traffic, visits, and promotion of local stores, or without goal guidance. For an online store, it makes sense to start with sales and a clearly defined purchase. Adding a product to a cart or viewing a page can help analyze the customer journey, but these actions do not confirm a completed order.

Step 1. Prepare your product range and check Merchant Center

First, compile a list of products the store is ready to sell: with accurate prices, current stock levels, clear attributes, and working product pages. For the first launch, we recommend choosing products whose availability you can confirm quickly and whose orders you can fulfill.

1. Check account links. Make sure the campaign uses the right Merchant Center account and that Google Ads has access to your store’s catalog. Do not assume that products will appear because they were included in earlier campaigns.
2. Review product status. Separate approved offers from those with issues. Google recommends checking the Product issues column to find missing attributes and other product data problems.
3. Match the data to the product pages. For selected items, compare the title, product variant, price, currency, image, availability, and link. Also check the final cart total after discounts are applied.
4. Check for updates. Change the price or stock level of a test item and confirm that the changes appear on the website and in Merchant Center. Record the actual delay so you can account for it during sales and when selling the last units.
5. Check checkout. Customers should be able to understand where the store delivers, how much delivery costs, and how to complete a purchase on a phone.

Google recommends providing detailed descriptions and high-quality images, updating prices and availability, and checking that products are approved. These recommendations appear in the Merchant Center data guide for PMax. If offers are not approved, address the reasons first: increasing the budget will not fix product data.

Step 2. Check the purchase, value, currency, and order number

Google recommends that retailers report the actual value of each transaction when purchases have different values. Using the same assumed value for every order does not reflect those differences. To optimize for purchase value, the store needs to report each order’s value as agreed with the developer and advertising specialist.

Prepare a short technical brief for your developer. Define when the website records a purchase, what amount it reports, whether it includes delivery and taxes, how it accounts for discounts, and which number identifies the order. If your store accepts cash on delivery, agree separately on how you will reconcile placed orders with orders that were actually paid for and received.

• Value. For each order, check the reported value against the order record itself, not the product price before discounts.
• Currency. Report the currency along with the value. For a purchase in hryvnias, specify the UAH code in the technical brief and verify it in a test event.
• Identifier. Use a unique, stable order number as the transaction ID. Reopening the confirmation page should not create a new purchase.
• Data sources. If purchases reach Google Ads through multiple routes, check that the same transaction is not counted twice among the goals the campaign uses for bidding. Reporting an order number alone does not replace this check.
• Cancellations and returns. Assign someone to reconcile and adjust advertising data using the configured measurement method. Until you have verified the adjustments, account for these amounts separately in the management report.

Practical pre-launch test: place an order, reopen its confirmation page, check what happens after a repeat redirect following payment, and place another order with a different value. Compare order numbers, currencies, values, and purchase counts. Also check a failed payment: it must not appear as a successfully paid purchase if payment is the action you chose as the campaign goal.

Start value-based bidding after the developer and advertising specialist confirm the test results. Google’s recommendation to report the value of individual transactions and use an appropriate bidding strategy appears in the tips for retail advertisers.

Step 3. Calculate affordable advertising spend

ROAS, or return on ad spend: the conversion value attributed to advertising in the report, divided by advertising costs. For example, UAH 100,000 in purchase value from UAH 20,000 in spend gives a ROAS of 5, or 500%. If the reported value is revenue, this metric shows the ratio of revenue to advertising. Assessing profit also requires the store’s costs.

We recommend calculating the contribution from completed orders toward covering costs: revenue after returns, minus the cost of goods sold and variable costs for those orders. Include payment fees, packaging, delivery paid for by the store, and costs from shipments that are not collected. After advertising costs are deducted, the remainder should cover fixed costs and generate profit.

Illustrative break-even ROAS example

Assume that, for this calculation, the store is not a value-added tax payer. All amounts use the same basis: hryvnias actually received and paid. After cancellations and returns, completed orders generated UAH 100,000 in revenue. The cost of goods sold is UAH 60,000; payment fees, packaging, and store-paid delivery total UAH 10,000; and other variable costs and turnover taxes under the assumptions in this example total UAH 5,000.

Under these assumptions, a 400% ROAS uses the entire contribution on advertising. Nothing is left for rent, fixed salaries, or profit. With UAH 20,000 in spend and a 500% ROAS, UAH 5,000 remains before fixed costs. So set a working target that accounts for those costs and the cash buffer you need.

You can compare this threshold with Google Ads only when the data uses a consistent basis. If the ad report includes placed orders that were later canceled, its ROAS will be overstated compared with completed sales. Likewise, you cannot compare revenue that includes value-added tax with costs that already exclude recoverable tax. Agree on the calculation basis with your accountant.

Plan cash flow separately. With cash on delivery, advertising costs may arise before the store receives payment for the goods. Even an acceptable revenue-to-cost ratio requires a reserve for stock purchases, delivery, and current payments.

Step 4. Create a clear campaign structure

Google recommends starting with a simple structure. Separate campaigns may be needed for different budgets, countries, languages, or ROAS targets. A practical approach for a store is to group products with similar economics and sales conditions, and split them when that allows you to make separate budget decisions.

An asset group brings together text, images, logos, and videos around a particular theme. For example, prepare different assets and relevant landing pages for pots and tableware. Use listing groups to select catalog products that belong to the relevant asset group. Check exactly which products are included before launch.

You can prepare custom labels to select products, for example by seasonality or commercial priority. Google describes the use of these labels and themed groups in its recommendations for campaigns with a product catalog. A group’s name does not determine its product range, so do not select all products if you plan to advertise only one category.

Check that all assets match the landing pages: customers should see the products and terms advertised on the page they visit. If you do not have your own video, Google may automatically create one from available assets. Review any such videos, especially after changing your product range or when a promotion ends.

Step 5. Set up audiences, location, and landing pages

Provide useful audience signals

Audience signals help Google understand more quickly who your store considers a potential buyer. You can use relevant visitor and customer lists, as well as custom segments. Provide your customer data only when you have the required consent and in accordance with the terms for using it.

An audience signal does not restrict ads to list members and does not guarantee that specific people will be reached. Google describes both provided segments and new segments identified by the system in its recommendations on audiences and performance analysis. In practice, select lists based on their relevance to the product; you do not need an arbitrary visitor-count threshold.

Match ad delivery to your delivery areas and service languages

For sales in Ukraine, define your location targeting based on where you can actually accept and deliver orders. Check both the selected areas and the location settings to make sure they match your sales plan. Align languages with your ads, landing pages, and customer support capabilities. After launch, review results by location and reasons for delivery refusals.

Check Final URL expansion

When Final URL expansion is enabled, Google may replace the specified page with another relevant page on your website. The text customization setting uses website content and existing ads to automatically create additional text. The advertiser is responsible for the accuracy of landing pages and generated assets.

Google recommends these features to help find additional sales. Before enabling them, check your website and decide which pages are suitable for ad traffic. If needed, use URL exclusions or rules based on URL parameters. For example, consider excluding job listings, administrative sections, and expired promotions. The feature is described in the PMax setup guide.

Separate unwanted queries from branded demand

Google documents negative keywords and brand exclusions as PMax controls; its retail recommendations also mention account-level negative keywords and campaign-level brand exclusions. Check the available settings and their scope before applying them: account-level exclusions need to be assessed for their effect on other advertising.

We recommend regularly reviewing the available search query data and adding well-founded exclusions. Make a separate decision about your own brand: analyze sales from searches for the store name with the understanding that the customer may already have known about the store. Turning off automatic text creation does not replace managing branded queries.

Step 6. Agree on the budget and launch a checked campaign

If your store reports purchase values correctly, Google recommends the “Maximize conversion value” strategy. You can set a target ROAS for it. Set the target based on product economics and reliable sales history; there is no universal rule to leave it blank for a first launch.

Agree on a budget with the owner as an amount the business can spend to evaluate the campaign without disrupting current payments. Google’s recommendations and forecasts can help with planning, but the decision should account for stock levels, order processing capacity, and cash reserves. Do not invent a mandatory sales threshold or a single minimum budget for every store.

Before enabling the campaign, the responsible specialist should confirm that selected products are approved, the purchase and its value are reported correctly, there is no duplicate counting, location targeting is correct, product groups are configured as intended, and landing pages and ad assets are suitable. The owner agrees to the affordable spend and the conditions for intervention.

How to check Performance Max over six weeks

Google recommends giving a campaign at least six weeks to collect data and compare results, and avoiding frequent changes. This is a recommendation for the system’s learning period. Measurement errors, an unavailable website, incorrect prices, or clearly unacceptable costs must be addressed immediately. The recommendation appears in the tips for all business types.

The schedule below is a practical plan for the owner and specialist. It helps you check the campaign regularly without making unjustified daily adjustments.

Keep one weekly table with advertising costs, purchase value in Google Ads, completed orders, revenue after returns, cost of goods sold, variable costs, and contribution after advertising. Mark recent orders whose payment or return status could still change the result. Compare periods with the same observation window.

Before increasing the budget, check that you have enough products and cash to replenish stock. Record the date and reason for any significant change so you can assess its impact later. Draw conclusions cautiously when there are few completed orders; the learning period does not automatically generate enough data.

What to consider when assessing advertising sales

Sales attributed to ads by Google Ads depend on the measurement setup and attribution rules, which assign value across ad interactions. Such a report does not prove that these sales would not have happened without PMax. Assessing incremental impact requires a separate comparison with a reasonable scenario without the campaign.

In practice, analyze ad reports alongside the store’s total completed sales, other advertising costs, promotions, and product range changes. If ROAS rises while the store’s contribution after all advertising costs falls, find out why before scaling up. Check possible explanations against the orders: a change in product mix, larger discounts, returns, or increased fulfillment costs.

How to prepare stock and orders for advertising in Torgsoft

If your store keeps its records in Torgsoft, check the selected product range before launch using the “Стан складу” mode (Ukrainian for “Stock Status”). It shows quantities, selling prices, cost prices, sizes, and other product data. Filters help find items running low, products without photos, or items discounted below cost. Use this check to select products for advertising and calculate affordable spend from current records. These features are described in the “Стан складу” help page.

The paid add-on feature “Синхронізація з Інтернет-магазином” (Ukrainian for “Online Store Synchronization”) lets you send product data to the website and receive orders for further processing; Torgsoft must be the primary record-keeping system, and connecting the website requires its developer’s involvement. Agree on the source of prices, warehouses, and the data to be sent. According to the documentation, setting transmission to include only available products removes items with zero or negative quantities from the website, so first check whether this behavior suits your catalog. Terms are described in the synchronization help page.

For scheduled automatic synchronization, Torgsoft must be running on a designated computer under the specified user account; the documentation sets an interval no more frequent than once every 10 minutes, taking other synchronization objects into account. Monitor the log and confirm that changes reach the website, then Merchant Center through a separately configured data feed. To reconcile ads with sales, compare order numbers, payments, and fulfillment: the “Задоволено” status described in the help page (Ukrainian for “Satisfied”) means that an invoice has been created, so by itself it does not confirm that the customer received the goods or that payment was received.