PMax Reporting: Getting Product-Level Data Out

Google's stated rationale for Performance Max's aggregated reporting is rooted in the philosophy of "automation-first." According to Google Ads Help documentation, by hiding granular product, placement, and query data, the system prevents advertisers from making manual optimisation decisions that might interfere with the machine learning's ability to find conversions. In their view, transparency leads to "over-optimisation" - a phenomenon where advertisers cut spend on placements that appear inefficient on a last-click basis but are actually providing valuable top-of-funnel signals.

The practical reality is more cynical. PMax reporting hides the very data an advertiser would need to challenge Google's placement mix. For example, if PMax is spending 40% of its budget on junk mobile app Display placements while reporting a high aggregate ROAS (often inflated by brand search), the advertiser would naturally want to exclude those placements. By hiding the split, Google maintains spend volume that might otherwise be cut, ensuring higher platform utilization regardless of individual placement quality.

This lack of transparency disrupts the traditional advertiser-platform relationship. In Search or Standard Shopping, you see every penny's destination. In PMax, you see the result, but the journey is obscured. This is why our attribution deep-dive is so critical for PMax - if you can't see the placements, you must at least be certain about the incrementality of the results. We have seen accounts where the "High ROAS" PMax campaign was simply claiming credit for sales that would have happened anyway through organic search or direct traffic.

Furthermore, Google's automation often prioritizes "low-hanging fruit." Without granular reporting, it's impossible to tell if PMax is finding new customers or simply following existing ones around the web with retargeting ads. This "circular logic" of reporting can make a campaign look incredibly successful on paper while doing very little to actually grow the business.

The Search Insights panel is the only granular data source built directly into the Google Ads web UI. It's often ignored because it's buried under the "Insights" tab rather than the "Keywords" tab where advertisers are trained to look.

Navigate to PMax Campaign > Insights > Search categories. Here, Google groups queries into "Search Themes." For a luxury watch retailer, themes might include "men's automatic watches" or "Rolex Submariner prices."

What it shows: Theme-level clicks, conversion value, and search volume growth. It's excellent for identifying intent drift - when the campaign starts serving for categories that are too broad or irrelevant.

The big limitation: You cannot see individual queries. If a theme is "watches," you don't know if it's "cheap plastic watches" (bad) or "luxury swiss watches" (good). You also cannot export this data easily to a CSV for long-term tracking without using third-party scrapers or API calls.

Search Insights also provides a "Search Category" view that helps you understand the broader market trends. If a particular category is growing in volume but your conversion rate is dropping, it's a signal that the competition is heating up or your landing page isn't meeting the intent of that new traffic. It's the closest thing we have to a "Search Query Report" in PMax, and while imperfect, it's the first place we look during an audit.

This is the single most important "hack" for PMax reporting. Google Ads Editor (the free desktop application) is built on a slightly different API layer than the web interface. For reasons Google hasn't officially explained, the Editor surfaces a Products tab for Performance Max campaigns that the web UI does not.

By downloading your account into Editor, you can see every SKU in your PMax campaign along with its individual clicks, cost, and conversions. This is the only way to perform a "Zombie SKU" audit without writing code.

The Editor Workflow:

We use this data to feed our custom label strategy. If Editor shows a product is high-spend but low-return within PMax, we move it to a 'Catch-all' Standard Shopping campaign or exclude it entirely. This "SKU-level pruning" can often save 15-20% of wasted spend within the first month of management.

Contrarian point: Many "experts" say you shouldn't touch product-level bids in PMax. They are right that you can't set individual bids, but they are wrong that you shouldn't manage the mix. Excluding a failing SKU isn't "meddling"; it's proper budget management. Google Ads Editor is the only free tool that lets you see which products are actually failing.

For advertisers managing more than £10k/month in PMax spend, manual Editor exports are too slow. This is where Google Ads Scripts come in. Scripts allow you to query the Google Ads API directly and dump the data into a Google Sheet on a schedule.

There are two primary resources we recommend for this:

• The Shopping Performance View: This allows you to pull SKU-level metrics filtered by advertising_channel_type = 'PERFORMANCE_MAX'. This is the programmatic version of the Editor hack.
• The Placement Report: While Google doesn't show placement performance directly, you can calculate 'Other' spend by subtracting Search and Shopping spend from the Campaign Total. This reveals the "Black Box" of Display and YouTube spend.

/**
 * PMax Product Performance Extractor
 * Extracts cost, clicks, and conversion value per Item ID
 */
function main() {
  const spreadsheetUrl = 'YOUR_SPREADSHEET_URL';
  const sheet = SpreadsheetApp.openByUrl(spreadsheetUrl).getActiveSheet();
  sheet.clear();
  sheet.appendRow(['Item ID', 'Product Title', 'Cost', 'Clicks', 'Conv Value', 'ROAS']);

  const query = `
    SELECT
      segments.product_item_id,
      segments.product_title,
      metrics.cost_micros,
      metrics.clicks,
      metrics.conversions_value
    FROM shopping_performance_view
    WHERE segments.ad_network_type = 'PERFORMANCE_MAX'
      AND segments.date DURING LAST_30_DAYS
  `;

  const report = AdsApp.search(query);
  while (report.hasNext()) {
    const row = report.next();
    const cost = row.metrics.costMicros / 1000000;
    const roas = cost > 0 ? row.metrics.conversionsValue / cost : 0;
    sheet.appendRow([
      row.segments.productItemId,
      row.segments.productTitle,
      cost,
      row.metrics.clicks,
      row.metrics.conversionsValue,
      roas
    ]);
  }
}

/**
 * PMax Placement Breakdown (Search vs Everything Else)
 * Calculates the 'Invisible' spend
 */
function main() {
  const query = `
    SELECT
      campaign.name,
      metrics.cost_micros,
      segments.ad_network_type
    FROM campaign
    WHERE campaign.advertising_channel_type = 'PERFORMANCE_MAX'
      AND segments.date DURING LAST_30_DAYS
  `;
  // Iterate and output to Logger or Sheet...
  // Search Network matches cost in Search Insights
  // Cross-network is the 'Black Box' (Display, YouTube, Discover)
}

Implementing these scripts requires basic familiarity with JavaScript, but the templates above are "plug-and-play." Simply navigate to Tools > Bulk Actions > Scripts, click the plus icon, paste the code, and authorise.

Once the data is in Google Sheets, you can build a dashboard that updates every morning. We recommend setting up "Alerting" within the sheet - for example, highlight any product that has spent more than £50 in the last 7 days with zero conversions. This allows you to react to inefficiencies much faster than a weekly manual check would allow.

If you're managing multiple accounts or need client-ready visualisations, third-party PPC management tools provide a "polished" version of the API data. They don't have access to "secret" data that isn't in the API, but they excel at making that data actionable through better UI and automated alerting.

For smaller brands, we usually recommend starting with Google Ads Editor and a basic script. The £200-£500/month cost of these tools is better spent on actual ad budget until you reach a scale where manual auditing becomes the bottleneck. However, once you are managing multiple accounts or spending over £50k/month, the time savings alone justify the cost.

Reporting is only valuable if it leads to action. We recommend a 30-minute weekly "PMax Hygiene Check" to ensure the automation hasn't drifted off course. This workflow ensures you're looking at the data Google hides, not just the dashboard they want you to see.

In our audits of UK eCommerce accounts, we see the same three reporting mistakes repeatedly. Each one stems from trusting the PMax dashboard at face value rather than digging into the API-level data.

• 1
  Confusing 'Asset Group' ROAS with Product Performance

  An asset group might contain 100 products. One product could be delivering a 10.0 ROAS while 99 others are at 0.5 ROAS. The asset group will report a 'Good' 4.0 ROAS, hiding the massive waste. Always verify SKU-level data via Editor.

• 2
  Ignoring the 'Search vs Shopping' Split

  PMax often shifts budget to 'Display' or 'Video' placements when Shopping performance dips. If you don't use a script to monitor the 'unaccounted' spend, you might not realise your 'Shopping' campaign has become a 'YouTube' campaign overnight.

• 3
  Over-valuing 'View-Through' Conversions

  PMax is aggressive with Display and Video. If you report on 'Conversions' rather than 'Last-Click' or 'Data-Driven' specifically, you are likely over-counting the impact of PMax's passive placements. Use the Attribution reports to see how often PMax is merely a "looker" rather than a "doer."

• 4
  Lack of Brand Exclusion Monitoring

  Without Search Insights, many advertisers don't realize that 60%+ of their PMax conversions are just their own brand name. This leads to a false sense of security and over-investment in a campaign that isn't actually driving incremental growth.