Why Performance Max is No Longer the Black Box It Once Was

Key Takeaways

• Performance Max is no longer a black box, given the increased insights it offers about ad placements, search term visibility, and channel-level performance.
• The tool has evolved from one that provides minimal insights to one that can give you more data about who is being targeted and why, which gives more transparency but also requires more management. 
• What matters most now is how you support Performance Max in its ability to target the correct audience members by knowing when to use it and what data you should provide it with.
  

Quick Answer:

While Performance Max used to provide very little information about where ads were displayed and how budgets were allocated, it has improved to give marketers more insights into search terms, ad placements, and channel reports. However, it still poses challenges for marketers by cannibalizing other campaigns or by chasing the easiest, rather than the highest-quality, conversions.

The Evolution of Performance Max

Performance Max has historically been known as a black box because marketers have had limited ability to understand the logic behind who Google targets, when, or why. The tool has long held this reputation for its lack of insight into budget, spending, and campaign management, but it has slowly and slightly become more transparent over time.

While it may no longer be a black box, with increased visibility today, many marketers still maintain unfriendly first impressions of the tool, given its reputation. Still, the value of PMax as a Google Ads campaign type and the sheer number of conversions it drives mean it’s important to understand when to use it compared to other campaign types and how to leverage the insights to drive actual returns.

PMax takes more control away from the user, letting campaigns run mainly on the algorithm, which offers its own pros and cons. However, today the tool offers more insight into search terms, devices, and ad placements than in the past, providing you with greater clarity into performance but also greater responsibility to manage campaigns and make adjustments.

In this guide, I unpack how Performance Max has changed, the idiosyncrasies it still has, when to use it or not use it, and what all this means for how we utilize it.

How Performance Max Works

Performance Max works by gathering information about conversion goals, budget, creative assets, and, if it’s a shopping campaign, a product feed to refine campaigns and push ads to the right audiences. You can also decide to provide PMax with your sales and CRM data as optional signal strength boosters, and the algorithm does almost everything else.

After setup, the tool determines where to surface your ads across Search, YouTube, Gmail, Maps, Display, Shopping, and Discover networks and how much to bid on each space. PMax often takes about four to six weeks to learn and understand your audiences and exit the learning phase for better ad placements and results.

Historically, Performance Max has been largely hands-off, which has its pros and cons. Instead of being able to choose where to spend, the algorithm decides for you based on what it learns, compared to standard shopping campaigns, which require more manual setup.

Ultimately, Performance Max works best when you can identify what campaigns are most valuable to use them for. In my experience, I consider these two sides of the coin:

• Beneficial: When you have clear audience signals, like in ecommerce, and conversion events are sales, or when you want to target across multiple touchpoints.
• Not Beneficial: When you are aiming for lead generation or similar types of campaigns that have minimal supporting data or conversion insights behind them.

What Made Performance Max a Black Box?

Performance Max’s reputation as a black box came from its lack of transparency around how campaigns are distributed and where money is spent. Instead of clear breakdowns and reports, you were left to trust the will of the algorithm without question.

Marketers have faced the following challenges consistently:

• Lack of visibility into budget allocation
• No control over ad placements
• Unavailable breakdowns of ad formats in reporting
• PMax works off of signals, so you can’t explicitly target a top-performing audience. Instead, Google will treat that “target audience” as a signal and look for people who are similar but not explicitly on that list.

All things considered, marketers had very little control and insight into the process, instead having to let Performance Max run its course. Relinquishing all this trust also comes with the fear of wasting budget on little return from actually beneficial leads and conversions. Although some of these fears have left a long-term mark on PMax’s reputation, the most recent changes may have provided more room for comfort.

Is Performance Max a Black Box Still?

Compared to a few years ago, Performance Max today is much clearer, offering more transparency and control. A greater availability of data and increased control over what elements of a campaign you can adapt mean that some of the biggest concerns with PMax have been addressed.

Within the past two years, it has offered the following improvements for clarity:

• Search term visibility – The tool allows you to see which keyword terms it is using to reach new audiences, giving you the ability to see what’s working and build in exclusions and negative keywords.
• Channel-level reporting – Google allows you to see reporting splits between Shopping, Video, Display, Discover, Gmail, Maps, and Search to understand how costs, clicks, and impressions are performing.
• Granular placement reports – Google now shows you where ads are being placed, giving you more power to identify waste and implement exclusions on where your brand is being displayed for better performance potential.

Then vs. Now

While PMax in the past offered very little insight into ad placements, budget allocation, and performance, it grants access to a lot more information and control today, offering you more capability in the following areas:

• You can input your goals into Google, whether it’s sales, leads, website traffic, or local store visits.
• You provide Google with the campaign information and creatives for it to use and display.
• You allow Google to decide where and when to display ads based on the algorithm’s insights and signals.

The added changes have opened up a middle ground between expanding reach with less oversight and still giving you the chance to make refinements for better performance. That said, the limited reporting windows and lack of by-channel breakdowns have left more to be desired.

A Word of Caution

The added improvements don’t necessarily mean Performance Max today has solved all its problems. One of the main lasting historical concerns is not so much that it doesn’t provide much data or signals, but that marketers don’t always trust where their money is going.

Overall, there is still an overarching critique that PMax may not target the highest quality leads, often jaded by its own goal of making conversion data look good. While it may in fact drive conversions, it sometimes drives conversions for low-revenue inventory or overlooks high-ROAS opportunities. In some cases, it can feel as though you’re relying on Google to just fill ad space without the actual best interests of your brand in mind.

Similarly, another major remaining gripe with the platform is the concern that PMax goes after warm traffic, likely cannibalizing performance from other campaigns. I’ve had colleagues notice that after launching PMax campaigns, their net sales/leads remain unchanged, while PMax captures conversions that would’ve happened anyway, requiring a mindful approach to targeting.

What This Means for You

More data and insights about PMax logic and performance place more control into your hands as a marketer, and given the lingering hesitations about ad quality, it’s important that you are equipped to feed the algorithm what it needs to succeed. From being mindful of keyword exclusions to offering many creative assets, knowing how to play the game is crucial for PMax success. 

Maximize campaign potential by doing the following:

• Identifying which keywords and platforms are worth excluding 
• Providing a high volume of creative assets, headlines, landing pages, and images to test
• Know when to avoid using PMax altogether and stick to standard shopping campaigns for your goals
• Upload first-party data of past buyers to provide the strongest signals to the algorithm
• Know when to consolidate or segment your Google Ads campaigns based on intent
• Prepare to counteract the risk of cannibalizing your own campaigns by providing keyword exclusions to keep your shopping campaigns running strong

When It’s Worth It To Use PMax

While Performance Max can offer a lot of convenience and reach without added burden, it isn’t the right answer to every scenario. I find it’s worth it to use PMax when:

• You have strong historical conversion data, such as in the eCommerce industry
• You want to run ads across all Google networks
• You have a reasonable budget to support machine learning
• You can reasonably allow for a two-week formal learning phase and an additional two-week period before the data is stable enough to evaluate

While this is not an exhaustive or definitive checklist, these are important considerations. However, if you have low monthly conversion volume, a low budget, or you need more granular control over keyword targeting, a standard shopping campaign will likely be a better fit. 

How We Use Performance Max

Typically, we use PMax for our eCommerce clients because it’s the easiest channel for finding conversion signals and provides Google with enough information for accurate targeting. Often, we use a balance between PMax and standard shopping campaigns to refine for the best signals. 

Our process usually starts with creating a catch-all shopping campaign with a low manual bid or a high ROAS target. This does two things:

1. Keeps PMax focused on top performers while giving every product visibility 
2. A restrictive bid on the catch-all guarantees we are not overpaying for the products it covers.

We also reserve Performance Max for scenarios where we can be confident we won’t cannibalize our own campaigns. The ongoing updates have given more control to avoid this, so we are careful to use the proper exclusions to prevent PMax from bidding on our own brand terms or keywords we are targeting through shopping.

Case Scenario:  For an eCommerce client, we tested different asset group signals within PMax and paused poor performers, allowing the best signals and audiences to remain active while staying visible via a catch-all shopping campaign. This allowed us to allocate more funds to top performers while also improving Click-Through Rates through stronger creative-to-product relevance.

February 15th to July 19th 2025 vs. February 15th to July 19th 2026 YoY

	Increase	Percent Change
CTR	1.6% to  2%	25%
Purchases	+209	43%
Clicks	+119,997	93%
Impressions	+3,301,917	53%
Cost	+ $55,335	92%
Revenue	+ $853,853	88%

Pull the Right Conversion Levers with collystring

PMax can be a challenging formula to figure out, but the increase in information it provides to marketers is a breath of fresh air. It will also be interesting to see how much additional functionality and reporting come from PMax in the future. Ultimately, the best approach is to balance shopping campaigns with Performance Max, using a catch-all shopping campaign and keeping PMax focused on top performers. 

Our Paid Search team is consistently testing and refining campaigns to figure out what signals work, how to adapt audiences, and how to better allocate your budget. We’ll guide you through the best strategy for your brand, blending expertise across shopping campaigns, display ads, and PMax campaigns to determine the optimal approach.

Schedule a call with collystring to learn more about how to improve your Google Ads.