Google Smart Bidding Exploration & Promotion Mode: What Changed June 2026

Google Smart Bidding Exploration is no longer optional context — it's becoming the default behavior across Performance Max campaigns. On June 16, Google announced three bidding and budgeting changes that shift how much autonomy its algorithms have over your spend. If you manage campaigns with tight efficiency targets, every one of these changes demands a response before they roll out.

Here's what changed, what it actually means for your campaigns, and what to do about it.

Smart Bidding Exploration: Google Wants to Bid on Queries It Can't Prove Will Convert

Smart Bidding Exploration expands the pool of queries Google will bid on by including searches with limited or no conversion history. Previously available as an opt-in for Search campaigns, it's now rolling out to all feedless Performance Max campaigns globally, with Shopping-based PMax entering beta.

Google's headline number: advertisers in early tests saw 27% more unique converting users when Exploration was enabled. That's a compelling stat — but it's Google's own data from controlled rollouts, and "unique converting users" is not the same as "more revenue at the same efficiency." The distinction matters.

What Exploration actually does is loosen the bidding algorithm's constraint set. Standard Smart Bidding relies on historical conversion data to predict which queries will convert. Exploration removes that requirement for a subset of auctions, letting Google bid on queries where it has a directional signal (user behavior patterns, contextual relevance) but no proven conversion track record.

The practical risk: your campaigns will enter auctions they previously avoided. For some advertisers, that surfaces genuine demand they were missing. For others, it's incremental spend on queries that look plausible to an algorithm but don't convert at your target efficiency.

What this means for feedless vs. Shopping PMax

Feedless PMax campaigns (lead gen, service-based) get Exploration globally. If you're running PMax for lead generation, expect your query coverage to expand whether you planned for it or not. Shopping PMax enters beta, which means you'll likely see an opt-in prompt in the coming weeks.

The critical difference: Shopping campaigns have product feed data as a conversion signal anchor. Feedless campaigns don't. Exploration on feedless PMax is Google bidding on unproven queries with fewer guardrails. Monitor your search terms reports closely — and if you haven't added negative keywords to your PMax campaigns yet, the recently launched native negative keyword support makes this the time to start.

Promotion Mode: Seasonality Adjustments, Repackaged and Improved

Promotion Mode, now in beta for both Search and Performance Max, lets you schedule temporary bidding adjustments for defined promotional periods — product launches, flash sales, seasonal pushes. You set a date range, increase your budget allocation, and expand your ROAS or CPA tolerance for that window. When the promotion ends, bidding reverts to your standard targets automatically.

If this sounds familiar, it should. Google has offered seasonality adjustments for years. Promotion Mode packages the concept more accessibly and adds explicit budget-increase scheduling, but the underlying mechanism is similar: you're telling the algorithm to expect a temporary shift in conversion rates and giving it permission to bid more aggressively.

The genuine improvement is workflow integration. Previously, running a promotion meant manually adjusting targets, increasing budgets, then remembering to revert everything afterward. Promotion Mode automates the revert, which eliminates the most common failure mode — leaving aggressive targets running two weeks after the sale ended.

Where Promotion Mode actually helps

• Scheduled launches with firm start/end dates. If you know your promotion window, Promotion Mode removes manual steps and prevents the "forgot to revert" problem.
• High-volume seasonal events. Black Friday, Prime Day, back-to-school — events where conversion rate spikes are predictable and temporary.
• Cross-campaign coordination. Applying consistent promotional adjustments across Search and PMax simultaneously, rather than editing campaigns individually.

Where it doesn't help: unpredictable demand surges, always-on discount strategies, or campaigns where you're already manually managing bids. Promotion Mode is a convenience feature for teams running automated bidding at scale. It's not a performance unlock.

Bidding Target Optimization: More Predictable, Potentially Lower Peaks

Rolling out August 17, 2026, Bidding Target Optimization changes how budget-limited campaigns with tCPA or tROAS targets allocate spend. Google's stated goal is "more predictable performance" — smoothing out the volatility that budget-constrained campaigns often experience when Smart Bidding alternates between aggressive and conservative phases.

The honest translation: if your campaigns currently overachieve on your targets (delivering CPAs below target or ROAS above target), this update may reduce that overachievement in favor of more consistent delivery. Google is optimizing for predictability, not peak performance.

For advertisers who've been quietly benefiting from Smart Bidding's tendency to outperform targets on budget-limited campaigns, this is a potential downside. The algorithm will aim closer to your stated target rather than beating it when conditions allow. As Search Engine Journal reported, this trade-off between consistency and peak performance is the core tension in the update.

A related change worth noting: Google is decoupling bidding strategy labels, returning tCPA and tROAS to standalone naming rather than bundling them under "Maximize Conversions" and "Maximize Conversion Value" with optional targets. This is a naming clarification, not a functional change — but it reduces confusion when analyzing strategy performance across accounts.

The Broader Context: PMax and Search Are Converging Fast

These bidding changes don't exist in isolation. Over the past six weeks, Google has also shipped:

• Native asset A/B testing for Performance Max (June 8) — you can now test creative variations within PMax without workarounds.
• Video asset limits increased from 5 to 15 — more room for creative testing in video-heavy PMax campaigns.
• Negative keyword support in PMax — a long-requested control that pairs directly with Exploration's expanded query coverage.
• DSA to AI Max migration timeline extended — the forced migration from DSA to AI Max has been pushed from September 2026 to February 2027, and Google restored the ability to create new DSA campaigns as of June 15.

The pattern is clear: Google is simultaneously loosening algorithmic constraints (Exploration, Promotion Mode) while adding advertiser controls (negative keywords, A/B testing, strategy label clarity). Whether that balance favors advertisers or Google's auction revenue depends entirely on whether you use the controls.

And with AI Mode ads reshaping how search queries surface commercial intent, these bidding changes are also positioning Smart Bidding to operate in contexts where traditional keyword-based conversion prediction doesn't apply.

What to Do Now: A Practical Checklist

1. Audit your PMax campaigns for Exploration status. Check whether Exploration is already enabled on your feedless PMax campaigns. If it is, review the last 14 days of search terms for new query patterns. Set up a weekly monitoring cadence.
2. Add negative keywords to PMax. With Exploration broadening query coverage, negative keywords are no longer optional. Start with your existing Search campaign negative keyword lists and apply them to PMax.
3. Benchmark your current tCPA/tROAS overachievement. Before Bidding Target Optimization rolls out on August 17, document how far your budget-limited campaigns are beating their targets. This gives you a baseline to measure the impact. If you're consistently hitting 20% below target CPA, expect that margin to narrow.
4. Evaluate Promotion Mode for Q3 campaigns. If you have product launches or seasonal events planned for July-September, Promotion Mode beta is worth testing now rather than waiting for GA. The automated revert alone justifies the switch from manual adjustments.
5. Revisit your bidding strategy labels. As Google decouples tCPA and tROAS naming, update your internal reporting dashboards to match. Small change, but it prevents confusion during the transition.
6. Don't ignore the DSA extension. If you were planning to migrate DSA campaigns to AI Max before September, the February 2027 extension gives you more runway. Use it to test AI Max on a subset before forced migration.

The Bottom Line

Google's June 16 bidding overhaul is less revolutionary than the announcement suggests and more consequential than most advertisers will realize until it hits their accounts. Smart Bidding Exploration expands where Google spends your money. Promotion Mode simplifies how you manage temporary pushes. Bidding Target Optimization trades peak performance for consistency.

None of these changes are inherently good or bad. They're shifts in default behavior that reward advertisers who actively manage controls — negative keywords, target calibration, search term monitoring — and penalize those who set targets and walk away. The 27% more unique converting users Google reports from Exploration is achievable, but only if you're filtering out the queries that inflate volume without delivering value.

For campaign managers, the action window is now through August 17. Use it.

Sources: Search Engine Journal, MediaPost, SE Roundtable, Common Thread Collective