Bidding Strategies: Manual, Smart Bidding, and When Each Applies — matching the strategy to the conversion data you actually have

verified · provenanceused 0× by assistantsconcept

Google Ads offers a spectrum from full manual control (Manual CPC) to full automation (Target ROAS), and the strategies are not interchangeable at any data level: several of them have explicit minimum-conversion thresholds, and switching triggers a "Learning" period that temporarily degrades performance. Picking the wrong strategy for the conversion volume you actually have is a direct way to burn budget.

The Strategy Families

- Manual CPC: you set your own maximum cost-per-click — a default bid at ad-group level, overridable per keyword or placement. You keep direct control of spend and can allocate more budget to the keywords you've found more profitable. — FONTE: https://support.google.com/google-ads/answer/2390250?hl=en - Enhanced CPC (eCPC): auto-adjusted your manual max CPC up or down based on estimated likelihood to convert. It was discontinued for Search and Display campaigns effective the week of March 31, 2025 — it is no longer a selectable option, and campaigns not manually migrated are now effectively on Manual CPC. — FONTE: https://support.google.com/google-ads/answer/2464964?hl=en - Maximize Clicks: automated; you set a budget and Google Ads sets your max CPC bids to get as many clicks as possible within it — you don't choose bid amounts. Click-focused, not conversion-aware. — FONTE: https://support.google.com/google-ads/answer/6336101?hl=en - Maximize Conversions and Maximize Conversion Value: Smart Bidding strategies that use machine learning to optimize bids at auction-time, driving the most conversions (or the most total conversion value) within your budget. — FONTE: https://support.google.com/google-ads/answer/7065882?hl=en - Target CPA: you set a desired average cost per conversion; Google Ads adjusts bids automatically to hit that target while maximizing conversions. It can be started with no conversion history — it is "effective for campaigns of all sizes." — FONTE: https://support.google.com/google-ads/answer/6268632?hl=en - Target ROAS: predicts the value of a potential conversion at auction-time and sets bids to maximize conversion value while holding a specified return target (e.g., 500% ROAS = $5 in sales per $1 spent). — FONTE: https://support.google.com/google-ads/answer/6268637?hl=en

When to use which

Manual CPC/Maximize Clicks fit accounts with too little conversion data for Smart Bidding to have a signal (see prerequisites below); Maximize Conversions/Value fit accounts optimizing for volume or total value without a strict efficiency ceiling; Target CPA/ROAS fit accounts that need a specific efficiency or ROI target and already meet the conversion thresholds. Conversion tracking must be enabled before any Smart Bidding strategy — see Conversion Tracking Fundamentals — the layer every audit and every bid strategy depends on. — FONTE: https://support.google.com/google-ads/answer/6268632?hl=en

How Smart Bidding Uses Auction-Time Signals

"Smart Bidding" is Google's umbrella term for strategies that use Google AI to optimize for conversions or conversion value "in every auction, a feature known as 'auction-time bidding'" — Target CPA, Target ROAS, Maximize Conversions and Maximize Conversion Value are all Smart Bidding strategies. — FONTE: https://support.google.com/google-ads/answer/7065882?hl=en

The models "train on data at a vast scale" and factor in a wide range of auction-time signals to estimate conversion likelihood per auction, including: device, physical location and location intent, weekday and time of day, remarketing-list membership, ad characteristics, interface language, browser and operating system, the actual search query, search-network partner, web placement, site behavior, product attributes, price competitiveness versus competitors, and seasonality. — FONTE: https://support.google.com/google-ads/answer/7065882?hl=en

Target CPA specifically adjusts bids using real-time signals "like device, browser, location, time of day, remarketing list, and more" to find the optimal bid each time an ad is eligible; Target ROAS uses the same class of real-time signals to bid higher on searches predicted to be high-value and lower on ones predicted to be low-value. — FONTE: https://support.google.com/google-ads/answer/6268632?hl=en, https://support.google.com/google-ads/answer/6268637?hl=en

Practical implication

because Smart Bidding already factors device, location and time-of-day into every auction, manually layering bid adjustments for those same dimensions on top of Smart Bidding is redundant with what the algorithm already does — a trap documented further in What Does NOT Work: Broad Match and Smart Bidding Without Guardrails — where automation burns budget instead of saving it.

Data Prerequisites Before Switching Strategies

Minimum data varies by strategy, and ignoring it produces an unreliable read on performance:

- Target ROAS: has campaign-type-specific minimum conversion thresholds, not a single general figure: Search/Shopping at least 15 conversions in the past 30 days; Display at least 15 conversions with valid conversion values in the past 30 days; App at least 10 conversions every day or 300 in 30 days; Demand Gen at least 50 conversions in the past 35 days (10 in the past 7 days); Hotel at least 50 conversions per week at campaign level. — FONTE: https://support.google.com/google-ads/answer/6268637?hl=en - Target CPA: no minimum — advertisers "can start using Target CPA with no conversion history," and it is "effective for campaigns of all sizes," but conversion tracking must already be set up first. — FONTE: https://support.google.com/google-ads/answer/6268632?hl=en - Before turning on Smart Bidding, wait for a few conversion cycles so the algorithm can adjust; it functions on minimal data but improves with an established conversion history. For long-cycle businesses (car dealerships, travel), the algorithm deliberately weighs recent data less heavily to avoid overreacting to clicks still awaiting conversion. — FONTE: https://support.google.com/google-ads/answer/10970825?hl=en

When it applies

an account below these thresholds should stay on Maximize Clicks or Manual CPC (or Maximize Conversions, which has no stated minimum) until it accumulates enough volume — jumping straight to Target ROAS on thin data is a documented setup for an unreliable, oscillating target. See the budget-to-strategy migration path in How to Set Budgets and Bid Strategies by Funnel Stage (Without Starving Smart Bidding of Data).

The Learning Period

Any new strategy, or any change to an existing strategy's target, puts the affected campaign into a "Learning" status while the algorithm recalibrates. This is triggered by: a new strategy, a bid-strategy setting change, a campaign/keyword composition change, or an ad-group target change (Shopping). — FONTE: https://support.google.com/google-ads/answer/13020501?hl=en

Calibration "can take up to 3 weeks or 1-2 conversion cycles" for the bid strategy to calibrate to the new objective, though it can be faster depending on the amount of conversion data present. Its length depends on three factors: the number of conversions obtained, the length of your conversion cycles, and the bid-strategy type. Conversion data from previous campaigns can drive faster results by speeding up this initial learning period. — FONTE: https://support.google.com/google-ads/answer/13020501?hl=en

Under the hood, when there is little to no conversion data Smart Bidding uses query-level data beyond the account's own bid strategy to build more accurate initial conversion-rate models, then uses Bayesian learning to continuously improve them; it can also use data-driven attribution (DDA) to weigh each keyword's contribution across the conversion path. — FONTE: https://support.google.com/google-ads/answer/10970825?hl=en

Why it matters for auditing

a performance dip right after a strategy or target change is very likely the expected learning-period effect, not a real regression — check the account's Learning status before diagnosing a drop. This is the first thing to rule out in How to Audit an Existing Google Ads Account (Priority Order) — the sequence that stops you from trusting broken data.

Portfolio vs Standard Bid Strategies

A portfolio bid strategy is an AI-powered, goal-driven strategy that groups together multiple campaigns, ad groups and keywords under one shared strategy and target, with Google's AI setting bids across the whole group to hit that goal; it is stored in your Shared library. A standard (single-campaign) bid strategy is applied individually, at the campaign/ad-group/keyword level, with no cross-campaign grouping. — FONTE: https://support.google.com/google-ads/answer/6263072?hl=en

- Portfolio strategies support both Smart Bidding types (Maximize Conversions, Maximize Conversion Value, Target ROAS, Target CPA) and automated non-Smart-Bidding types (Maximize Clicks, Target Impression Share). — FONTE: https://support.google.com/google-ads/answer/6263072?hl=en - They are not available for Performance Max campaigns. — FONTE: https://support.google.com/google-ads/answer/6263072?hl=en, https://support.google.com/google-ads/answer/6263058?hl=en - A portfolio strategy with a shared budget distributes spend across its campaigns and stops one campaign's budget cap from constraining the others; it also gives you a single place in the Shared library to change bidding settings for every campaign attached to it. — FONTE: https://support.google.com/google-ads/answer/6263072?hl=en

When it applies

portfolio strategies pool conversion signal across campaigns, which is useful precisely when a single campaign wouldn't hit the minimum-conversion thresholds above on its own — but never for Performance Max, which cannot use portfolio strategies at all.

Smart Bidding Exploration

Smart Bidding Exploration is a mechanic layered on value-based Smart Bidding (Target ROAS / Maximize Conversion Value) that changes *how* the algorithm looks for conversions. Instead of bidding only where a proven conversion history exists, a flexible ROAS target lets the campaign open itself up to bid on less obvious but potentially valuable queries more often — within your existing targeting — to reach qualified users it wouldn't have captured otherwise. Google calls it "our biggest update to bidding in over a decade." — FONTE: https://blog.google/products/ads-commerce/smart-bidding-exploration-ai/

- What the flexible target does: loosening the ROAS constraint lets the model bid into auctions it would otherwise skip as too uncertain, surfacing *incremental* conversions from a wider range of unique query categories, without changing core campaign targeting. — FONTE: https://blog.google/products/ads-commerce/smart-bidding-exploration-ai/, https://business.google.com/us/accelerate/announcements/bidding-and-budgeting-updates-to-scale-your-growth/ - Launch (Search): introduced for Search campaigns (announced May 21, 2025). Google's reported results were +18% unique search query categories with conversions and +19% conversions (Google internal data, Global, Mar 11–Apr 11 2025); a later figure cites +27% more unique converting users for Search campaigns using it (Google internal data, Jan 1–Feb 20 2026). — FONTE: https://blog.google/products/ads-commerce/smart-bidding-exploration-ai/, https://business.google.com/us/accelerate/announcements/smart-bidding-exploration-expansion/

Extension to Performance Max and Shopping (2026)

Announced for 2026, Smart Bidding Exploration moved beyond Search into Google's most automated, feed-driven surfaces:

- Performance Max, no product feed: Smart Bidding Exploration is expanding to all Performance Max campaigns without a product feed, giving them the same ability to capture new unique users that Search text ads have. — FONTE: https://business.google.com/us/accelerate/announcements/bidding-and-budgeting-updates-to-scale-your-growth/ - Performance Max with product feed + Shopping: a separate beta — for Shopping ads on both Performance Max campaigns with a product feed and standard Shopping campaigns — has been opened. — FONTE: https://business.google.com/us/accelerate/announcements/bidding-and-budgeting-updates-to-scale-your-growth/, https://business.google.com/us/accelerate/announcements/smart-bidding-exploration-expansion/

Why the extension matters

it carries the same explore-into-uncertainty trade-off into feed-driven PMax and Shopping — deliberately bidding on unproven queries to find incremental conversions. That interacts directly with What Does NOT Work: Performance Max Failure Modes: a loosened ROAS target on a feed-driven PMax campaign with thin conversion data can spend into unproven queries faster than the model can confirm they convert. Treat the flexible target as a lever to *widen* reach after the conversion-volume thresholds above are met — not a fix for thin data — and pair it with the brand-cannibalization guardrails in How to Launch Performance Max Without Cannibalizing Brand or Wasting Spend.

Related

- Conversion Tracking Fundamentals — the layer every audit and every bid strategy depends on — every Smart Bidding strategy in this page requires conversion tracking to already be enabled; broken tracking invalidates the strategy choice before it even starts. - How the Google Ads Auction Actually Works — the pricing engine behind every bid decision you'll make — Smart Bidding sets the bid input into the same per-impression auction described there; this page explains how that bid is computed, not what happens to it once submitted. - How to Set Budgets and Bid Strategies by Funnel Stage (Without Starving Smart Bidding of Data) — the intended migration path (Maximize Clicks → Maximize Conversions → Target CPA/ROAS) as an account accumulates the conversion volume this page's thresholds require. - What Does NOT Work: Broad Match and Smart Bidding Without Guardrails — where automation burns budget instead of saving it — what goes wrong when broad match is paired with Smart Bidding below the conversion thresholds documented here, and why repeated target changes keep resetting the Learning period. - What Does NOT Work: Performance Max Failure Modes — where Smart Bidding Exploration's flexible ROAS target can amplify feed-driven PMax failure modes if switched on before the conversion-volume thresholds are met. - How to Launch Performance Max Without Cannibalizing Brand or Wasting Spend — the PMax surface Smart Bidding Exploration now reaches; loosening the ROAS target widens the query set, so the brand-cannibalization guardrails there apply. - How to Audit an Existing Google Ads Account (Priority Order) — the sequence that stops you from trusting broken data — checking bid strategy and Learning status is a required step before attributing a performance change to anything else.

Why this page matters for the wiki's objective

choosing a bid strategy that doesn't match the account's actual conversion volume — or reading a Learning-period dip as a real regression, or opening a ROAS target on thin data — is one of the most direct ways to burn budget covered by this wiki; this page gives the exact thresholds and mechanics needed to set up and audit that choice correctly. FONTI: ["https://support.google.com/google-ads/answer/7065882?hl=en","https://support.google.com/google-ads/answer/2390250?hl=en","https://support.google.com/google-ads/answer/2464964?hl=en","https://support.google.com/google-ads/answer/6336101?hl=en","https://support.google.com/google-ads/answer/6268632?hl=en","https://support.google.com/google-ads/answer/6268637?hl=en","https://support.google.com/google-ads/answer/13020501?hl=en","https://support.google.com/google-ads/answer/10970825?hl=en","https://support.google.com/google-ads/answer/6263072?hl=en","https://support.google.com/google-ads/answer/6263058?hl=en","https://blog.google/products/ads-commerce/smart-bidding-exploration-ai/","https://business.google.com/us/accelerate/announcements/bidding-and-budgeting-updates-to-scale-your-growth/","https://business.google.com/us/accelerate/announcements/smart-bidding-exploration-expansion/"]

Verified against

22 claims checked against these sources · 5 refuted and removed

Source: Sinapsi — verified compositional memory, queryable by LLMs. Query this wiki live from your assistant over MCP, or build your own verified wiki (public, or private for your team). CC BY 4.0 — reuse with attribution to Sinapsi.