Direct answer: DoubleClick Search bid strategies are automated bidding systems — now delivered through Search Ads 360 (the 2018 rebrand) — that use machine learning to set bids across campaigns toward a goal you define: a cost per action, a return on ad spend, an impression share position, or a budget to spend efficiently.
Most run as portfolio strategies: one strategy governing campaigns across different accounts and ad platforms (Google Ads, Microsoft Advertising, Yahoo! JAPAN Ads, Baidu).
Key takeaways
- DoubleClick Search became Search Ads 360 in 2018, and has since been rebuilt into a new-experience platform.
Old docs describing "DS3" screens may no longer apply.
- Choose from your goal type first — visibility, traffic, conversions, or revenue — not from what sounds most advanced.
- SA360 can bid on its own every 6 hours (intraday) or hand the final bid to Google Ads at auction time (auction-time bidding, needs conversion-data sharing).
- Value-based bidding needs value data: Google recommends 4+ weeks of conversion value data for ROAS, longer at low volume.
- Give it time: wait 2–3 conversion-delay cycles before judging; change targets at most once a week.
- Automation doesn't fix broken measurement. Tracking and attribution quality set the ceiling.
DoubleClick Search → Search Ads 360: what changed
DoubleClick Search was Google's enterprise search platform — part of the DoubleClick Digital Marketing suite — sitting above the individual ad engines (Google Ads, Bing/Microsoft, Yahoo!, Baidu) with one big draw: portfolio bidding across all of them at once.
A retailer with hundreds of thousands of keywords across engines couldn't manage bids by hand, and no single engine's automation could optimize across engines.
That premise hasn't changed.
The name, the interface, and the machine learning have.
In 2018, Google folded DoubleClick Digital Marketing and the Analytics 360 Suite into the Google Marketing Platform: DoubleClick Search → Search Ads 360, DoubleClick Campaign Manager → Campaign Manager 360, DoubleClick Bid Manager → Display & Video 360.
Then Google rebuilt SA360 on a new foundation — the new Search Ads 360 experience — with a new interface, account structure, and parts of the bidding/reporting model changed.
Documentation for the legacy platform can describe screens that no longer exist, so always check whether a source means legacy DS or the new experience.
Old name → current equivalent
| If your deck/docs say… | The current equivalent is… |
|---|---|
| DS3 / DoubleClick Search | Search Ads 360 (new experience) |
| Maximise conversions (target CPA) | Target CPA (portfolio or standard) |
| Maximise revenue (target ERS/ROAS) | Target ROAS / Maximize Conversion Value |
| Maximise clicks (monthly spend) | Budget bid strategy: Maximize Clicks |
| Keyword position target | Target Impression Share |
| DS bid strategy (external layer) | Intraday bidding (every 6 hours) — or Google Ads auction-time bidding layered on |
Why the naming matters technically: DoubleClick-era bidding was mostly an external layer — it computed bids and pushed them into the engines on a schedule.
Modern SA360 can still work that way (intraday), but it can also hand the final bid decision to Google Ads at auction time.
That distinction is the single most important technical concept in SA360 bidding today, covered below.
How SA360 automated bidding actually works
It learns from performance history. Strategies set bids using the history of biddable items — keywords, product groups — measured against the campaign's selected conversion goals, including offline conversions.
Budget bid strategies also factor in campaign budgets.
A new strategy can't perform on day one, which is why importing historical conversion data before launch matters more than most teams assume.
Changes are incremental by design. Strategies adjust, watch the effect, then adjust again in small steps — Google states they deliberately avoid large swings because those make outcomes unpredictable.
If you expected aggression on day one, this is why it doesn't happen.
Predictable trends get exploited. A recurring weekend peak, for example, can earn bid or budget adjustments — the stronger and more consistent the pattern, the more confidently the strategy acts.
This needs ample history, another reason to feed strategies volume.
Low-traffic items get a second chance. Long-tail keywords without enough data can drift so low they stop serving.
For intraday strategies, SA360 automatically tries to secure minimal traffic: Google documents that a biddable item with no clicks in 60 days gets daily bid increases until it earns a click — provided the strategy isn't budget-constrained.
Intraday bidding vs Google Ads auction-time bidding
Intraday bidding is SA360's native system: one bid every six hours from portfolio-level modeling.
Google Ads auction-time bidding (ATB) sets a bid at each auction using query-time signals.
You can enable ATB inside an SA360 strategy — SA360 keeps defining the portfolio goal and target while Google Ads sets the individual bid.
Note: bids can only be adjusted for intraday strategies; budgets can be overridden for both.
Enabling ATB requires turning on conversion-data sharing first (at sub-manager level), and it's supported for Target CPA, Target ROAS, Maximize Conversions, and Maximize Conversion Value.
Practitioner recommendation (not official Google guidance): where ATB is supported and data sharing is acceptable, it's generally the stronger default — an auction-time bid sees signals a six-hour refresh can't.
Keep intraday where you need cross-engine portfolio control that ATB doesn't extend to, or where governance restricts data sharing.
For Microsoft Advertising ATB, Google documents a minimum of 30 conversions per campaign in the last 30 days, and only Floodlight and Adobe Analytics conversions are supported.
Every bid strategy: what it does, when to use it, what it needs
Target CPA
Sets bids to win as many conversions as possible at your target cost per action.
Google is explicit: it's an average, not a ceiling — some conversions cost more, some less.
Use when: conversions are worth roughly the same, you have a defensible efficiency target, and cost-per-outcome is the constraint (lead generation is the archetype).
Don't use when: conversion values differ a lot — a B2B advertiser whose enterprise demos are worth 20× a newsletter signup will systematically over-buy the cheap action.
Move to value-based bidding or split the goals.
Target ROAS
Sets bids to win as much conversion value as possible at your target return on ad spend.
Use when: conversions carry meaningfully different values and you're passing those values in reliably — eCommerce with variable baskets is the clearest fit.
Don't use when: values are placeholders, identical for every conversion, or inconsistently populated — ROAS bidding on fabricated values optimizes confidently toward the wrong thing.
Data: Google recommends at least 4 weeks of conversion value data (longer at low volume) for both ATB and intraday ROAS.
Target Impression Share
Bids to place ads at the absolute top, top of page, or anywhere on the results page.
Use when: presence is the objective — brand-term defense, competitive conquest, launch windows.
Don't use when: you're accountable for CPA or ROAS — it buys position, not outcomes, and will spend to hold it.
Always pair with bid limits. No minimum conversion data required.
Budget bid strategies: Maximize Clicks / Maximize Conversions / Maximize Conversion Value
These set both bids and budgets to spend a planned amount as efficiently as possible — they're the plans-layer tool.
Maximize Clicks: most clicks within target spend — for traffic goals and early market tests; never when you have conversion tracking and a performance goal (clicks are a proxy, and proxies invite cheap low-intent traffic).
Maximize Conversions: most conversions within a fixed budget you intend to spend fully — CPA is an output here, not an input, so don't use it with a hard efficiency requirement.
Maximize Conversion Value: most conversion value within a fixed budget — for revenue mandates; needs reliable value data.
Note: campaigns under a budget bid strategy can't be included in Google Ads experiments.
Multiple Target CPA / Multiple Target ROAS
Apply differentiated targets across a group instead of forcing one target on everything.
Use when: segments have genuinely different economics — margins, geographies, lifetime values — and one target would starve profitable segments or overspend weak ones.
Don't use when: your segmentation is arbitrary — differentiated targets fragment data, and fragmented data slows learning.
Each segment needs its own sufficient history.
Manual CPC and Enhanced CPC
Not SA360 portfolio strategies — Google Ads bidding methods for campaigns outside automated strategies.
In modern SA360 deployments they're for narrow cases: brand campaigns on a fixed cost, regulated categories with rigid bid ceilings, diagnostic baselines, or volume too low to model.
Warning: Enhanced CPC is often treated as a safe middle ground.
It isn't a portfolio strategy and delivers none of the cross-campaign optimization that justifies SA360 — running ECPC across an enterprise account usually means paying for SA360 without using it.
Quick comparison
| Strategy | Optimizes for | Best fit | Data guidance | Main risk |
|---|---|---|---|---|
| Target CPA | Conversions at a cost target | Lead gen; equal-value conversions | More volume = more stability; ~20 conv./week for intraday ROI | Over-buying low-value conversions |
| Target ROAS | Conversion value at a return target | eCommerce; variable values | 4+ weeks of value data; longer at low volume | Volume collapse if target too high |
| Target Impression Share | Ad position | Brand defense; visibility | No minimum | Spends for position regardless of ROI |
| Maximize Clicks (budget) | Clicks within spend | Traffic goals; market tests | No minimum | Low-intent traffic |
| Maximize Conversions (budget) | Conversions within budget | Fixed budget, volume mandate | Needs conversion tracking | CPA uncontrolled |
| Maximize Conversion Value (budget) | Value within budget | Fixed budget, revenue mandate | Needs reliable value data | ROAS uncontrolled |
| Multiple Target CPA/ROAS | Differentiated targets | Distinct segment economics | Each segment needs its own data | Data fragmentation |
| Manual CPC / ECPC | Advertiser-set bids | Edge cases | N/A | No portfolio optimization |
Portfolio vs standard bid strategies
A standard strategy applies to one campaign.
A portfolio strategy applies to many at once — across client accounts and ad platforms — optimizing toward a shared target.
Supported: Google Ads, Microsoft Advertising, Yahoo! JAPAN Ads, and Baidu Search; Google Ads and Microsoft Shopping and Performance Max; and Dynamic Search Ads on Google, Microsoft, and Yahoo! JAPAN.
Why portfolios usually win at scale: data pooling (ten campaigns at 8 conversions/week each are ten starved strategies; pooled, they're a learnable signal); budget flows to whatever delivers against the target instead of sitting in silos; one target in one place simplifies governance; and cross-engine consistency has no equivalent inside Google Ads alone.
Keep campaigns out when: economics differ so sharply they'd distort the shared target; the campaign is an isolated test; or budget is legally ring-fenced.
The blended-average trap is the classic error: brand terms at €20 CPA pooled with generic prospecting at €120 can hit a €50 portfolio target while masking that prospecting is wildly inefficient.
The portfolio average looks healthy; the business isn't.
SA360 bidding vs Google Ads Smart Bidding
| Dimension | Google Ads Smart Bidding | Search Ads 360 |
|---|---|---|
| Scope | Within Google Ads | Across Google Ads, Microsoft, Yahoo! JAPAN, Baidu |
| Bid timing | Auction-time (Target CPA/ROAS, Max Conversions/Value) | Intraday every 6h natively; auction-time when ATB enabled |
| Conversion sources | Google Ads + imported conversions | Floodlight, Google Ads, GA4, Adobe Analytics, offline import |
| Budget automation | Campaign / shared budgets | Budget bid strategies in plans, with pacing and forecasting |
| Planning layer | — | Bidding and Planning: spend plans, pacing charts, allocation tables |
Note: within Google Ads, Target Impression Share and Maximize Clicks are automated but not Smart Bidding — no auction-time bidding.
Also watch for Google's relabeling: "Maximize conversions with a Target CPA" now displays as Target CPA (and likewise for ROAS); behavior is unchanged.
Rule of thumb: if all spend sits in one Google Ads account, Smart Bidding covers most bidding capability alone.
SA360 earns its keep with multi-engine spend, many accounts, non-Google conversion sources, or centralized planning.
Manual vs automated: an honest assessment
Where automation has a structural advantage: conversion-based strategies set per-auction bids using time of day, device, location, browser, OS, the specific ad — and Smart Bidding adds query-level conversion data across the account, which rescues low-volume keywords no human could model individually.
At enterprise scale this compounds beyond anything manual management can touch.
Where manual still wins: near-zero volume (nothing to model); hard legal/contractual bid ceilings; broken measurement (automation optimizes confidently toward a false signal — manual at least fails visibly); and three-day event windows that end before learning finishes.
The 7-step decision framework
Work through these in order.
Don't skip to step four.
- Confirm measurement is trustworthy. Conversions must fire correctly, deduplicate properly, and sit under an attribution model the business endorses.
Supported sources: Floodlight, Google Ads conversions, GA4, Adobe Analytics, offline import.
If this layer is wrong, everything downstream is wrong.
- Name the goal type: visibility (impression share/position), traffic (clicks), or conversions/revenue.
- Choose the conversion action deliberately. Upper-funnel actions (Lead, Add to Cart) give volume and fast feedback but may not reflect revenue; lower-funnel actions (Sale, Purchase) measure ROI accurately but give less data, slower.
The most consequential choice in the whole process — revisit it if the strategy won't stabilize.
- Volume or value? All conversions worth about the same → Target CPA (or Maximize Conversions if budget is the binding constraint).
Different values you can measure → Target ROAS (or Maximize Conversion Value under a fixed budget).
Practitioner tip: anchor the initial target within ±20% of the last 30 days' actuals — an unreachable target restricts auctions rather than producing efficiency.
- Name the binding constraint. Efficiency (hit CPA/ROAS, spend follows) → target strategy.
Budget (spend this amount, efficiency follows) → budget bid strategy.
- Check data against the requirement: Maximize Clicks / Target Impression Share / Google Ads ATB — no minimum.
Intraday ROI portfolios — ~20 conversions/week.
ROAS — 4+ weeks of value data, longer at low volume.
Conversions should arrive frequently, ideally daily.
For judging Smart Bidding, Google suggests windows containing 30+ conversions (50+ for ROAS).
- Decide portfolio vs standard, intraday vs ATB. Pool campaigns with shared economics.
For eligible Google Ads campaigns, evaluate ATB against the conversion-data-sharing requirement.
Decision matrix
| Your situation | Recommended strategy |
|---|---|
| eCommerce, variable order values, reliable revenue tracking, efficiency mandate | Target ROAS |
| eCommerce, fixed seasonal budget, revenue mandate | Budget: Maximize Conversion Value |
| Lead gen, uniform lead value, CPA mandate | Target CPA |
| Lead gen, tiered lead quality with offline values | Target ROAS or Multiple Target CPA |
| Fixed quarterly budget, must be fully spent on conversions | Budget: Maximize Conversions |
| Brand defense; presence is the objective | Target Impression Share with bid limits |
| New market entry, no conversion history | Maximize Clicks, migrate once data accumulates |
| Multi-brand/multi-margin, distinct economics | Multiple Target CPA or ROAS |
| Regulated category, hard bid ceiling | Manual CPC or automated with strict bid limits |
Industry examples
Illustrative practitioner scenarios — not Google case studies, no performance claims.
eCommerce retail. Thousands of SKUs across margin tiers means variable value by definition: Target ROAS on Search and Shopping with revenue (ideally margin) as the value; Multiple Target ROAS where margin tiers diverge sharply; seasonal peaks handled via the planning layer and budget bid strategies, not manual target hikes.
B2B lead generation. Form fills converting to pipeline at different rates by source — optimizing to raw fills teaches the system to buy the cheapest, weakest leads.
Import qualified-lead or closed-won values offline and run Target CPA on the qualified action (or Target ROAS once value assignment is stable).
Long sales cycles mean slow learning: exclude immature data windows when judging.
Multi-brand enterprise. The scenario SA360 exists for: several brands in separate client accounts across engines, each held to its own economics via portfolio strategies, with plans and budget bid strategies pacing committed spend.
Agency, multiple clients. The win is governance: consistent strategy naming, centralized reporting at manager/sub-manager level, forecasting for target negotiations.
The common failure in inherited accounts is dozens of near-identical ad-hoc strategies — consolidating into fewer, better-fed portfolios is usually the highest-value first action.
Migrating from DoubleClick Search to SA360
Two migrations get conflated.
The 2018 rebrand required no advertiser action — accounts, strategies, and data carried over.
The legacy → new-experience migration was a genuine platform move with operational consequences.
If your runbooks predate it, expect gaps.
Pitfalls to check:
- Don't assume strategy behavior is identical — re-verify each migrated strategy's type, target, bid limits, and conversion goals.
- Conversion-goal drift: a strategy silently pointed at the wrong goal performs confidently and wrongly.
Confirm what each one optimizes toward.
- Post-migration learning isn't failure — setting changes can push strategies into learning mode.
Plan an unstable window and don't intervene inside it.
- Don't compare across the boundary — pre/post data may not be like-for-like.
Establish a fresh baseline.
- Schedule the nightly sync so client-account changes are reflected in SA360; unsynced accounts mean bidding on stale reality.
- Sequence by risk — low-spend, low-complexity portfolios first; re-audit bid limits set against an older cost environment.
- One change at a time: never run a major bid-strategy change and a major structural change in the same week — when performance moves, you won't know what caused it.
2026 behavior note: industry reporting indicates that since August 2026, budget-capped Target CPA/ROAS campaigns are being lifted toward their stated targets — relevant to portfolio and shared-budget setups.
Treat documented figures as starting points and verify against the current SA360 Help Center for your account.
Best practices
- Respect the learning period: wait 2–3 conversion-delay cycles before judging a new strategy; review reports only after a couple of weeks; exclude the most recent delay window from evaluation (judging last week's numbers with a 14-day delay makes a healthy strategy look broken).
- Change targets sparingly: at most once per week, never drastically.
Frequent or large changes don't damage the strategy itself, but a major target change produces a similarly sized swing in spend or volume.
- Add keywords gradually — bulk additions, especially history-less ones, dilute what the strategy is learning from.
- Use bid limits deliberately: only where a genuine ceiling exists.
Blanket limits as a comfort measure block the auctions the strategy was built to win.
- Check whether budget is the real constraint. A budget-capped strategy looks like an underperforming strategy — Google's own diagnostics call this out specifically.
- Forecast before changing targets — use the bid strategy forecast to see the cost/volume trade-off at candidate targets instead of adjusting and observing.
Mistakes and myths that undermine bidding
- Optimizing to the wrong conversion action — the single most consequential error; everything else is downstream of it.
- Mixing brand and non-brand in one portfolio — brand's cheap conversions flatter the average and mask non-brand inefficiency (the blended-average trap).
- Too many strategies, too little data each — fragmentation is the most common structural problem in inherited accounts.
Consolidate.
- Placeholder conversion values — value-based bidding on invented values optimizes precisely toward fiction.
- Panicking during learning — intervening mid-learning restarts learning; repeat it and the strategy never stabilizes.
- Set-and-forget for a quarter — automation manages bids, not strategy.
Markets, margins, and competition move.
- Stale bid limits and skipped syncs — silently cap strategies or bid on a stale picture of the account.
- Evaluating on immature data — guarantees false conclusions wherever conversion delay is long.
Myth vs fact: automation doesn't remove account management — it relocates it to goals, conversion quality, targets, budgets, and structure.
A lower CPA target doesn't always produce a lower CPA — an unrealistic target restricts auctions and can suppress volume without hitting the goal.
Automation doesn't fix poor structure — structure determines how data pools, and poor structure starves strategies.
And SA360 bidding vs Google Ads Smart Bidding aren't competing systems — they're composable, with SA360 setting portfolio goals and Google Ads setting auction-level bids.
Troubleshooting + launch checklist
Spend dropped sharply? Check for a recently tightened target, a max bid limit below market rates, or — most urgent — broken conversion tracking (no recorded conversions reads as performance collapse).
CPA well above target? Confirm learning is complete, exclude the immature delay window, check volume sufficiency and whether new history-less campaigns/keywords just joined.
Volume too low? Check keyword coverage and whether budgets are the real cap; use the forecast to preview a relaxed target.
Performance swinging week to week? Usually low volume — consolidate — or too-frequent target edits.
Long-tail getting no traffic? Intraday strategies already attempt rescue via daily bid increases; if budget-constrained, that mechanism can't run — budget is the root cause.
Launch/quarterly checklist: tracking verified across all sources in use; conversion action deliberately chosen with its volume/delay understood; values real and consistently populated; attribution model endorsed by the business; strategy type matches the binding constraint; portfolio grouping reflects shared economics; data meets documented minimums; target anchored to recent actuals; bid limits justified and current; budgets confirmed non-limiting; ATB evaluated with the data-sharing decision made; nightly sync scheduled; learning window defined with a no-touch commitment; evaluation window excludes immature data; review cadence set at ≤1 target change/week.
Frequently asked questions
Is DoubleClick Search the same as Search Ads 360?
Yes.
DoubleClick Search was rebranded as Search Ads 360 in 2018 under the Google Marketing Platform, and has since been rebuilt into a new SA360 experience with an updated interface and account structure.
What bid strategies does Search Ads 360 offer?
Target CPA, Target ROAS, Target Impression Share, Multiple Target CPA, Multiple Target ROAS, and budget bid strategies (Maximize Clicks, Maximize Conversions, Maximize Conversion Value) — most available as portfolio strategies across campaigns and accounts.
Which SA360 bid strategy is best?
There is no single best.
Target CPA for similar-value conversions with a cost constraint; Target ROAS for differing values you track reliably; a budget bid strategy when a fixed budget is the constraint; Target Impression Share when visibility, not outcomes, is the goal.
How much data does an SA360 bid strategy need?
No minimum for Maximize Clicks, Target Impression Share portfolios, or Google Ads auction-time bidding.
Intraday ROI portfolios: ~20 conversions/week recommended.
ROAS: 4+ weeks of conversion value data, longer at low volume.
More conversions generally mean better, more stable results.
How long is the learning period?
Google recommends 2–3 conversion-delay cycles before assessing a new strategy, and reviewing reports after a couple of weeks.
Long sales cycles mean longer learning windows.
How often does SA360 adjust bids?
Intraday bidding recalculates every six hours.
Google Ads auction-time bidding sets a bid at each individual auction.
Enabling ATB requires conversion-data sharing.
What is the difference between portfolio and standard bid strategies?
Standard applies to one campaign.
Portfolio applies to many at once — across client accounts and ad platforms — pooling data and shifting investment toward what delivers against the shared target.
Can one SA360 bid strategy cover Google and Microsoft campaigns?
Yes — a portfolio strategy can span Google Ads, Microsoft Advertising, Yahoo! JAPAN Ads, and Baidu.
Google Ads and Microsoft ATB can even be mixed within one strategy.
Can I still use Manual CPC or Enhanced CPC?
Yes, but neither is an SA360 portfolio strategy, so neither delivers cross-campaign or cross-engine optimization.
They suit low-volume campaigns, hard cost ceilings, and diagnostic baselines.
Is Search Ads 360 worth it over Google Ads alone?
Mostly when you run search across multiple platforms or many accounts, need conversion sources like Floodlight or Adobe Analytics, or require centralized planning and pacing.
Single Google Ads account? Smart Bidding covers most of it.
Pricing goes through Google Marketing Platform sales.
Should I switch bid strategies if performance is poor?
Not first.
Verify learning is complete, exclude immature data, confirm tracking is intact, check budgets and bid limits.
Switching restarts learning — diagnose before you switch.
Do SA360 bid strategies work with Performance Max?
Google documents support for Google Ads and Microsoft Advertising Performance Max campaigns, alongside Search, Shopping, and Dynamic Search Ads on supported platforms.
Confirm for your account in the SA360 Help Center.
Final recommendations
Bid strategy selection is the third decision, not the first: measurement quality first, conversion-action selection second, strategy type as a consequence of those plus your binding constraint.
Audit measurement before touching bidding.Consolidate into fewer, better-fed portfolios.
Match the strategy to the constraint — efficiency mandate means a target strategy, budget mandate means a budget bid strategy.
Evaluate auction-time bidding for eligible campaigns against the data-sharing requirement.
And hold a change cadence: one target change per week at most, evaluated on mature data.
DoubleClick Search bid strategies didn't disappear — they matured into SA360's bidding system, and the advertisers who get the most from it invest in the inputs, not in finding the perfect strategy name.
Thanks for reading! DoubleClick Search Bid Strategies: Full SA360 Guide you can check out on google.