Google Ads Bid Adjustments: Device, Location, and Audience Guide 2026
Table Of Contents
TL;DR: Bid adjustments let you pay more or less for clicks depending on where users are, what device they're on, and which audience segment they belong to β without changing your base bid. Used correctly, they narrow the gap between your CPL and your tCPA target. Used wrong, they fight against Smart Bidding and hurt performance. This guide covers what to adjust, by how much, and when to leave it to the algorithm. If you're scaling campaigns, verified Google Ads accounts with established Quality Scores respond to bid adjustments faster than zero-history accounts.
| β Right fit if | β Wrong fit if |
|---|---|
| You have clear performance differences by device or geo | You have fewer than 100 conversions/month per segment |
| You run Standard Search or Display campaigns | You're on Smart Bidding with sufficient conversion data |
| You're on manual CPC or Enhanced CPC | You're using PMax (no bid adjustment access) |
| Your campaign data shows 2x+ CVR difference by segment | You're in the learning phase (first 3 weeks) |
Bid adjustments are multipliers applied to your base bid for specific conditions: device type, geographic location, audience list membership, time of day, or call interactions. A +30% device adjustment on mobile means you're willing to pay up to 30% more for a click from a mobile user.
What Changed in Google Ads Bid Adjustments in 2026
- Smart Bidding deprecation warning: Google now shows an in-UI warning when you apply bid adjustments to campaigns using tCPA or tROAS β reminding you that Smart Bidding already accounts for these signals
- Audience bid adjustments now available for Customer Match lists in Search without requiring "Observation" mode explicitly β Customer Match targeting simplified in interface
- Location adjustments expanded: New granularity for adjustments at the postal code level (previously city minimum for most markets)
- Device adjustments for Connected TV added in Display campaigns β CTV can now be adjusted separately from Desktop in eligible accounts
- Demographic adjustments β age and gender bid adjustments now surface prominently in the Audiences tab, previously buried in demographics report
Bid Adjustment Ranges by Dimension
| Dimension | Range | Applies To |
|---|---|---|
| Device (Mobile) | -100% to +900% | Search, Display, Shopping |
| Device (Tablet) | -100% to +900% | Search, Display, Shopping |
| Device (Desktop) | 0% (baseline) | β |
| Device (TV screens) | -100% (exclude only) | Display |
| Location | -90% to +900% | Search, Display, Shopping |
| Audience (Observation) | -90% to +900% | Search, Display |
| Ad Schedule (hourly) | -90% to +900% | Search, Display |
| Call interactions | -100% to +900% | Search |
| Demographic (age/gender) | -90% to +900% | Search, Display |
Device Bid Adjustments
Device adjustments are the most commonly used. The logic is straightforward: if mobile converts at half the rate of desktop but your bid strategy doesn't account for this, you're overpaying for mobile clicks.
When to Apply Device Adjustments
- Pull 90-day device performance report: Google Ads β Reports β Predefined reports β Basic β Performance by device
- Calculate Device CVR Index: Divide each device's conversion rate by desktop CVR. Desktop = 1.0 baseline.
- Apply adjustment: Device CVR Index below 0.7 = negative adjustment. Above 1.3 = positive.
Example calculation: - Desktop: 5.2% CVR β index 1.0 (baseline, no adjustment) - Mobile: 2.8% CVR β index 0.54 β apply -45% bid adjustment - Tablet: 4.1% CVR β index 0.79 β apply -20% bid adjustment
For CPL-optimised campaigns, also factor in the mobile CPC difference. According to WordStream (2025), average CPC across all industries is $5.26 for desktop. Mobile CPCs often run 15-30% lower, which partially offsets lower CVR.
Related: Google Ads CPC Benchmarks 2026: Industry Data and How to Lower Your Cost Per Click
β οΈ Important: Applying a -100% device adjustment to mobile is the same as excluding that device. Only do this if your product literally cannot be used on mobile (e.g. a software installer that requires desktop). Excluding mobile from lead gen campaigns in 2026 typically cuts reach by 55-65% because mobile accounts for the majority of search traffic.
Progressive Adjustment Strategy
Don't jump to a -50% adjustment based on 30 days of data. Start conservative:
- Week 1-2: Apply -20% if mobile CVR is more than 30% below desktop
- Week 3-4: Review performance, adjust by Β±10% increments
- Month 2+: Stabilise at whatever adjustment produces equal CPL across devices
Location Bid Adjustments
Location adjustments let you increase bids in high-value markets and reduce them (or exclude) in low-performing areas. This is especially valuable for:
- Local service businesses (dentists, plumbers) competing in a metro area
- National advertisers with geographic CPL differences
- Media buyers running geo-split tests
Finding Your Top-Performing Locations
Google Ads β Reports β User location report (where the user physically was when they clicked). Filter for 90 days + minimum 50 clicks. Sort by conversion rate or CPL.
Apply positive adjustments (+20% to +50%) for locations where CVR is significantly above account average. Apply negative adjustments (-30% to -60%) for locations with poor conversion rates but unavoidable impressions.
Related: Google Ads CPA and Target CPA Bidding: Complete Strategy Guide for 2026
Case: Legal services advertiser, $400/day budget, national Search campaign. Problem: The account's average CPL was $140 (below the industry $131.63 per WordStream 2025 benchmark β but some states were dragging the average up). Texas and Florida CVR was 4.1% while Nevada was 1.2%. Action: Applied +35% location adjustment for TX and FL, -50% for NV and three other low-converting states. Result: Overall CPL dropped from $140 to $112. Conversion volume increased 18% at the same budget because budget was redistributed toward high-converting geos.
Need accounts for multi-geo campaigns? Browse verified Google Ads accounts β accounts with billing addresses matching your target geo often get better auction participation in local-intent queries.
Audience Bid Adjustments (Observation Mode)
Audience bid adjustments in "Observation" mode let you see how different audience segments perform without restricting your targeting β and then increase or decrease bids for those segments.
Audiences Worth Adjusting
| Audience Type | Typical Adjustment | Why |
|---|---|---|
| Remarketing (past visitors) | +30% to +100% | 3-5x higher CVR than cold traffic |
| Customer Match (existing customers) | +20% to +50% | Already know your brand |
| Similar Audiences (lookalikes) | +10% to +30% | Warmer than fully cold traffic |
| In-market segments (relevant) | +15% to +40% | High intent signal |
| Demographics: age 18-24 | -20% to -40% (varies) | Often lower purchase intent for high-ticket items |
Adding Audiences in Observation Mode
Google Ads β Your campaign β Audiences β Edit audience segments β Add audience β Observation. Select audiences from: - Your remarketing lists (Website visitors, YouTube viewers) - Customer Match lists (uploaded from CRM) - Google's in-market and affinity audiences
After 3-4 weeks, check performance in the Audiences tab. For segments with 20%+ better CVR, apply a positive adjustment. For segments dragging the average, apply negative.
Related: Budget and Bid Settings in Twitter Ads: Should You Choose Auto or Manual?
β οΈ Important: Observation mode audiences don't restrict who sees your ads β they just let you measure and adjust bids per segment. Don't confuse with "Targeting" mode, which limits your campaign to only those audience segments. Using "Targeting" mode on Search campaigns significantly shrinks reach and should only be done with explicit intent.
Bid Adjustments vs Smart Bidding: When to Use Which
This is the most misunderstood area in Google Ads bid management.
| Scenario | Recommended approach |
|---|---|
| Manual CPC campaigns | Use bid adjustments fully |
| Enhanced CPC | Use bid adjustments β ECPC partially overrides them |
| tCPA / tROAS (Smart Bidding) | Avoid bid adjustments β Smart Bidding already factors device/location/audience |
| tCPA with < 30 conv/month | Use device adjustments only β Smart Bidding lacks data for these signals |
| PMax campaigns | No bid adjustments available β algorithm handles all allocation |
According to Google, 86% of campaigns use automated bidding strategies (Google Ads Blog, 2026). For these campaigns, bid adjustments can confuse the algorithm by adding a second layer of signal that conflicts with Smart Bidding's own modelling.
Exception: Device adjustments on tCPA campaigns can still help when you have very clear evidence (500+ conversions per device) that one device is systematically under-performing. But the adjustment should be modest (β€ Β±30%) to avoid destabilising the learning phase.
The Learning Phase Problem
Changing bid adjustments by more than 20% at once resets the Smart Bidding learning phase. According to Google, the recommended learning period is 3 weeks + 60 conversions before adjusting anything. If you make large bid adjustment changes mid-flight, expect 1-2 weeks of performance volatility.
Case: B2B SaaS, $300/day Search campaign, tCPA strategy, 45 conversions/month. Problem: Mobile CPL was $220 vs desktop $140. Smart Bidding was treating both devices equally. Action: Applied -40% mobile device adjustment. Waited 3 weeks for learning phase to complete. Result: Mobile CPL dropped to $158. Total CPL fell from $170 to $143 (blended). Conversion volume was unchanged β Smart Bidding redistributed spend to higher-converting desktop without significant reach loss.
Ad Schedule Bid Adjustments
For businesses with clear conversion windows (B2B: weekdays 9-5; hospitality: evenings/weekends), ad schedule adjustments help concentrate budget.
Best practice: Pull 13-week performance by hour of day β identify hours with CPL β₯ 2x account average β apply -30% to -60% for those hours. Don't exclude (avoid -100%) unless the hour truly produces zero conversions.
Quick Start Checklist
- [ ] Pull 90-day device performance report, calculate CVR index per device
- [ ] Apply device adjustments where CVR index < 0.7 or > 1.3 (start at Β±20%)
- [ ] Add key audience segments in Observation mode (remarketing, Customer Match)
- [ ] Wait 4 weeks, then apply audience bid adjustments where CVR differs by 20%+
- [ ] Pull User Location report, identify locations with CPL 2x+ above average
- [ ] Apply location adjustments: +20-35% for top geo, -30-50% for poor performers
- [ ] For Smart Bidding campaigns: limit adjustments to device only and keep β€ Β±30%
- [ ] After any adjustment > 20%: note the date, expect 1-2 weeks learning volatility
- [ ] Review all adjustments monthly β market conditions change seasonally
Scaling campaigns across multiple devices and geos? Verified Google Ads accounts with a clean history respond more predictably to bid adjustment changes β no residual flags or policy holds slowing down the learning phase.
What to Read Next
How to Choose Music and Sounds in TikTok: Trending Audio, Licensing, and Sound Strategy for 2026
Updated: April 2026 TL;DR: Sound selection determines up to 80% of your TikTok reach β the algorithm uses audio signals to...
Prompt Engineering: Query Structures, Roles, Restrictions, and Practical Examples
Updated: April 2026 TL;DR: Prompt engineering is the skill of structuring AI queries to get precise, repeatable results β and it...
AI Agents: How Action Chains and Tools Work Under the Hood
Updated: April 2026 TL;DR: An AI agent is an LLM that can plan, use tools, and execute multi-step workflows autonomously. Unlike...
