How to Optimize Your Bids for Mobile vs. Desktop Devices

Device bid adjustments let you raise or lower what you pay for a click based on whether the searcher is on a phone, a tablet, a computer or a TV screen. To optimize your bids for mobile vs. desktop devices properly, you need at least 30 conversions per device over the last 60 to 90 days, a conversion action that captures calls as well as form fills, and the discipline to move in 10 to 15 percent increments rather than swinging from +50 to -40 percent.

Most accounts get this wrong in the same direction: they cut mobile because the cost per conversion looks ugly, then lose the cheapest volume they had. The data usually is not wrong, but it is almost always incomplete.

What Device Bid Adjustments Actually Do Now

In Google Ads, a device bid adjustment is a percentage modifier applied at the campaign or ad group level, ranging from -100 percent (effectively excluding a device) up to +900 percent. Computers, mobile, tablet and TV screens each get their own slider, and adjustments set at the ad group level override the campaign level.

The important detail is how much authority those numbers still carry. With Manual CPC or Maximize Clicks, the modifier does exactly what it says and multiplies your base bid. With Smart Bidding strategies such as Target CPA or Maximize Conversion Value, device is already one of the live signals Google evaluates at auction time, so your percentages act as hints rather than hard math. Google’s own documentation on bid adjustments is worth a read before you touch anything, because eligibility varies by strategy.

• Manual CPC and Maximize Clicks: device adjustments apply directly and predictably.
• Enhanced CPC: your adjustment sets the starting point, then Google layers its own multiplier on top.
• Target CPA, Target ROAS, Maximize Conversions: only -100 percent has a guaranteed effect, which is a full device exclusion.
• Performance Max and Demand Gen: no device sliders at all, only device targeting exclusions in some cases.

Read Your Device Data Before You Change Anything

Open the Devices report inside a campaign, set the date range to the last 90 days, and add columns for impressions, clicks, CTR, average CPC, conversions, cost per conversion and conversion value. Then segment by network, because a Search campaign that also runs on Search Partners can hide wildly different mobile behaviour inside a blended average.

Three numbers matter more than the headline cost per conversion:

1. Conversion rate by device, which tells you whether the gap is an intent problem or an experience problem.
2. Average order value or lead value by device, since phone traffic often converts at a lower basket size while producing more total transactions.
3. Conversion volume per device, because any device with fewer than about 30 conversions in the window is a sample too small to act on confidently.

If you are not already watching these weekly, the shortlist in our breakdown of PPC metrics worth tracking daily will tell you which numbers deserve a dashboard slot and which ones are noise.

Why Mobile Performance Looks Worse Than It Is

This is the gap most articles skip. Mobile CPA is frequently inflated by measurement, not by behaviour, and the fix is instrumentation rather than a bid cut.

Cross-device journeys are the biggest distortion. Someone taps your ad on a phone during a lunch break, researches, then buys that evening on a laptop, and unless your conversion tracking stitches those sessions together, mobile eats the cost while desktop banks the credit. Google reports cross-device conversions by default in most modern accounts, but only when users are signed in and consent signals allow it.

• Untracked phone calls. Service businesses often see 40 to 70 percent of mobile conversions arrive as calls. Without call tracking, those conversions simply do not exist in your Devices report.
• Directions and store visits. A tap on a location extension is high intent and rarely counted as a primary conversion action.
• Consent mode gaps. Mobile users decline cookie banners at higher rates on small screens, which suppresses recorded conversions.
• Attribution windows that are too short. A 7 day window will systematically punish mobile if your average sales cycle runs two or three weeks.

Before you drop a mobile bid by 30 percent, confirm your tracking is actually capturing what phones produce. Otherwise you are optimizing against a reporting artefact.

The Math Behind a Device Bid Adjustment

When you are on a manual or clicks-based strategy, the calculation is straightforward. Divide your target cost per conversion by the device’s actual cost per conversion, subtract one, and multiply by 100.

Say your target CPA is $60. Desktop is delivering conversions at $48, so the raw suggestion is +25 percent. Mobile is at $82, which suggests -27 percent. In practice you should apply roughly half of any calculated move on the first pass, wait 14 days, then reassess, because auction dynamics shift as soon as your bid does.

• Cap early moves at 15 to 20 percent in either direction to keep data comparable between periods.
• Never combine a device change with a budget, keyword or landing page change in the same week, or you lose the ability to attribute the result.
• Treat tablets separately. Tablet volume is small in most accounts, and a -20 percent tablet adjustment is often a quiet, safe saving.
• Use -100 percent sparingly. Full device exclusion also removes the impression data you would need to reverse the decision later.

Working With Smart Bidding Instead of Against It

Once a campaign moves to Target CPA or Maximize Conversion Value, the lever changes from percentages to inputs. The algorithm already reads device alongside location, time of day, query wording, audience list and browser, so your job becomes feeding it better information.

Three inputs move the needle more than any slider:

1. Conversion values that reflect real profit. If mobile leads close at 15 percent and desktop leads at 25 percent, pass different values so bidding can price them correctly.
2. Complete conversion imports, including calls longer than 30 seconds and offline closes uploaded through GCLID matching.
3. Device-aware segmentation, meaning a separate campaign when mobile economics differ enough to justify their own target and budget.

Splitting campaigns by device is the blunt instrument, and it costs you shared learning, so reserve it for cases where the performance gap holds steady across two or three months. The trade-offs between letting the machine decide and keeping your hands on the controls are covered in more depth in our comparison of manual and automated bidding strategies, and the newer Smart Bidding Exploration setting deserves the same cautious testing approach.

Fix the Mobile Experience Before Cutting Mobile Bids

A poor mobile conversion rate is usually a landing page problem wearing a bidding costume. Google’s field data has long shown that bounce probability climbs sharply as load time stretches past three seconds, and the Core Web Vitals thresholds give you concrete targets: an LCP under 2.5 seconds and an INP under 200 milliseconds on real mobile connections.

Common culprits on phone traffic include multi-step forms with desktop-sized fields, hero images above 300KB, chat widgets that cover the call button, and sticky headers that swallow a third of the viewport. For WordPress and WooCommerce sites, backend bloat matters too, which is why cleaning up query overhead through WordPress database optimization and tightening WooCommerce product page structure often lifts mobile conversion rates more than any bid change.

Test one page element at a time on mobile, give it a fortnight, and compare against your desktop baseline. If mobile conversion rate rises 20 percent after a form redesign, the bid cut you were about to make would have cost you the growth.

A Monthly Device Bidding Routine

1. Pull 90 days of device data per campaign, segmented by network and conversion action.
2. Check volume thresholds and set aside any device with fewer than 30 conversions.
3. Verify tracking coverage for calls, cross-device conversions and consent-affected sessions.
4. Calculate indicated adjustments, then halve them before applying.
5. Log the change with a date note in the account change history.
6. Review after 14 days against conversions and value, never against CPC alone.

Agencies that manage this on a fixed cadence catch device drift early, and if you would rather hand the routine to someone local who knows your market, there are practical advantages to working with a local agency on paid search.

Frequently Asked Questions

Do Device Bid Adjustments Still Work With Smart Bidding?

Only -100 percent reliably applies when you use Target CPA, Target ROAS or Maximize Conversions, since those strategies set bids per auction using device as one signal. Positive or moderate negative adjustments are treated as preferences. To influence device spend under Smart Bidding, adjust conversion values, improve tracking or split campaigns by device.

What Is a Safe Starting Device Bid Adjustment?

Start within plus or minus 15 percent and hold it for 14 days before reviewing. Larger opening moves change auction position enough to distort the comparison data you need, and they often cut volume faster than they cut cost.

How Much Traffic Comes From Mobile Devices?

Roughly 60 percent of global web traffic now arrives on phones, and for local service searches the share commonly sits between 70 and 80 percent. Excluding mobile entirely removes most of the available query volume in those verticals, which is why mobile exclusions are rarely the right answer.

Should I Build Separate Mobile and Desktop Campaigns?

Split campaigns only when device CPA or conversion rate differs by more than about 40 percent for three consecutive months and each split would still see 30 or more conversions monthly. Separate campaigns give you independent budgets and targets, but they fragment conversion data and slow down bidding learning.

Can Maximize Clicks Be Used for Device Optimization?

Yes, and the Max Clicks bid strategy is the easiest place to test device adjustments because your percentages apply directly to the bid. Use it with a maximum CPC cap during a two to four week diagnostic period, then move back to a conversion-based strategy once you know how each device behaves.

Get a Device Bidding Review From SEO Quirk

If your mobile numbers look weak and you are not sure whether the problem is bidding, tracking or the landing page itself, we will audit all three and tell you which one to fix first. Talk to SEO Quirk about a device performance review of your Google Ads account.