Match Types
Part of Search ad experiments
Comparing bidding goals under similar campaign conditions
Test a different search bidding goal while keeping conversion signals and campaign conditions comparable. Allow for adaptation and conversion delay.
Compare bidding goals by changing the strategy or target you mean to test while holding the audience, conversion definition and offer steady. Decide beforehand whether the business seeks more accepted enquiries, a lower cost per accepted enquiry or more order value within its spending limit. An improved platform metric is useful only if it supports that decision.
Specify what differs
Record the original and trial settings precisely. "Try automation" is too vague: click-focused, conversion-focused and value-focused bidding pursue different recorded outcomes. Name each Google Ads strategy; for example, compare Maximise conversions with Maximise conversion value.
For these Search strategies, both arms use a budget and no target. Maximise conversions pursues conversion count, while Maximise conversion value pursues total conversion value. Keep the budget setting unchanged and, if conversion values are used, keep their rules and inputs steady.
If the planned difference is a target, name it and record its value and unit. Target CPA sets the cost per conversion you are willing to pay; Target ROAS sets the return on ad spend you want. Do not treat their numeric targets as directly comparable.
If the trial values orders differently, check that those values reflect real differences the business wants bidding to favour. If testing two targets under one strategy, record both targets and the acceptable business range.
Check which conversion goals the campaign uses. Keep the goal setup aligned between arms unless it is part of the planned difference; otherwise document the difference and describe the broader test accurately.
Bidding Strategy Comparison: Maximise Conversions vs Maximise Conversion Value
- Bidding Goal
- Maximise conversions
- Bidding Goal
- Maximise conversion value
- Primary Objective
- Maximise number of accepted enquiries
- Primary Objective
- Maximise total order value within budget
- Conversion Value Input
- Required for conversion value strategy
- Target CPA or ROAS
- Not used in either strategy (budget-only)
Pros and Cons of Automation Bidding Strategies
- Pros – Maximise Conversions
- Focuses on quantity of accepted leads; ideal for lead generation campaigns
- Cons – Maximise Conversions
- May increase cost per conversion if quality varies; doesn't account for value
- Pros – Maximise Conversion Value
- Prioritises high-value conversions; suitable for e-commerce or service-based businesses
- Cons – Maximise Conversion Value
- Requires accurate conversion values; may underperform if values are misassigned
Protect the comparison
A cleaner comparison changes one variable at a time and avoids simultaneous ad or landing-page tests. Test in an established campaign with enough conversion volume to support a meaningful comparison; this is a planning guide, not a guarantee that a particular test will be conclusive.
For Target CPA, Google recommends at least 15 conversions in the last 30 days. Treat that as planning guidance, not a guarantee that a comparison will be conclusive.
In Google Ads, the "Test your bid strategy" test is one way to compare bidding. Confirm both arms are eligible and serving. Keep keywords, match types, negatives, locations, ads, pages and conversion settings aligned except for the planned difference. An even split need not produce equal spend or impressions because the strategies can enter and win auctions differently.
Hold steady where possible / Record if it changes
- Conversion action, value and counting rule
- A new or reclassified goal
- Search coverage and locations
- Added keywords, negatives or geography
- Offer, ad and destination
- A promotion or page revision
- Spending boundary
- An unplanned budget or target change
Key Performance Indicators for Bidding Strategy Testing
- Conversion delay consideration
- Exclude recent days with incomplete conversions
- Budget consistency
- Must remain unchanged across both arms
Steps to Conduct a Controlled Bidding Strategy Test
- Define the business objectiveChoose between more leads, lower cost per lead, or higher order value
- Fix all campaign variablesKeep audience, offer, landing page, keywords, match types and negatives consistent
- Set up two campaign armsUse identical settings except for the bidding strategy or target
- Run uninterrupted comparisonAllow for ramp-up and exclude incomplete conversion data
- Evaluate against business outcomesCompare platform metrics with confirmed sales or accepted enquiries in records
Allow adaptation and conversion delay
Conversion-focused bidding can need time to adapt. Allow for an initial ramp-up period, then run an uninterrupted comparison period, and exclude recent days whose conversions may still be incomplete when judging results. Set a review date using the account's actual conversion cycle; an early swing is not a verdict.
Review the preselected measure, available confidence interval and spend. Compare them with accepted enquiries, retained orders or confirmed sales in business records. A lower platform cost per form submission could still accompany fewer useful leads. A higher reported return depends on the accuracy and completeness of the values supplied.
For example, if Maximise conversion value reports a higher return but mature business records show no improvement in confirmed sales value, that platform result has not shown the chosen business benefit. If Maximise conversions produces more form submissions but fewer accepted enquiries, it has not delivered more useful leads.
If conversion goals, pages or delivery changed during the run, document the disruption. Conclude only what the remaining comparison and mature business outcomes support.
