Directly represents the lead Northstar wants to acquire.
Google Ads Search · Level 4 of 5
Measurement & Bidding
Good bidding cannot rescue bad measurement. Build the conversion signal first, then choose a bidding strategy and target that match the campaign's data maturity, economics and budget reality.
6 decision lessons~30 minScenario-basedProgress saved locally
0% complete0 of 6 lessons complete
Optimize toward the business outcome, not the easiest event to collect.
A conversion action is useful only if you understand what it represents. High-volume micro-actions can be valuable for observation, but making them optimization goals can teach Smart Bidding to chase activity that is easier rather than activity that matters.
Useful diagnostic behavior, but not equal to a qualified lead.
Useful for funnel analysis and troubleshooting drop-off.
Quick check
Northstar can track demo requests reliably. Which event should anchor campaign optimization?
Rule of thumb
Use the deepest reliable business outcome you can measure. More data is not automatically better data.
Primary and secondary actions answer different questions.
Primary actions are included in the Conversions column and can be used by bidding when they are part of the campaign goal. Secondary actions remain observable in All conversions without asking the campaign to optimize toward them.
| Action | Role | Use |
|---|---|---|
| Qualified demo request | Primary | Campaign optimization + core reporting |
| Form start | Secondary | Diagnose lead-form drop-off |
| Pricing-page view | Secondary | Observe research behavior |
| Newsletter signup | Secondary / separate goal | Useful business event, but different intent |
Quick check
A pricing-page view is useful for analysis but should not steer bidding. What is the cleanest setup?
Automation gets more useful when the signal is trustworthy and repeated.
There is no magical universal conversion threshold that makes a bidding strategy “safe.” Think in terms of signal quality, recent volume, conversion lag, budget, campaign stability and how restrictive your target is.
Prefer less restrictive automation and avoid forcing an aggressive target immediately.
Evaluate CPA/ROAS patterns before introducing tighter efficiency constraints.
Targets can become useful guardrails when they reflect achievable performance.
Quick check
A brand-new lead-gen campaign has reliable tracking but almost no campaign history. Which launch posture is most defensible?
Choose bidding around the outcome you value — then decide how much constraint you need.
Clicks, conversions and conversion value are different optimization objectives. Target CPA and Target ROAS add efficiency constraints; they do not create efficiency from nothing.
Useful when conversion volume matters and a strict CPA constraint is not yet appropriate.
Best when the target is economically meaningful and supported by sufficient signal.
Requires meaningful conversion values and is suited to revenue/value objectives.
Quick check
An ecommerce campaign tracks purchase value accurately and wants to prioritize revenue efficiency. Which objective family fits best?
A target is a constraint. Budget and achievable volume decide how much room the system has.
If Northstar targets £40 CPA with roughly £120/day, the budget theoretically supports about three conversions per day at target efficiency. That is a planning relationship — not a promise of exactly three leads every day.
| Scenario | Daily budget | CPA target | Planning implication |
|---|---|---|---|
| Current plan | £120 | £40 | ~3 conversions/day at target economics |
| Tighter target | £120 | £20 | Requires twice the efficiency; may restrict delivery |
| Very small budget | £30 | £40 | Less than one target-CPA conversion/day on average |
Quick check
Actual stable CPA is £46, but the business target is £40. Which change is more sensible than instantly setting tCPA to £25?
Five accounts. Five measurement and bidding decisions.
Choose the strongest response for each situation. Your result is scored when you review the lab.
1 · Lead generation
A form-start fires 180 times/month; qualified submitted leads fire 32 times/month and tracking is reliable.
2 · Ecommerce value
Purchase values range from £20 to £600 and revenue is measured accurately.
3 · New campaign
Tracking is solid, but the campaign has almost no conversion history.
4 · Micro-conversions
Newsletter signup and pricing-page view are useful signals but are not the campaign's sales outcome.
5 · Target pressure
A stable campaign delivers £42 CPA against a £40 business target. There is also obvious irrelevant-query spend.
Level 4 complete: your measurement signal now has a bidding strategy attached to it.
Next, diagnose what actually happens after launch — without treating every metric movement as an instruction to change something.
Continue to Level 5: Campaign Optimization →