In this article
Define quality without looking at the score
Write a qualification rule your sales team can apply consistently. It might include service eligibility, a real requirement, purchasing authority and an achievable timeframe. Avoid circular definitions such as “high quality means a high intent score.” That only confirms that the score agrees with itself.
Specify when the judgement is made. A new enquiry and a lead reviewed after a sales call have different amounts of available evidence. Comparing them without age controls can make a campaign appear to attract weaker leads simply because its leads are newer.
This is a pilot evaluation guide. It does not assume universal access, published scoring weights or a native 1ClickReport connection to this particular field. Check the current Google account and official product information for eligibility.
Create a score-versus-outcome table
For a practical review, use anonymous lead identifiers and only the information needed for the decision. Keep raw personal messages out of general-purpose analysis when an aggregated outcome table is sufficient. Define missing and unresolved outcomes rather than silently marking them unqualified.
| Field | Purpose |
|---|---|
| Lead date | Establish the observation window |
| Campaign | Link the outcome to acquisition context |
| Score or band | Record the signal being evaluated |
| Qualification status | Independent sales judgement |
| Outcome date | Distinguish mature and immature records |
| Sale or value | Assess commercial usefulness where available |
Keep a small review sample for disagreements. A lead may have a promising signal but fail because it is outside the service area. Another may have modest engagement but become a valuable customer. Those cases teach you whether the score aligns with the business's eligibility rules.
Evaluate precision and coverage separately
Precision asks what fraction of prioritized leads meet the business outcome. Coverage asks how many useful leads are captured by that prioritization. A policy that routes only the very highest scores can look precise while ignoring many viable customers.
For a hypothetical sample, 30 of 50 high-scoring leads become qualified. That is 60% observed qualification within that band. If another 40 qualified leads sit outside the band, routing only the high-scoring group would miss those 40. Neither figure establishes incremental revenue; both help evaluate the routing policy.
Include the age of the records and the unresolved denominator. If half the low-scoring leads have not been contacted, you do not have a fair comparison with fully reviewed high-scoring leads. The sales process itself affects what gets measured.
Avoid letting the score change the truth label
When sales staff can see the score, it can influence their judgement or attention. If practical, review some records without exposing the score first. Otherwise, document that the outcome assessment is influenced by the routing policy and cannot be treated as entirely independent.
Check service capacity. A small team might respond faster to prioritized leads, creating a real difference in outcomes. That can make the prioritization useful, but it does not prove the model identified inherently better people. Your decision may be about workflow improvement rather than prediction accuracy.
Do not use customer attributes as improvised explanations without a lawful, appropriate business purpose. The most useful discrepancies often concern offer eligibility, destination promise and response speed. Those are concrete operational factors the team can improve.
Connect campaign cost to the reviewed outcome
A low platform cost per lead can coexist with an expensive cost per qualified opportunity. Allocate spend to the same campaign and acquisition period used by the lead review. Keep unattributed and unresolved leads separate rather than distributing them arbitrarily across campaigns.
Calculate cost per qualified lead only where both spend and qualification counts are meaningful. If the campaign spent $1,200 and generated 12 independently qualified leads, the illustrative cost is $100 per qualified lead. State whether all leads had enough time to be reviewed. Report raw enquiry cost as a separate measure.
Use 1ClickReport with an evidence export
Ask the AI marketing analyst to review connected campaign data alongside an anonymized qualification summary you provide. Do not claim a native CRM or lead-score integration that you have not verified.
Compare campaign spend with this anonymized lead review. Keep raw leads, independently qualified leads, unresolved records and sales separate. Evaluate score-band precision and coverage only for mature records. Flag campaigns whose low reported CPL does not translate into qualified opportunities. Do not infer missing sales outcomes or treat the pilot score as the ground truth.The result should identify one decision: change routing, improve offer eligibility, investigate response delays or collect a more mature sample. It should not automatically recommend a bidding change based on a small or biased review.
Frequently asked questions
Is Google's lead intent scoring available in every account?
The cited episode describes a pilot. Confirm current access in your account rather than assuming broad availability.
Does a high score mean a lead is sales-ready?
Not by itself. The lead still needs to meet your independent business criteria.
Can I use a manual CRM export?
Yes, for this review workflow. Use appropriately authorized, minimized information and state the limitations of manual mapping.
Sources and research notes
Platform statements are linked below. The checklists, worked examples and analyst prompts are original editorial frameworks. Examples labeled hypothetical are not customer results. Research checked October 13, 2026; account availability may differ.