For B2B SaaS, lower cost per form fill does not necessarily mean lower customer acquisition cost. Start with a conversion event that predicts customers, verify the CRM-to-Google Ads connection, and evaluate results over your actual sales cycle.
This checklist separates current platform rules from strategy choices. Google documentation was checked on 9 September 2026. The suggested tests are not promises of a particular CAC reduction.
1. Optimizing for form submissions without checking lead quality
The mistake: Treating every demo request as equally valuable, then judging the campaign only on cost per submission.
Why it matters: A cheaper form fill can still be a poor fit for your product, market or sales process. Google cannot infer your sales team's qualification decision unless you supply a usable signal.
Fix: Agree on one meaningful bidding milestone with sales: a qualified lead, an attended demo or another event that reliably precedes revenue. Keep raw submissions available for diagnosis. Audit campaign goals so the same prospect is not counted repeatedly as a primary success at every funnel stage.
Connect your CRM using a supported integration or Google Ads Data Manager, then validate identifiers, timestamps, deduplication and upload diagnostics. Google recommends enhanced conversions for leads for new offline-conversion setups. Its current documentation also directs offline uploads toward the Data Manager API, with legacy Google Ads API access subject to restrictions. Do not build a new integration from an old upload-clicks tutorial without checking the migration guidance. Google's offline conversion documentation
Hashing customer data is not a substitute for a privacy review. Your privacy owner should review what is collected and shared before enabling customer-data features.
2. Treating 30 conversions as a universal Smart Bidding rule
The mistake: Assuming Maximize Conversions becomes usable at exactly 30 monthly conversions, or automatically moving a low-volume account to Manual CPC.
Fix: Distinguish eligibility from evidence. Google's Maximize Conversions documentation requires conversion tracking and explains that the strategy aims to spend the available budget. It does not impose a universal 30-conversion minimum.
Low volume still makes results harder to interpret. Check conversion delay, tracking reliability, budget exposure and whether unnecessary campaign splits are fragmenting the data. Consolidate compatible campaigns only when their goals and economics align.
Target ROAS is a different strategy with campaign-specific requirements. Google's current guidance lists 15 conversions in the past 30 days at the conversion-tracking level for Search and Shopping, alongside conversion-value setup. Meeting that threshold is not proof that a pipeline-value model is ready. Target ROAS requirements
After a material change, allow for your conversion delay before judging performance; do not confuse this with a mandatory calendar-month waiting period. Google's target-adjustment guidance
3. Ignoring conversion windows and import deadlines
The mistake: Assuming every closed-won deal can be attributed to its original click, however long the sales cycle takes.
Fix: Check both the configured conversion window and the limit for the import method. Google documents a maximum of 90 days from the associated click for GCLID imports, and 63 days for enhanced conversions for leads. A shorter conversion window can further limit attribution. Uploading after the applicable deadline does not extend it. Offline import limits
If your deal closes after that window, retain the full customer journey in your CRM and import an earlier meaningful milestone in time. Do not backdate a sale or relabel a weak lead as revenue to make it fit.
Separate search intent from measurement limits. A category definition query and a software-pricing query may need different offers and evaluation plans. With a constrained budget, start with clearly commercial use cases; test research-stage traffic only with a defined budget and a way to evaluate its contribution. Neither broad match nor top-of-funnel traffic is automatically unprofitable.
4. Sending visitors to a page that does not answer their query
The mistake: Sending every campaign to the same page regardless of product, audience or purchase stage.
Fix: Check the experience against the actual query. A useful destination answers:
- Is this built for my company and use case?
- Can it meet our integration, security and implementation requirements?
- What happens after I request a demo or start a trial?
- What evidence supports the claims on the page?
A homepage can be appropriate for a brand query. A specific integration or use-case query may need a focused landing page. Test the change against qualified leads and downstream results, not a promised industry-wide conversion-rate lift. There is no required number of landing pages.
5. Counting brand efficiency as proof of new demand
The mistake: Blending brand and non-brand results and calling the combined CPA your acquisition performance.
Fix: Where separate budget control is needed, use separate campaigns, not just separate ad groups: Google Ads budgets are controlled at campaign level or through shared campaign budgets. Google's campaign-budget guidance
Review search terms and brand exclusions where supported, so campaign names do not become your only definition of brand traffic. Report brand and non-brand results separately.
Brand advertising may protect visibility, support a specific offer or capture demand. It may also overlap with organic visits. Use auction evidence and, where practical, controlled tests to evaluate incremental value. A low brand CPA alone does not establish that the ads created customers who otherwise would not have arrived.
Customer Match observation is not an ICP restriction. It allows audience analysis without limiting reach to that list. Nor should you recommend manual audience bid modifiers as a Smart Bidding control: Smart Bidding handles auction-time bids and does not support those manual adjustments. Observation settings, bid-adjustment support
6. Using target ROAS with unvalidated pipeline values
The mistake: Assigning arbitrary values to leads, or adding full deal value at several funnel stages and treating the total as revenue.
Fix: Decide what each value represents. Actual closed-won revenue and estimated opportunity value are different measures and should be labelled separately in reports.
For illustration only, an opportunity with a potential first-year contract value of $20,000 and a historically calibrated 10% win probability has an expected value of $2,000. That is a modelled estimate, not $2,000 of booked revenue and not a benchmark for your business.
Validate probabilities using your own completed cohorts. Check whether values are consistently available, arrive in time and distinguish materially different outcomes. Use appropriate conversion goals and adjustments to avoid double-counting the same customer journey. Test value-based bidding only when the inputs and campaign eligibility support it.
7. Recommending a retired attribution model
The mistake: Recommending position-based attribution, or saying data-driven attribution requires 300 conversions per month.
Fix: Google Ads no longer supports position-based, first-click, linear or time-decay attribution. Data-driven attribution and last click remain available. Current attribution models
Google states that all conversion actions are eligible for data-driven attribution regardless of volume, though additional data improves the model. It is the default for most conversion actions, not a reward unlocked at 300 monthly conversions. Data-driven attribution eligibility
Attribution is not a complete record of buyer influence or proof of incrementality. Compare Google Ads with CRM cohorts, analytics and optional self-reported discovery answers. Reconcile conversion definitions, dates, windows and eligible channels before treating differences as errors.
Where to start: a measurement-first review
Work through these checks in order, keeping a record of each change:
- Define a qualified lead with sales and inspect the active campaign goals.
- Trace a permitted test record from the website into the CRM and through import diagnostics.
- Check the import deadline against the time it takes to reach the selected milestone.
- Review bidding eligibility, values and budget exposure before changing strategy.
- Separate brand reporting, then test query-to-page fit.
- Compare completed acquisition cohorts after allowing for conversion delay.
Keep the denominator explicit: media-only CAC is ad spend divided by new customers attributed under the chosen method. A fully loaded CAC also allocates relevant sales and marketing costs. Compare like with like; pipeline value is not customer revenue.
There is no evidence here for a guaranteed percentage reduction or a fixed time to the first customer. The aim is to make the next spending decision more reliable.
If you're considering switching agencies
If you are evaluating a new partner, compare the proposed scope, measurement plan and evidence of relevant experience. Our SaaS PPC agency buyer's guide provides a sourced, fit-based shortlist rather than an audited performance ranking.
Free SaaS Google Ads audit
Request a free Google Ads audit to review your conversion setup, campaign structure and next tests with a senior strategist. The recorded review is typically delivered within 48 hours after read-only access and scope are confirmed; it is yours to keep.
For the service approach, see our SaaS & B2B Google Ads management page. To separate management fees from media spend, use our PPC management pricing guide.