Key Takeaways
- Define the business outcome before choosing dashboard metrics.
- A contact click is not a confirmed enquiry; an enquiry is not automatically a sale.
- ROAS uses revenue, while a useful return calculation also accounts for delivery costs.
A campaign can generate clicks without generating suitable work. Start by defining what success means for the business: qualified enquiries, booked consultations, orders or repeat purchases. Then decide which records can establish each outcome.
Create a small measurement plan
| Stage | Evidence | What it does not prove |
|---|---|---|
| Visibility | Search impressions or advertising reach | That someone visited or wanted to buy |
| Visit | A recorded landing-page session | That the visitor submitted a request |
| Contact action | A phone, email or booking-link click | A completed conversation or enquiry |
| Confirmed enquiry | A successful form submission or a recorded inbound contact | That the lead is suitable or has bought |
| Qualified opportunity | A reviewed CRM or sales record | That the work has been won |
| Sale | A confirmed business transaction | That the entire value is profit or caused by one campaign |
Use consistent definitions in the report. Google's key-event guidance explains how important events are represented in GA4, but the business must still choose what its event actually establishes. Test the trigger against a controlled submission rather than assuming the label generate_lead proves delivery.
Keep the comparison fair
Compare complete periods using the same sources, country, device and landing-page filters. Record changes to tracking, consent, budgets and offers. Small samples need absolute counts: one enquiry becoming two is an increase of one, even though the percentage doubles.
Search Console clicks, GA4 sessions, hosting visitors and third-party traffic estimates have different definitions. Do not force them to match. A discrepancy is a reason to check collection and definitions, not permission to count every hosting request as a human customer.
Calculate revenue return and contribution separately
ROAS = attributed advertising revenue ÷ advertising spend. It describes a revenue ratio, not profit.
For a clearly defined campaign contribution calculation: marketing ROI = (attributable contribution before marketing − campaign cost) ÷ campaign cost × 100. State what costs and attribution assumptions you included; finance may use a different definition for its reporting.
Here is a hypothetical example, not a client result. A campaign costs £1,000 and is attributed £3,000 of revenue. Delivering those sales costs £1,800, leaving £1,200 before marketing. The revenue-to-spend ratio is 3:1. After campaign cost, the remaining contribution is £200, giving 20% on this defined calculation. If management, creative or other campaign costs were omitted, add them before interpreting the return.
Attribution still does not establish incrementality: some customers might have bought without the campaign. Describe that uncertainty instead of presenting an attributed total as a proven causal result.
Connect analytics to the enquiry record
Use a stable campaign naming convention and appropriate landing links. Do not put personal information in URLs or analytics event parameters. Keep contact details in the authorised business system, and report aggregate outcomes where possible.
Record the enquiry's status and the team's follow-up. Separate spam, duplicates and unsuitable requests from genuine opportunities. If analytics misses some visitors because of consent or blockers, keep the gap visible; do not bypass their choice to make the report look complete.
The traffic but no-enquiries guide helps test collection and delivery. For automated follow-up, use the email automation guide so a sequence stops when its purpose has been fulfilled.
Decide what to change next
Match the action to the evidence. Irrelevant queries need a targeting review. A broken form needs a repair. Good enquiries that receive no timely reply need an operational change. A small dataset may justify further observation rather than a new website.
For Slough digital-marketing projects, we separate audience, service area, budget and follow-up responsibilities before selecting channels. Our analytics and reporting service can help define the measurement plan. Share the campaign question, without exporting private customer records into a public report.
Frequently Asked Questions
Is a phone-link click a lead?
It is a contact action, not proof that a call connected or became a qualified enquiry. Report it separately from confirmed contacts and reviewed opportunities.
Why can ROAS look good while profit is low?
ROAS compares revenue with advertising spend. Product or service delivery costs, management, creative work and other costs can substantially change the commercial result.
Can GA4 prove that a campaign caused every attributed sale?
No. Attribution assigns credit under a measurement model. Establishing incremental sales requires a stronger comparison or experimental design, with its limitations made clear.
