Revenue
How to Track Which Marketing Channels Actually Drive Revenue
Go beyond the traffic leaderboard. Compare customers, revenue and acquisition costs to see which channels deserve a closer look.
Premely team
On this page
The traffic winner is not always the revenue winner
Imagine reviewing these three channels at the end of a month. Google delivered most of the visitors. LinkedIn delivered the most revenue. Both statements are true, but they suggest very different next actions.
| Channel | Visitors | New customers | Revenue |
|---|---|---|---|
| 8,200 | 43 | £5,400 | |
| 1,400 | 36 | £6,800 | |
| 600 | 25 | £7,100 |
Revenue per visitor makes the contrast easier to see: about £0.66 for Google, £4.86 for Reddit and £11.83 for LinkedIn. That does not prove you should move all your budget to LinkedIn. It tells you that a traffic-only leaderboard is missing an important part of the story.
The useful question is not “which channel is biggest?” It is “which channel brings the right customers at an acquisition cost we can support?” Answering it means connecting acquisition evidence to verified outcomes, then checking the economics. That is the kind of review Premely’s connected revenue view is designed to support.
Define the report before joining the numbers
Choose a reporting period, a currency and a customer definition. Then decide whether you are reporting payments received during that period or revenue from visitors acquired during it. Those are different reports. A payment received today may belong to a customer first acquired six months ago.
- 01Capture source and campaign
- 02Record the conversion
- 03Preserve a permitted matching key
- 04Confirm the payment
- 05Apply the attribution rule
- 06Report unmatched revenue separately
This sequence describes evidence and rules, not proof that marketing caused a purchase.
Use a consistent channel dictionary
Agree how paid search, organic search, email and referral traffic are named. Keep campaign parameters consistent in spelling and case. Treat an unknown source as unknown. Do not silently label every unmatched payment “direct,” because that makes a measurement gap look like a marketing channel.
Choose one attribution rule
A first-touch rule gives credit to the earliest qualifying touch you captured. A last-touch rule gives it to the latest qualifying touch before the outcome. More complex models divide credit. None of them can recover unobserved activity, and each answers a slightly different question.
Write the rule and lookback window next to the report. For a comparison across tools, align the available evidence as well as the model name. Two systems that both say “last touch” may still see different sessions or use different windows.
Compare your own channels
Use this calculator to check the arithmetic before automating the report. Add as many channel rows as you need. It compares supplied totals; it does not discover the missing link between a visitor and a payment.
Revenue attribution calculator
Compare visits, spend and revenue for the same period. Edit the illustrative example to see each channel’s return. This compares your totals, not individual payment journeys.
Customer and refund detailsOptional inputs for conversion rates, acquisition costs and net revenue
Use revenue and new customers from the same acquisition cohort. Do not mix older customers’ renewals into a new-customer comparison.
Full channel breakdownNet revenue, conversion rates, acquisition costs and more
- Net revenue
- £19,000.00
- Refunds
- £300.00
| Channel | Conversion rate | Visitor to customer | Revenue / visitor | Spend / new customer | ROAS | Net revenue |
|---|---|---|---|---|---|---|
| 2.50% | 0.52% | £0.66 | £46.51 | 2.70× | £5,200.00 | |
| 7.00% | 2.57% | £4.86 | £22.22 | 8.50× | £6,700.00 | |
| 10.00% | 4.17% | £11.83 | £48.00 | 5.92× | £7,100.00 |
More metrics: Google
- Cost / conversion
- £9.76
- Revenue share
- 27.98%
- Revenue / new customer
- £125.58
- Net revenue / visitor
- £0.63
- Net ROAS
- 2.60×
- Period break-even CAC
- n/a
- Contribution after ad spend
- n/a
More metrics: Reddit
- Cost / conversion
- £8.16
- Revenue share
- 35.23%
- Revenue / new customer
- £188.89
- Net revenue / visitor
- £4.79
- Net ROAS
- 8.38×
- Period break-even CAC
- n/a
- Contribution after ad spend
- n/a
More metrics: LinkedIn
- Cost / conversion
- £20.00
- Revenue share
- 36.79%
- Revenue / new customer
- £284.00
- Net revenue / visitor
- £11.83
- Net ROAS
- 5.92×
- Period break-even CAC
- n/a
- Contribution after ad spend
- n/a
What stands out
- Most traffic
- Most customers
- Most revenue
- Highest revenue per visitor
- Highest ROAS
Potentially under-invested: LinkedIn A candidate to investigate, not an instruction to increase spend.
Screening rule: at least 10 new customers, positive net RPV above the blended rate, and net revenue share above spend share. Highest qualifying net RPV leads; ties remain ties. This does not establish statistical significance, marginal returns or scalable demand.
n/a means a denominator is zero or an optional input is missing. Spend / new customer is channel-level acquisition cost, not fully loaded company CAC. Contribution after ads excludes fixed costs and taxes.
Save or share your result
Nothing is sent by this tool unless you choose to copy, save or share. A share link includes your inputs. Anyone with the link can read them.
The starting example adds illustrative spend, conversions and refunds to the table above. Change those figures to your own, using one period and one currency. The currency menu changes formatting, not exchange rates. A blank denominator produces “n/a,” not an invented zero return.
Read the result as a set of questions
Revenue per visitor helps compare the value associated with different traffic. Customer conversion rate shows how often that traffic becomes a new customer. Revenue per customer helps separate conversion quality from order size. Together they explain why a lower-volume channel can outperform a busy one.
CAC in this calculator is channel spend divided by new customers, not fully loaded company acquisition cost. ROAS is revenue divided by advertising spend, not profit. Include refunds and contribution margin before deciding whether a campaign can support more investment.
| What you see | What to inspect next |
|---|---|
| High traffic, weak customer conversion | Intent, landing-page promise, device mix and qualification |
| Few customers, high revenue per visitor | Order concentration, sample size and repeatability |
| Strong gross ROAS, weak net ROAS | Refunds, cancellations and the timing of adjustments |
| A promising low-spend channel | Available audience and marginal returns, not only historical averages |
A potentially under-invested channel is a hypothesis. A small audience may already be exhausted. One unusually large customer may dominate the result. Recent acquisition cohorts may not yet have converted. Investigate those explanations before changing spend, then test a modest increase against a defined success criterion.
Attribution describes assigned credit; it is not an incrementality experiment. If you need to know whether the advertising created additional demand, consider a controlled test appropriate to your scale rather than treating a source label as causal proof.
Build a weekly review that produces one useful change
- Check capture and matching health before reading channel rankings.
- Compare mature cohorts, and label recent cohorts as incomplete.
- Review customers, net revenue, cost and conversion together.
- Inspect the biggest change against campaign launches, site changes and refunds.
- Choose one action, name an owner and define when to review it.
Keep a brief decision log: what you saw, what you believed it meant, what you changed and what happened next. This helps distinguish a repeatable insight from a persuasive chart viewed on a good day.
The aim is a dependable decision loop, not a new spreadsheet ritual. Start with the smallest report that changes a real decision, then automate the stable definitions. If the definitions are still changing every week, automation will only make inconsistent answers arrive faster.