Revenue Attribution Calculator

How revenue attribution works

Revenue attribution connects an outcome to the acquisition evidence that preceded it. A source label says where captured traffic came from. A payment record says that money changed hands. An attribution rule decides how to connect the two. This calculator handles the arithmetic after you have assembled those channel totals; it does not discover or repair the underlying payment-to-visitor connection.

Start by choosing a reporting period and a single currency. Decide whether the rows describe revenue received during that period or revenue earned by a particular acquisition cohort. For customer acquisition comparisons, use newly acquired customers and revenue associated with that same group. Recurring payments from older customers should not quietly inflate the apparent return on this month’s acquisition spend.

Enter one row per channel or campaign, using the same definition throughout. Use visitors for the population you are evaluating, unique conversions for a chosen intermediate outcome, and new customers for the acquisition result. Spend should represent the costs you intend to compare. If you include only advertising spend, do not describe the result as fully loaded company CAC.

The example is editable and requires no login. You can add rows, copy the result, save a text report or explicitly create a share link. Shared links contain the entered figures in the fragment after the URL’s hash. Anyone with the link can read them. Use aggregate figures and avoid customer information or confidential campaign names.

Revenue per visitor: the formula and the meaning

Revenue per visitor = attributed revenue ÷ visitors. If a channel brings 1,000 visitors and £2,000 in revenue from the defined population, its RPV is £2. This combines two forces: how often visitors become customers and how much revenue those customers produce. It is often more revealing than either traffic or conversion rate alone.

Net RPV uses revenue after the refunds entered in the calculator. That adjustment matters when a campaign brings high initial sales but a substantial amount comes back later. Keep the refund policy consistent: either analyse cash movements in the period or adjust the original acquisition cohort, and explain which approach you chose.

A higher RPV does not automatically mean a channel deserves more budget. It may reflect a small sample, one unusually valuable customer or a channel serving people who were already close to purchasing. Look at customer count, acquisition cost and repeatability alongside the average. The next visitor may not be worth as much as the average visitor you have already acquired.

When visitors are zero, RPV is undefined. The calculator shows “n/a” instead of inventing an answer. If revenue exists without captured visitors, investigate the cohort, tracking coverage or matching process. The presence of money does not justify manufacturing a denominator.

CAC, RPV and ROAS answer different questions

CPA measures the cost of the conversion event you entered, such as a completed trial signup. CAC measures spend per new customer. If trials are cheap but rarely become paid accounts, CPA can look healthy while CAC is unsustainable. Always name the conversion rather than letting a generic label conceal the difference.

ROAS measures revenue against advertising spend. It is not profit and does not account for all costs of serving a customer. Optional gross margin makes the economics more useful: the calculator multiplies net revenue by the margin, then subtracts ad spend to estimate contribution after advertising. Fixed costs and taxes are excluded.

Period break-even CAC divides that pre-ad contribution by new customers. It is the acquisition spend per customer the supplied period can support under your margin assumption, not a lifetime spending allowance. Longer payback decisions require a separate retention and cash-flow model. Without a margin input, these contribution metrics stay “n/a.”

A worked example: the smaller channel can be more valuable

The starting data gives Google 8,200 visitors, 43 new customers and £5,400 revenue. Reddit has 1,400 visitors, 36 customers and £6,800 revenue. LinkedIn has 600 visitors, 25 customers and £7,100 revenue. These are illustrative figures, not results achieved by Premely or a customer.

Google wins on traffic and customer count. LinkedIn wins on total revenue and RPV, at about £11.83 per visitor. Reddit’s £800 spend produces 8.5× gross ROAS, ahead of LinkedIn’s roughly 5.92× and Google’s 2.7×. Those different winners show why a single leaderboard is not enough to choose the next action.

After the example’s £300 combined refunds, net revenue is £19,000. Across 10,200 visitors, gross blended RPV is about £1.89. The blended rate comes from total revenue divided by total visitors, not the average of the three channel rates. Averaging rates without their denominators would give the smaller channels too much weight.

Common mistakes that make the answer misleading

Mixing currencies. Adding pounds and dollars creates a meaningless total. The currency selector changes display labels only. Convert source figures using a documented policy before entry, or run a separate report for each currency.

Mixing new and existing customers. Renewal revenue can make an acquisition campaign look extraordinarily effective when its denominator includes only new customers. Define the cohort and use corresponding revenue, customers and spend. For an all-customer report, use a tool and denominator that explicitly support that population.

Ignoring time. A fresh trial cohort has not had the same opportunity to convert as a mature one. Give each group the same observation window, and label incomplete cohorts. Compare acquisition-date and payment-date reports separately rather than moving between them without notice.

Treating zero spend as an infinite return. Organic work still has costs, even if no media invoice exists. With zero entered spend, ROAS is undefined. The calculator displays “n/a.” Include relevant content or production costs if they belong in the decision you are evaluating.

Confusing credit with causation. A first-touch or last-touch rule allocates observed credit. It does not prove a purchase would disappear without the channel. Use the report to form a hypothesis, then design an appropriate controlled test when the decision requires causal evidence.

How to automate the work once the definitions are stable

A manual calculator is useful for checking your logic. The repetitive work lies upstream: maintaining campaign names, collecting trustworthy outcomes, connecting payments, handling adjustments and keeping unmatched revenue visible. Automate those pieces only after agreeing the definitions, or you will produce inconsistent answers more quickly.

Premely brings captured traffic, conversion context and connected revenue into a focused review. The quality of attribution still depends on supported integrations, valid matching evidence and the measurement window. It is not a promise that every historical payment will acquire a source label.