Local Marketing ROI: Cost per Lead, Customer Acquisition and Break-Even Calculations

By BizLookupPro · Published 2026-10-28

Compare the full cost of a marketing activity with the contribution from the work linked to it. Keep raw contacts, suitable enquiries, customers and paid jobs separate, and use the same period, currency and cost basis throughout.

Revenue on its own can make a weak channel look good. A campaign can produce a big sales total and still leave too little, after delivery costs and marketing spend, to support the business.

This guide gives you a worked model and a decision worksheet. Every figure is a fictional arithmetic example. None is a benchmark, a forecast, tax advice or a reported client result.

Define the outcomes before calculating rates

A phone-link click belongs to a different certainty level from a received request. A returning customer belongs to a different acquisition category from a new customer.

Write the definitions down once. Otherwise a supplier and an owner can both calculate cost per lead correctly and still disagree, because they divide by different things.

Build the full cost of the trial

Separate cash spending from an economic allocation of time. Both can matter, but they answer different questions.

A useful cost register can include placement/ad fees, setup, creative work, management, tracking costs and the time spent handling the channel. Allocate shared costs deliberately rather than quietly assigning all existing business overhead to one small experiment.

Do not count the same labour twice. If job-delivery labour is already part of variable fulfilment cost, it should not also appear as marketing administration.

Keep one currency and a consistent treatment of tax, refunds and fees. Ask the appropriate bookkeeping adviser how the model should align with your actual records; the calculations below deliberately do not decide recoverability or tax treatment for a jurisdiction.

The core calculations

Use the same agreed cost basis in related comparisons.

Cost per suitable enquiry = total channel cost ÷ suitable enquiries.

Observed acquisition cost = total channel cost ÷ new customers counted under the agreed rule.

Job contribution before marketing = revenue from the job − relevant variable delivery cost.

Contribution after marketing = associated job contribution − total channel cost.

Contribution-based marketing return = contribution after marketing ÷ total channel cost.

If a denominator is zero, record the result as undefined, with no observed outcome. Do not record it as zero cost. With no suitable enquiries, the spend is still a cost, and you have no usable cost per suitable enquiry.

These calculations describe the trial as recorded. They do not prove that the channel caused every outcome attributed to it.

Worked example: a directory placement trial

A fictional specialist cleaning service is evaluating one placement. The amounts are illustrative USD values and use a consistent basis throughout.

The $325 trial cost includes $275 cash and $50 allocated time. Report both if cash availability and economic sustainability need separate decisions.

The relevant calculations are:

• Cost per suitable enquiry: $325 ÷ 20 = $16.25.
• Observed acquisition cost: $325 ÷ 6 = $54.17, rounded.
• Contribution per counted job: $180 − $95 = $85.
• Contribution from six counted jobs: 6 × $85 = $510.
• Contribution after the trial cost: $510 − $325 = $185.
• Contribution-based return: $185 ÷ $325 ≈ 56.9%.

The $185 is not the company's final accounting profit. Other overhead, finance, tax, capacity and unallocated costs have not been settled by this small channel model.

The owner should also ask how the source was established, whether these were incremental customers and whether the sales cycle has had time to mature.

Why revenue and ROAS tell a different story

The example produces $1,080 revenue from the six counted jobs. Dividing that revenue by the $275 cash advertising fee gives an illustrative revenue-on-ad-spend ratio of about 3.93.

That is a different calculation from the contribution return. It excludes variable delivery cost and uses cash advertising spend rather than the fuller $325 economic channel cost.

Both numbers can be arithmetically correct and still answer different questions. Whenever you show a figure, name its denominator and the costs it includes.

A supplier reporting an attractive revenue multiple should be able to explain how revenue was associated with the activity and which costs were left out.

Test the effect of job economics

Suppose the same six jobs still produce $180 each, but variable delivery cost is $145 rather than $95. Contribution falls to $35 per job.

Six jobs then contribute $210 before marketing. After the $325 trial cost, the contribution is −$115. The revenue total and revenue-on-ad-spend figure remain unchanged.

That is why a channel has to be judged against the business's delivery economics, not just an enquiry count or a big sales total.

Calculate a practical break-even point

With $325 trial cost and $85 contribution per new counted job, the model needs at least four such jobs to cover that trial cost: round $325 ÷ $85 up to the next whole job.

At $35 contribution, it needs at least ten. These are trial-level arithmetic thresholds before the other costs excluded from the model.

You can also estimate a maximum enquiry cost from an assumed suitable-enquiry-to-paid-customer rate:

Contribution available per suitable enquiry = expected contribution per customer × the assumed conversion rate.

Using $85 and a provisional rate of 6/20 gives $25.50 per suitable enquiry before allocating other required costs or a desired margin. This is a planning allowance based on the assumptions, not a safe bid or a guaranteed market value.

If the cohort is small or unfinished, test several rates rather than treating the observed fraction as a permanent characteristic of the channel.

Keep the enquiry cohort and outcome window aligned

Choose which enquiries belong to the trial and when outcomes are assessed. Work won this month can come from last month's requests, and some recent enquiries may still be undecided.

For the example, two quoted requests might remain open at the review date. Report that fact and revisit the cohort later. Do not mark them lost merely to complete a report, and do not count their potential value as collected revenue.

A useful review table records received date, source evidence, suitability, quotation state, outcome date and counted value/cost. Keep private customer details in the protected business system rather than the public reporting sheet.

Handle attribution without inventing certainty

A customer can see a listing, search the brand later, ask a friend and finally telephone. The last recorded route is not necessarily the introduction that caused the decision.

Use permitted tagged links, customer-reported source where appropriate and operational enquiry records. Preserve unknown and conflicting information. Avoid forcing a request into whichever campaign needs a better result.

GA4 acquisition reports describe collected traffic-source information; they are not a complete record of offline jobs or causal marketing impact. GA4 Traffic acquisition documentation

The UTM guide explains a maintainable link convention. The GA4 guide explains why contact clicks, accepted forms and actual business enquiries can differ.

Compare channels on a fair basis

Use comparable definitions, periods and cost allocations. Check whether one channel supplies urgent small jobs while another supplies slower recurring contracts. Those can need different outcome windows and contribution models.

Distinguish a free listing's maintenance cost from an optional placement fee. If the business would keep the free profile anyway, evaluate the incremental paid trial as well as the whole channel relationship.

Capacity matters too. Additional low-contribution work can displace better work or require extra staffing. A channel that looks profitable in a quiet month may need different handling when the team is full.

Make the next decision explicit

At the review date, choose maintain, improve, continue a bounded test or stop spending. Record the reason and the uncertainty.

Examples of useful decisions include repairing a poor enquiry route before buying more exposure, improving service scope because requests are unsuitable, or continuing a trial until a normal sales cycle has matured.

Do not renew automatically because a report looks busy. Ask what the next period is meant to establish, and what spending and time limit applies.

A reusable channel review sheet

Keep the arithmetic attached to its assumptions. A modest, clear model that supports a real decision beats an impressive return figure nobody can reconcile to the business records.