top of page

Paid Advertising Forecasts: How to Check an Agency’s Numbers

Writer: DIGITALTHROB
DIGITALTHROB
Oct 20
6 min read

A forecast of $9,000 revenue from $3,000 in ads can still leave your business with very little room for error. In the example below, fees and delivery costs reduce that headline to $900 before fixed overheads. One fewer completed job removes all of it.


Before approving an agency’s forecast, establish two things: whether the calculation works and whether you can afford to discover its assumptions are wrong. We use a fictional service business to show how that discussion should change the spending decision.


Separate correct maths from credible assumptions


A spreadsheet can calculate perfectly while relying on an enquiry rate nobody has achieved. Check the workings and the evidence separately.


For every input, ask the agency to record its source, date range, definition and limitation. A cost-per-click estimate concerns traffic cost. It does not validate how many visitors will enquire or how many jobs your team will complete. Historical sales data is stronger only when the offer, customer type, follow-up and measurement are comparable.


Our hypothetical business sells one $1,500 service. All figures in this guide are invented planning assumptions in USD, before applicable sales taxes. They are not DigitalThrob results or industry benchmarks. Price and fulfilment costs need checking against the business’s actual costing; the enquiry, qualification and completion rates remain unvalidated.


That last sentence belongs in the decision brief. Labelling a figure “base case” does not make it the most likely outcome.


Make every funnel stage reproducible


The proposed plan buys ads for 30 days. In this worked example, all modelled enquiries also arrive on days 1–30. A lead is one genuine, deduplicated enquiry; qualified means confirmed service, location, price and timing fit. For this forecast, a customer means a completed, fully paid job within 45 days of the enquiry. Deposits and signed work remain separate until then.


The base calculation is:


  1. $3,000 media ÷ $5 per click = 600 clicks.

  2. 600 clicks × 5% enquiry rate = 30 leads.

  3. 30 leads × 60% qualification rate = 18 qualified leads.

  4. 18 qualified leads × one-third completion-and-payment rate = 6 customers.

  5. 6 customers × $1,500 = $9,000 collected revenue.


Each rate has a different denominator. “Thirty conversions” tells us too little unless we know which stage those conversions represent. Our guide to tracking qualified leads from Google Ads explains how to connect enquiries with later business outcomes.


Models can produce fractional customers. A calculation of 5.4 is a scenario output, not five confirmed jobs plus an almost certain sixth. Keep the model’s precision for comparisons; report observed customers as whole, evidenced outcomes.


Find the headroom behind the revenue headline


Assume each $1,500 job costs $600 in variable fulfilment labour and materials. Its contribution before marketing and fixed overheads is therefore $900. Marketing is deducted afterwards, once.


The proposed cost is $3,000 media, $750 management covering the campaign and this cohort’s final review, $500 setup and $250 creative: $4,500 total. Confirm that scope in writing; later management, software or production charges cannot be omitted if they are required.


Six jobs produce $5,400 contribution, leaving $900 after the listed marketing costs. Five jobs produce $4,500 and leave nothing for fixed overheads. Four leave a $900 shortfall. The base case has one job of headroom before even those overheads are considered.


Media-only acquisition cost is $3,000 ÷ 6 = $500 per customer. Including the listed marketing costs, it is $750. Neither label should conceal additional sales labour, commissions or other relevant costs. Add them before deciding what return is acceptable. Future repeat purchases also need evidence; they cannot automatically rescue weak first-job economics.


Use scenarios to expose the downside


Keeping media at $3,000 and listed marketing costs at $4,500 gives three conditional calculations:


  • Low: $6 per click produces 500 clicks. At 4% enquiries, 50% qualification and 20% completion/payment, that becomes 20 leads, 10 qualified leads and 2 customers. Revenue is $3,000; contribution after marketing is −$2,700.

  • Base: $5 per click, 5% enquiries, 60% qualification and one-third completion/payment produce 6 customers. Revenue is $9,000; contribution after marketing is $900.

  • High, before capacity: $4 per click produces 750 clicks. At 6.4% enquiries, 62.5% qualification and 40% completion/payment, that becomes 48 leads, 30 qualified leads and 12 jobs. Revenue would be $18,000; contribution after marketing would be $6,300, if delivery were possible.


These are not probability ranges or statistical confidence intervals. The high calculation improves several assumptions together, while even the low case includes sales. With no customers, the listed marketing loss is $4,500. Cancellation costs or other committed expenditure could increase the exposure.


Test the weak assumption, then choose the next action


Change one base input at a time to see what could reverse the decision. Qualification falling from 60% to 40% leaves 12 qualified leads and four customers: −$900 after marketing. The owner should investigate rejection reasons and offer fit before buying more traffic.


If qualification holds but completion/payment falls from one-third to one-sixth, three customers leave −$1,800. Review where qualified leads stall: quoting, acceptance, delivery or payment. Increasing the click budget cannot identify that bottleneck.


If cost per click rises from $5 to $6 while all other base rates hold, 500 clicks become 25 leads, 15 qualified leads and five customers. Marketing costs are just covered. That makes traffic cost worth monitoring, but the response still depends on actual downstream quality.


Choose an investigation that addresses the weakest evidence. A smaller test may reveal broken enquiries or obvious poor fit; it may produce too few completed jobs to estimate a dependable completion rate. Agree what it can answer before treating its results as permission to scale.


Apply capacity and a consistent maturity window


This business can deliver only eight additional jobs during the forecast period. The high calculation therefore exceeds capacity. At eight completed jobs, revenue is $12,000 and contribution after the unchanged $4,500 marketing cost is $2,700. Later delivery or extra staff requires a revised model, including changed costs and customer expectations.


The 45-day rule also needs an exact clock. For each enquiry, count only jobs completed and fully paid within 45 days of that enquiry. With enquiries assumed to arrive on days 1–30, the final cohort review is on day 75, when the latest leads have reached 45 days. Use each lead’s own 45-day window, so earlier leads do not receive extra observation time in the comparison. If a person clicks during the campaign but enquires later, their enquiry date starts the clock and their review may fall after day 75; keep the original advertising interaction recorded separately.


Keep later outcomes visible separately. At day 30, unresolved leads are pending; they are neither failures nor assumed wins. Cash planning needs a separate schedule showing when deposits and balances arrive against advertising bills, fees and delivery payments. Eventual positive contribution does not fund today’s cash gap.


Write a decision brief that actually decides


Here is the completed brief we would put in front of this fictional owner:


  • Decision: hold the proposed launch. The unvalidated base case leaves just $900 before overheads, and the owner can afford a maximum $2,000 marketing loss. The $4,500 zero-customer exposure exceeds that limit.

  • Before reconsidering: the owner checks job contribution and available cash; the agency documents the evidence behind each rate and all non-cancellable costs. Both confirm the eight-job capacity limit.

  • Possible next proposal: a revised pilot whose total committed marketing exposure, including fees and setup, fits within $2,000. A smaller media budget alone is insufficient if fixed charges already consume the allowance. No launch is approved until the revised scope and learning question are written down.

  • Controls if approved: the agency monitors the agreed spending cap and enquiry route; the owner records qualification, delivery, payment and rejection reasons. Broken enquiries, exhausted capacity or threatened overspend trigger a pause.

  • Next funding decision: review comparable mature outcomes, updated costs and cash before authorising another cycle. Five jobs merely cover the original listed costs; six do not automatically justify continuation. Recalculate the next cycle’s actual fees and required contribution, and hold if unresolved evidence or affordable loss still blocks it.


The decision uses the business’s exposure and evidence, rather than the most attractive scenario. When comparing agency proposals, ask each agency to make those trade-offs equally visible.


Our paid advertising management starts with the relationship between the offer, measurement and business outcomes. Bring your costs, capacity and existing lead records so we can discuss what a useful forecast needs to establish.


 
 
 

Comments


Discover clics solution for the efficient marketer
More clics

Never miss an update

bottom of page