How to compare an agency with an in-house search ads team
The visible comparison is often an agency fee versus a marketer salary. That leaves out setup, account handoff, retained oversight, tools, training, cross-functional support, and the value created by different campaign performance. An agency can have a lower entry cost but a media-spend fee that scales with budget. An in-house model can require more hiring and onboarding cost but become structurally cheaper at a larger or longer-running account.
This calculator holds monthly media spend constant, then connects each model to its own average CPC and click-to-conversion rate. It values those conversions with contribution after variable product, payment, fulfillment, and refund cost. The result shows total cost, fully loaded CPA, contribution ROAS, ROI, the budget where cost matches, the budget where economic profit matches, and the month when cumulative advantage can change.
Every default is a fictional demonstration value. It is not a fee, compensation, CPC, conversion-rate, or performance benchmark for any market, platform, industry, or agency.
Complete cost boundary
Include setup, retainer, media fees, loaded labor, tools, training, and internal support time.
Comparable performance
Use average CPC and conversion rate from the same period, campaign scope, and conversion definition.
Contribution, not revenue
Value a conversion after costs that rise with each sale, lead, booking, or completed service.
Align the inputs before trusting the result
A mathematically precise result is still misleading when the two alternatives use different scopes. The agency case should include strategy and migration setup, fixed retainer, media-spend percentage, required recurring services, and the internal hours that remain for reviews, approvals, asset handoffs, and meetings. If performance bonuses or mandatory creative services are quoted separately, convert them to the same monthly or upfront boundary before entry.
The in-house case should use allocated FTE rather than automatically charging one full salary. Annual loaded compensation should reflect employer cost, not base salary alone. Add recruiting, onboarding, account transfer, bidding and analytics tools, training, and support from design, development, data, legal, or approval teams. Use one consistent tax basis across vendor and internal costs.
Aligned input boundary for agency and in-house search ads| Area | Agency case | In-house case | Evidence |
|---|
| Setup | Strategy, account, tracking, and handoff | Hiring, onboarding, and knowledge transfer | Quote, recruiting ledger, project records |
| Recurring management | Retainer, media fee, retained oversight | Allocated labor, tools, training, support | Contract, payroll cost, time logs |
| Performance | Observed or evidenced CPC and conversion rate | Pilot or internal CPC and conversion rate | Platform, CRM, order, and offline outcome data |
| Value | Same contribution per conversion | Same contribution per conversion | Product or service contribution ledger |
Core formulas and metric definitions
Clicks, conversions, and contribution
Monthly clicks = monthly ad spend ÷ average CPC.
Monthly conversions = monthly clicks × conversion rate.
Monthly contribution = monthly conversions × contribution per conversion.
Profit, ROAS, and ROI
Net profit = total contribution − media and management cost.
Contribution ROAS = total contribution ÷ total cost.
ROI = net profit ÷ total cost × 100.
Google Ads defines average CPC as total click cost divided by total clicks and conversion rate as conversions divided by eligible ad interactions in the same period. A conversion rate can exceed 100% when several actions or every occurrence is counted, which is one reason both operating models need the same conversion action and counting settings. Platform conversion value can be revenue or a configured value; it should not be assumed to equal contribution.
Fully loaded CPA divides media, setup, and management cost by conversions in the analysis horizon. Management CPA excludes media and isolates the operating-model layer. If conversions are zero, CPA is undefined rather than infinite. If total cost is zero, ROAS and ROI are also left undefined.
Cost break-even and economic break-even answer different questions
The cost-only monthly spend break-even uses agency setup, fixed management, media-spend fee, and in-house setup and management cost. It asks where the two operating structures cost the same before performance is considered. A higher in-house fixed cost can favor an agency at a smaller account, while a percentage-based agency fee can make in-house management cheaper as spend scales.
The economic monthly spend break-even adds CPC, conversion rate, and contribution per conversion. Lower CPC or higher conversion rate can offset a more expensive operating structure. Better in-house product knowledge or first-party data can move the threshold the other way. The two break-even figures are equal only when agency and in-house CPC and conversion rate are equal.
Cost, economic, and cumulative crossover interpretations| Output | Included | Question answered | Main caution |
|---|
| Cost-only spend break-even | Setup, fixed management, media fee | At what spend does management cost match? | Does not value performance differences |
| Economic spend break-even | Cost plus CPC, conversion, contribution | At what spend does net profit match? | Needs defensible performance evidence |
| Cumulative crossover month | Upfront gap and recurring profit gap | When can the initial decision reverse? | Assumes a stable monthly run rate |
A missing finite threshold is not an error. Under the entered linear assumptions, one model may lead across every non-negative spend level. The result labels that direction instead of displaying a negative, decision-irrelevant threshold.
Step-by-step workflow
- Choose one decision horizon and shared media budget. Use the contract term, hiring ramp, or another period that matches the actual decision.
- Convert the agency quote into fully loaded cost. Include setup, retainer, media-spend fees, mandatory recurring services, and retained internal oversight.
- Allocate the in-house workload. Apply the search-ads FTE to annual loaded compensation, then add hiring, tools, training, and support hours.
- Enter comparable CPC and conversion rates. Prefer observed agency history and an in-house pilot or a documented internal baseline.
- Estimate contribution per conversion. Reconcile reported conversions to completed sales, qualified leads, bookings, or another outcome and deduct variable fulfillment cost.
- Read current advantage and thresholds together. A small current lead can reverse just beyond the selected spend or horizon.
- Stress the assumptions. Review the fixed sensitivity table and rerun separate budgets, seasons, channels, or product groups where a constant run rate is weak.
Worked USD example from the fictional defaults
The English example uses USD 20,000 of monthly spend, USD 4 average CPC, a 4% conversion rate, and USD 250 contribution per conversion for both models over 12 months. The agency case uses USD 4,000 setup, USD 2,000 monthly retainer, a 12% media fee, and eight oversight hours at USD 60 per hour. The in-house case uses USD 15,000 setup, 0.5 FTE at USD 60,000 annual loaded compensation, USD 700 monthly tools, USD 2,400 annual training, and eight support hours per month.
Fictional USD agency and in-house search ads result| Result | Agency | In-house | Interpretation |
|---|
| Monthly management cost | USD 4,880 | USD 3,880 | In-house recurring cost is USD 1,000 lower |
| Twelve-month total cost | USD 302,560 | USD 301,560 | In-house is USD 1,000 lower in the horizon |
| Total contribution | USD 600,000 | USD 600,000 | Equal CPC and conversion rate produce equal value |
| Net profit | USD 297,440 | USD 298,440 | In-house leads by USD 1,000 |
| Fully loaded CPA | about USD 126.07 | about USD 125.65 | Media, setup, and management cost included |
The threshold view changes the discussion
Cost-only and economic spend break-even are about USD 19,305.555556 per month. The agency would need about 4.006667% conversion at the current CPC, its maximum media fee would be about 11.583333%, and the in-house loaded-compensation ceiling would be USD 62,000 per FTE at the entered allocation. The lower agency setup cost gives an initial lead, but in-house recurring profit is USD 1,000 higher per month, so cumulative profit crosses at month 11.
Use the reverse calculations as decision tests
Required agency conversion rate
This is the agency conversion rate needed to match in-house net profit at the current spend, CPC, cost, and contribution assumptions. Treat it as an evidence question, not a service-level guarantee. Ask whether account history, a comparable cohort, landing tests, and offline outcome data support the target. A result above 100% signals that conversion rate alone cannot realistically close the gap.
Maximum agency media fee
This is the media-spend percentage that makes agency and in-house net profit equal while other inputs remain unchanged. A negative result means that even a zero media fee would not close the fixed-cost and performance gap. Retainer, performance bonus, mandatory creative, and taxes still need to remain inside the same quoted boundary.
Maximum in-house loaded compensation
This is the annual loaded compensation per full-time equivalent that makes the entered FTE allocation match agency profit. Compare employer cost rather than salary alone. If allocated FTE is zero, compensation has no effect on the model and no threshold is returned.
Practical planning scenarios
Agency renewal
Enter the renewal quote and three to six months of aligned CPC, qualified conversions, and contribution. Use the fee and conversion thresholds to frame due-diligence questions and negotiation ranges.
First performance-marketing hire
Include hiring, onboarding, account transfer, loaded compensation, tools, and support. Compare the cumulative crossover with available cash and the realistic time required for a new owner to stabilize the account.
Budget expansion
Rerun the model at several spend levels. Do not assume CPC and conversion rate remain constant as budget expands; use scale-test evidence where available.
Brand and product segmentation
Brand search, non-brand search, high-margin products, and lead-generation campaigns can have very different CPC and contribution. Model material groups separately before combining the operating decision.
Sensitivity table and model limits
The agency downside scenario lowers agency conversions and increases agency management cost by the selected stress rate. The in-house downside does the same to the internal model. The shared performance and shared management-cost cases test a broad market or cost shock. These are deterministic stresses, not predicted probabilities or confidence intervals.
Important limitations
- The model repeats one monthly budget, CPC, conversion rate, and contribution value across the horizon.
- It does not forecast auction learning, seasonality, competition, brand-search overlap, or declining marginal efficiency at higher spend.
- It does not score agency or employee capability, turnover, account access, creative quality, or data quality.
- It does not replace incrementality experiments or multi-touch attribution that separates paid impact from organic demand.
- It does not determine tax, accounting, privacy, employment, advertising-policy, contract, data-ownership, or handoff compliance.
Pre-comparison checklist
- Agency and in-house data cover the same period, campaign set, location, device, and brand-search boundary.
- Average CPC uses comparable cost, credit, invalid-click, and tax treatment.
- Conversion action, counting method, conversion window, duplicates, cancellations, refunds, and offline outcomes are aligned.
- Contribution per conversion deducts variable cost and does not double-count fixed cost already entered elsewhere.
- The agency quote includes minimum fees, mandatory tools or creative, reporting, and retained internal oversight.
- The in-house case includes loaded compensation, allocated FTE, hiring, training, tools, and cross-functional support.
- Account ownership, tracking data, source assets, confidentiality, and exit handoff are documented.
- More than one performance, cost, budget, and time scenario has been reviewed.
Official measurement references
Google Ads official Help definitions for average CPC, conversion rate, conversion value, and ROI were checked on August 5, 2026. They are used only to define measurement boundaries. They do not provide agency fees, loaded compensation, future CPC, conversion performance, or an operating-model recommendation.
Frequently asked questions
Can I compare only the percentage agency fee with salary?
No. Agency setup, retainer, mandatory services, and retained oversight can materially change the result. In-house hiring, loaded compensation, tools, training, and support must also use the same complete boundary.
Should I simply choose the higher ROAS?
Contribution ROAS is useful, but it does not capture capability risk, account ownership, organizational learning, hiring delay, or contract exit. Review total profit, thresholds, crossover time, and qualitative controls together.
What if I do not know CPC or conversion rate?
Prefer a short pilot, aligned account history, or an agency range supported by comparable evidence. Run conservative, base, and upside cases rather than inventing one market average.
Why do the cost and economic break-even budgets differ?
The cost threshold excludes performance, while the economic threshold values CPC, conversion rate, and contribution. They match when CPC and conversion rate match; otherwise the economic threshold moves or may disappear.
What does no finite crossover mean?
Under the entered upfront and recurring monthly assumptions, time does not reverse the leading model. Real budgets, staffing, contract terms, and performance can change, so long horizons should still be rerun with staged assumptions.
Can I use this outside Google Ads?
Yes. The model is platform-neutral for CPC-based search advertising. Reconcile each platform definition for cost, clicks, conversions, credits, taxes, and attribution window before comparison.
Replace the fictional defaults with aligned evidence
Enter the agency quote, fully loaded in-house cost, comparable CPC and conversion rate, and contribution per completed outcome. Review the current lead together with budget thresholds, required performance, allowable cost, sensitivity, and the time when cumulative economics can change.