How-to
7 Steps to Calculate ROI for Agency Services
Last Updated: October 27, 2026
Quick Summary
Agency ROI is the net financial return from an agency engagement divided by the total investment in that engagement, multiplied by 100. To calculate it, total your full investment including fees, ad spend, tools and internal hours, then subtract that from attributed revenue plus cost savings.
- Total the full investment, not just the retainer, including ad spend, tools and internal labor.
- Agency ROI comes from two buckets: revenue generated and money you stop losing.
- Pick a measurement window before the campaign starts, such as 90 days for lead-driven services.
- Add a sensitivity row to your calculator to test what happens if attributed revenue drops.
- Compare ROI against your own history and alternatives rather than a single industry benchmark.
Table of Contents
- Step 1: Define What Agency ROI Must Answer
- Step 2: Total the Full Investment, Not Just the Retainer
- Step 3: Identify Revenue Generated and Cost Savings
- Step 4: How to Measure Marketing Campaign ROI With Attribution
- Step 5: Build a Digital Marketing ROI Calculator
- Step 6: Compare Against Marketing ROI Benchmarks
- Step 7: Review Agency Marketing ROI Examples
- Schedule a Free Consultation to Review Your Numbers
- Frequently Asked Questions
Step 1: Define What Agency ROI Must Answer
Calculating return on investment for agency services starts with one question: did the money you spent come back with more attached? That sounds simple until you try to answer it. Agency ROI is the net financial return from an agency engagement divided by the total investment in that engagement, and it exists to answer three things: was this worth it, should we renew, and where should the next dollar go?
Business owners may wonder whether a monthly retainer was worthwhile. The invoice is the easy part. The hard part is proving what came back.
Your ROI model needs to answer:
- What did we spend in total, not just the retainer?
- What revenue and savings can we trace to the work?
- Over what time period?
- What would we have earned without the agency?
Skip any of those and you're guessing with extra steps.
The Formula You'll Use
The core formula is net return divided by investment. The Agency for Healthcare Research and Quality's ROI tool defines ROI as net financial returns from improvement actions divided by the financial investment in those actions.
Harvard Business School Online describes the math this way: subtract the investment cost from the income generated, divide by the investment cost, then multiply by 100. A result above zero means the work paid for itself.
Simple version:
ROI = (Income generated - Total investment) ÷ Total investment × 100
One caveat worth knowing. A workforce program ROI analysis notes that the classic gain-minus-cost formula assumes a particular way of valuing an investment. It won't capture every cost or benefit.
Step 2: Total the Full Investment, Not Just the Retainer
Here's where most ROI calculations fall apart. The retainer is rarely the whole cost.
Add up everything the engagement actually consumes:
- Monthly agency fees for the full reporting period
- One-time setup, onboarding, or strategy fees
- Ad spend managed on your behalf
- Software, tools, or licenses the agency requires
- Your own team's hours spent on reviews, approvals, and asset delivery
- Content production, photography, or video costs
That last category catches people. If your marketing director spends six hours a month in agency meetings, that's a real cost. Price it at a loaded hourly rate and add it in.
Watch out: comparing a retainer against revenue without counting internal labor and ad spend inflates ROI and makes every renewal decision look better than it is.
If you're unsure what services actually cost beyond the retainer line, a guide to pricing can help you identify the components. And if you're weighing contract structures, a comparison of project-based and retainer pricing can explain how each changes your total investment picture.
Step 3: Identify Revenue Generated and Cost Savings
Agency ROI comes from two buckets: money that comes in and money you stop losing.
Revenue generated is the easier bucket to name and the harder one to prove:
- New leads that became closed deals
- Revenue from returning customers reactivated by campaigns
- Average order value increases from landing page or site changes
- Pipeline value from organic search and paid traffic
Cost savings is the bucket most businesses ignore:
- Hours saved through automation or agency-handled production
- Lower cost per lead than your previous approach
- Reduced spend on channels that weren't working
- Fewer wasted ad dollars after negative keyword and audience cleanup
Time saved counts, but only if you convert it to a monetary value. Multiply hours saved by a realistic loaded hourly rate. Don't count hours you never redeploy.
The rule: if you can't name the dollar figure and where it came from, leave it out. An honest ROI with fewer line items beats an impressive one you can't defend.
Step 4: How to Measure Marketing Campaign ROI With Attribution
Attribution is how you decide which results belong to the agency. Get this wrong and every other number is fiction.

Pick a measurement window before the campaign starts, not after. A common approach is a 90-day window for lead-driven services and a shorter one for direct-response campaigns. Write it down.
Then choose a model:
- Last-touch: credits the final click before conversion. Simple, but undervalues research and brand work.
- First-touch: credits the first interaction. Good for demand generation.
- Linear: splits credit across all touchpoints. Fairer, harder to explain.
- Position-based: weights first and last touches most heavily.
Most small and mid-sized businesses do fine with last-touch for paid search and a first-touch view for organic. Run both and compare.
Then filter out noise: internal traffic, bot clicks, leads that never qualified, and revenue from customers who were already in your pipeline.
Schedule a Free Consultation →
When Sales Cycles Are Long
If your average deal takes six months to close, a 90-day window will make every agency look like a failure. The fix is to measure leading indicators inside the window and lagging revenue outside it. Track qualified pipeline value, proposal requests, and demo bookings as early signals, then reconcile closed revenue on a rolling 12-month basis. Write both numbers down so you can defend the program before the revenue lands.
When Multiple Channels Contribute
Most deals touch search, email, social, and a sales call. No single-touch model tells the whole story. Use a position-based model for budget decisions and a linear model for post-mortems. If your CRM can store campaign source on the contact record, pull a multi-touch report monthly. If it can't, start with last-touch and add a self-reported "How did you hear about us?" field on your forms, imperfect, but better than guessing.
When Results Lag Delivery
SEO, content, and brand work often show returns months after the invoice. Set a lag assumption in your model: for example, assume 25% of attributed revenue arrives in months 1-3, 50% in months 4-9, and 25% in months 10-12. Then compare actuals to that curve each quarter. If the curve is wrong, adjust it, but write the assumption down first so you're not moving the goalposts after the fact.
For a fuller look at how to present and verify this data, see a guide to transparent digital marketing reporting.
Step 5: Build a Digital Marketing ROI Calculator
A spreadsheet beats a gut feeling every time. Here's a simple structure we'd use.
| Field | What to Enter | Example |
|---|---|---|
| Total investment | Fees + ad spend + tools + internal hours | Sum all line items |
| Attributed revenue | Closed revenue in the window | From CRM |
| Cost savings | Hours saved × loaded rate | Automation gains |
| Net return | Revenue + savings - investment | Calculated |
| ROI % | Net return ÷ investment × 100 | Calculated |
| Payback period | Months until cumulative return exceeds investment | Calculated |
Build it in five steps:
- Create one tab per quarter.
- Lock the measurement window at the top.
- Pull revenue from your CRM, not from memory.
- Add a sensitivity row: what happens if attributed revenue drops by a quarter?
- Review monthly, not annually.
That sensitivity row is the difference between a defensible ROI model and a hopeful one. If the project only works when everything goes right, you need to know that now.
If spreadsheets aren't your thing, a walkthrough on tracking marketing ROI without expensive software can cover a leaner setup.
Step 6: Compare Against Marketing ROI Benchmarks
A number in isolation tells you nothing. Benchmarks give it meaning.
Marketing ROI benchmarks vary widely by industry, channel, and how a business counts revenue. Rather than chase a single industry figure, compare against your own history and your own alternatives:
- Your ROI from the same period last year
- Your ROI from the previous agency or in-house effort
- Your cost per acquired customer across channels
- Your payback period against your cash flow needs
If paid search pays back in four months and organic takes eleven, both can be good. They serve different purposes. The mistake is judging them by the same clock.
For context on how ROI figures are commonly reported by industry, First Page Sage's SEO ROI statistics breaks results down by sector, including time to break even.
One more benchmark to track: retention. A campaign that brings in customers who stay is worth more than one that brings in customers who leave. Factor retention rate into your model when the data allows.
Step 7: Review Agency Marketing ROI Examples
Concrete examples make the model click. Here are three hypothetical patterns, with the mechanics spelled out so you can replicate the math.
Example 2: SEO for a B2B firm. The investment is mostly labor and content: $4,000 per month in agency fees plus $1,000 in internal review time, or $5,000 monthly.
Example 3: Web redesign. The investment includes design, development, and your team's review hours, say $15,000 one-time plus $1,000 in internal time.
Notice the pattern: the same formula, different clocks. That's why step 1 matters so much.
Add a Sensitivity Row to Every Example
Before you present any of these numbers, run a best/base/worst case. In Example 1, what happens if attributed revenue drops by 25%? ROI falls from 357% to 257%, still strong. In Example 2, what happens if closed revenue arrives three months later than planned? The program still breaks even by month 15 instead of month 12.
A few non-financial returns deserve a line in the model too: brand visibility, competitive advantage, and reduced dependence on paid channels. Don't assign them a dollar figure you can't defend. Note them and move on.
Schedule a Free Consultation to Review Your Numbers
Most businesses don't have an ROI problem. They have a measurement problem. The spend is real, the results are real, but nobody connected the two in a way anyone can defend at a budget meeting.
Real Web Marketing Inc. brings 35+ years of combined experience and hands-on involvement from founder John Eberhard, with direct access to experienced marketers rather than junior account managers.
As Brenda Silva of Silva Construction put it: "To have you as my online partner, it puts a sense of security there."
Get started with Real Web Marketing Inc. and work toward a number you can defend. Schedule a free consultation.
Frequently Asked Questions
What costs should be included when calculating agency ROI?
Include everything you spend, not just the retainer. That means agency fees, ad spend, software and platform subscriptions, implementation and setup costs, internal labor hours your team puts toward the work, and any one-time expenses like content production or web design. AHRQ's ROI tool defines ROI as net financial returns divided by the financial investment, so an incomplete cost total inflates your result. Track operating costs across the full reporting period, then compare that number against the revenue and savings the work produced.
How long does it take to see ROI from a marketing agency?
It depends on the service. Set a measurement window before work starts and judge results against it. Build a payback period into your model, meaning the month when cumulative returns equal cumulative investment. Reviewing actual results against projected returns at that checkpoint keeps the investment decision grounded in data.
How do you measure ROI when agency results are difficult to attribute?
Use a baseline and a defined attribution window. Record your lead volume, conversion rate, and revenue before the engagement, then compare the same metrics over a matching period afterward. Where multiple channels overlap, assign credit with a stated rule, such as last-click or a simple split, and keep that rule consistent. Document your assumptions so the model stays defensible. For services like SEO or social media, consider non-financial measures alongside financial value.
What is a good ROI for digital marketing services?
There is no universal number, and any provider quoting one should explain its source. The better approach is to compare your result against your own baseline and against the assumptions in your business case. First Page Sage publishes SEO ROI statistics broken down by industry, which can serve as a reference point, though it does not disclose methodology in the summary. Treat benchmarks as a starting range, then run a sensitivity analysis: if returns land 20% below projection, does the investment still make sense?