Generic services RFPs miss the things that determine whether an agency retainer succeeds. This template centres on the four areas where most engagements fail: how the account team is staffed and substituted, how scope is set so monthly retainer hours do not silently overrun, how the reporting cadence holds the agency accountable, and how attribution maps to the metrics your finance team already trusts. Built on the standard 10 section RFP frame, with agency specific requirements bolted into each.
The marketing world over uses the RFP. Sending a 30 page document to five agencies for a $40K social media project is heavy procurement infrastructure for a light decision. The trigger to use the structured RFP rather than a 3 way pitch competition or sole sourcing should be one of the following.
If none of the above apply, a 3 way pitch with a brief, a shared budget range, and a 2 hour chemistry call per agency will reach the same decision in two weeks rather than eight. For framework guidance read the ANA agency-resources hub which publishes the industry-standard search playbook.
Most agency retainers go over hours not because the agency is greedy but because scope was set in vague language during the pitch. The fix is to write the monthly deliverables list with the same specificity a SaaS pricing tier uses. Each deliverable has a count, a cadence, a definition of done, and a defined revision allowance.
| Service line | Deliverable | Cadence / count | Revisions included |
|---|---|---|---|
| Paid media | Campaign builds (Meta, Google, LinkedIn) | Up to 6 new campaigns / month | 2 rounds per campaign |
| Content marketing | Long-form blog posts (1,500+ words) | 4 / month | 2 rounds per post |
| Marketing emails (HTML build + copy) | 4 / month | 2 rounds per email | |
| Creative | Static social assets (Meta / LinkedIn) | 16 / month, 4 variants of 4 concepts | 1 round per concept |
| Creative | Short-form video (15 to 60s) | 2 / month, includes 1 round of edits | 1 round per asset |
| SEO | On-page audits + recommendations | 1 priority site section / month | Review call included |
| Strategy | Quarterly review + planning workshop | 1 / quarter, 2 hours | Pre-read 48 hours prior |
| Reporting | Monthly performance report + 30min readout | 1 / month | Standard template, agreed Q1 |
This is a sample scope; substitute the deliverables your business actually needs. The point is the structure. When the agency proposes a $15K monthly retainer against this table they are signing up to exactly what is listed, and you both have a baseline against which to discuss change requests. Anything not on the list is either an explicit upcharge (typically billed at the blended retainer hourly rate) or a swap (drop X to add Y this month, agreed in writing). The 4As publishes guidance on monthly retainer scoping in their agency business practices library.
The number one complaint about agency engagements is that the senior people who pitched are not the people who deliver. Pitch teams are stocked with founders and creative directors; delivery teams default to mid level account leads and junior designers. The agency is not lying when it pitches with seniors; the seniors really do pitch. They just do not stay on the account once it is sold.
The RFP control is to require named team disclosure with seniority and monthly hours, and a contractual no substitution clause for the first 6 months. Specifically include:
Agencies push back on substitution penalties because their account staffing is fluid and substitutions happen for reasons that are not predatory (someone leaves the agency, parental leave, project conflicts). The negotiable version is: penalty applies only when the substitution is not explained, when the replacement is more than one level junior, or when the substitution occurs in the first 90 days. Reasonable agencies accept this; agencies that fight every version of the clause are signalling that bait and switch is their operating model.
Agency reporting is a place where the wrong defaults lead to wasted time. Default agency reports are 30 page slide decks full of vanity metrics. The reports that make a difference are short, comparable month over month, anchored on the metrics that the CFO will see in the quarterly review, and accompanied by an opinion on what to do differently next month.
One page summary + appendix. Summary shows: spend, primary KPI (leads or pipeline or revenue depending on funnel stage), CPA, CPL, and one paragraph of agency opinion on what changed and what to do next. Appendix has the detail. Standard structure agreed at engagement start so each month is comparable to the last.
90 minute working session, not a presentation. Pre-read sent 48 hours prior. Format: agency proposes 3 strategic shifts for next quarter, client team weighs in, decisions logged. The QBR is the strategic recalibration; the monthly is the tactical review.
Half day or full day session, ideally in person or video. Outputs: next year channel mix, retainer size, scope adjustments, named team confirmation, and KPI targets. Inputs include the previous 12 months of monthly reports so the workshop is data-led not opinion-led.
Looker Studio, Tableau, or whatever your stack already runs. Live numbers from the channels the agency operates. Not a replacement for the monthly report but a way to spot anomalies in real time. Agency should set this up in the first 60 days; ongoing maintenance is included in the retainer.
Agency attribution is the area where pitches are most likely to make claims you cannot validate. An agency that pitches a 4x ROAS without telling you the attribution model is pitching a story not a number. Settle the model in the RFP. Three approaches to choose from, listed in order of complexity.
The RFP requirement is: agencies disclose which attribution model their proposal assumes, and they commit to reporting against your existing finance team's model rather than introducing a new one. Where the agency model disagrees with the finance model, the variance is explained, not hidden. Useful public reference on the trade offs: Nielsen on marketing mix modeling and Google Analytics 4 conversion attribution.
Agency pricing falls into four models. Each rewards different agency behaviours and creates different misalignments. The RFP should ask the agency to propose against your preferred model, justify it, and disclose the blended hourly rate even when pricing is structured as a retainer or fixed fee.
Most common for ongoing engagements. Fixed monthly fee against an agreed scope (see Part II). Best when the work is predictable; risky when scope inflates silently. Always anchored to a defined hours allotment; overflow billed at agreed hourly. Demand the blended hourly rate so you can sanity check the retainer math.
Best for discrete deliverables with clear scope (a brand refresh, a website launch, a campaign). Agency carries the risk on internal cost; client carries the risk on scope creep. Change order process must be clear. Common cost benchmarks: brand refresh $40K to $150K depending on category, website launch $50K to $300K, campaign launch $30K to $100K.
Hourly billing against a blended or roled rate card. Honest but creates an incentive misalignment (agency profit grows with hours). Best for genuinely unpredictable scope (exploratory work, fast moving project). Demand weekly burn rate reports and a not to exceed cap.
Some or all of the fee tied to outcomes (CPA, ROAS, qualified pipeline). Sounds appealing; complicated to operate. Agency only accepts if the attribution model is clean and the targets are realistic. Often blended (60 percent base retainer, 40 percent performance). Useful for direct-response work; rarely viable for brand work.
Strategic fit and relevant case studies dominate. Cost is meaningful but rarely the differentiator inside a competitive set, because four agencies pitching a $120K retainer will land within 15 percent of each other.
See the full mechanics in RFP Evaluation Criteria and the master template at requestforproposaltemplate.com.