RICE prioritization for agency teams

Agencies have backlogs too. They're not product backlogs — they're queues of internal initiatives like "build a productized brand-audit offering", "implement a project management system upgrade", "raise rate cards by 10% on new contracts", "launch a senior advisory tier", "automate our reporting deliverables with AI". These compete for the same scarce resource: principal attention and senior staff time. RICE — with adapted inputs — produces a defensible ranking across them.

When RICE works for agencies

RICE is a good fit when:

• You have 8–20 internal initiatives competing for the same scarce capacity
• Items are roughly comparable in scope (not "build a new office" vs. "improve a single email template")
• You want a defensible ranking that survives partner debate

RICE is a poor fit when:

• The decision is fundamentally strategic (entering a new category, acquiring another firm) — use a SWOT or premortem
• Initiatives are deeply non-comparable in time horizon — score within timeframes (annual vs. multi-year) separately
• The agency is in survival mode where one or two decisions dominate the calendar

Defining Reach for agency initiatives

Three options, used in different contexts:

• Client-level Reach = number of client accounts affected (active engagements, retained relationships)
• Staff-level Reach = number of staff affected (when the initiative changes how work is delivered)
• Revenue-weighted Reach = sum of ARR/contract value affected (when revenue impact is the primary lens)

Pick one definition per scoring round. The most common error is unwittingly mixing definitions: scoring one item by client count, the next by staff count, the next by revenue. The resulting RICE numbers cannot be compared.

Defining Impact for agency initiatives

Tie Impact to a measurable agency outcome. The four most useful Impact targets:

• Utilization rate change in percentage points (e.g., +5 points = a "2" on the RICE scale)
• Gross margin change in percentage points
• Client retention annualized basis points (1% retention improvement is material)
• Staff retention annualized percentage points (turnover reduction)

Map to the 0.25–3 scale based on your agency's typical performance:

• 0.25 = barely measurable
• 0.5 = small but real (e.g., +1 utilization point)
• 1 = solid improvement (e.g., +3 utilization points)
• 2 = material (e.g., +5–7 utilization points or +3 margin points)
• 3 = transformative (e.g., +10+ utilization points or +5+ margin points)

Different agencies need different rubrics. Calibrate against your historical range.

Defining Confidence

Same 50/80/100% scale as product RICE, with agency-specific evidence types:

• 100%: you've piloted the change with a subset of clients or staff and have measured the outcome
• 80%: a peer agency has reported a similar change with a similar outcome, or you have a strong analogy
• 50%: informed intuition; no direct or analog evidence

Anything under 50% means the initiative needs a pilot or research step before it can be scored honestly.

Defining Effort

Effort = person-months of senior-staff time required. Distinguish billable-time loss (initiatives that pull senior staff off client work) from non-billable time (initiatives that consume nights/weekends or ops bandwidth). Both count, but for an agency the billable-time loss is the more painful currency — every senior week pulled off client work has direct margin cost. Surface this in the Effort calculation rather than averaging it away.

The override discipline

For agencies, the most common override is client-relationship-driven: a low-RICE initiative that a flagship client has named as critical. Override the ranking when the relationship value is real, but document the override and review whether the pattern is justified.

Also override for succession capacity: initiatives that distribute knowledge or expand the bench (mentorship programs, IP documentation) are systematically under-scored by RICE because their Reach and Impact compound slowly. Allocate 20–30% of capacity to these explicitly, separate from the RICE-ranked stack.

A worked example pattern

Agency leadership team has 12 candidate internal initiatives for the year:

1. Each initiative gets a one-page brief from its sponsor (PM equivalent — usually a partner or practice head)
2. Initial RICE scores drafted async by the ops lead
3. Leadership team review session — 90 minutes focused on disputed scores
4. Final ranking published with rubric, override list, and capacity allocation
5. Quarterly check-in: shipped initiatives compared against original Reach and Impact estimates; calibration adjusted

The pattern matches product RICE — the difference is who's in the room and what evidence counts.

Related

• RICE framework — the full framework page
• The RICE framework for prioritization — long-form guide
• RICE vs ICE — when to use each
• RICE for SaaS product teams — for software-product RICE