SOP: [Service Name]

One-liner: [Outcome-focused sentence for what this playbook delivers — e.g. “Ops and data review actions for pre-, mid-, post-experiment stages.”]

Last updated: [Date]

Domain / Service line: [Data Platform & Analytics | AI Automation & Copilots | Training] / [e.g. Intake Optimization]

Tech anchors: [Snowflake, dbt, Fivetran, Omni/Rill, M365, Slack — whatever delivery needs to know]

How to use this doc

[1–3 sentences: who this is for and when to use it. E.g. “This document serves as the playbook to run [service] from kickoff to handoff. Use it for staffing, phase checklists, and SOW copy. What’s in [brackets] varies by engagement; the rest is mandatory for each run.”]

Optional: Note what varies vs what’s mandatory (e.g. “Purple/case-specific vs black/mandatory” if you use color in Notion).

Scope & configuration

Default bundle (included):

[Module A] — [what it delivers]
[Module B] — [what it delivers]
[Module C] — [what it delivers]

Optional add-ons:

[Module D] — [when to use]
[Module E] — [when to use]

Prereqs

Data & access: [sources, warehouse, SSO, APIs — what client must provide]
Environment: [client cloud vs Brainforge, environments needed]
SMEs: [names/roles for sign-offs — e.g. “Marketing owner (attribution), analytics/engineering contact”]

➡️ Phase 0 / Pre- [e.g. Audit · Discovery · Pre-Experiment]

[One line: what this phase is for — e.g. “How large is this problem area? Scope out qualitatively and quantitatively.”]

Duration: [e.g. 1 week]

Outcome: [e.g. Discrepancy report, scope agreed, hypothesis and criteria defined]

Checklist:

[Step 1]
[Step 2]
[Step 3]
Gate: [Sign-off or deliverable required before Phase 1]

DRIs (if applicable): [e.g. “DRI: Data engineering for pipeline; DRI: PM for client sign-off”]

Links/templates: [e.g. Link to proposal template, audit checklist, or “A/B Criteria Design” doc]

➡️ Phase 1 / Live [e.g. Pilot · Build · Rollout]

[One line: what this phase is for — e.g. “Day-to-day operations from launch to sign-off.”]

Duration: [e.g. 2–4 weeks]

Outcome: [e.g. Working pilot, validation, experiment live]

Before go-live, ensure:

[Pre-flight 1 — e.g. 90% of users on required version]
[Pre-flight 2 — e.g. Dashboard and success metrics defined]
[Pre-flight 3 — e.g. Bug tracker and escalation path in place]
Gate: [Sign-off or deliverable required before Phase 2]

Daily / weekly must-haves:

[e.g. Overview of success + check metrics for treatment vs control]
[e.g. SOP for doomsday thresholds and escalation when something is in danger]
[e.g. Link to day-to-day check-in doc or Bayesian A/B template]

DRIs (if applicable): [Who owns rollout, who owns dashboard, who owns client comms]

Roll-out timeline (if applicable):

Day 0–1: [e.g. Internal testing, catch technical breaks]
Day 2–5: [e.g. Controlled rollout to X% of users at 50/50 split]
Day 5+: [e.g. Full release, 50/50 split]
Day X: [e.g. Turn off experiment, analysis, decide ship or revert]

Links/templates: [e.g. Bayesian A/B template, technical bug tracker, dashboard]

➡️ Phase 2 / Post- [e.g. Scale · Handoff · Analysis]

[One line: what this phase is for.]

Duration: [e.g. 4–8 weeks]

Outcome: [e.g. Rollout complete, documentation, training, analysis readout]

Checklist:

[Step 1]
[Step 2]
Handoff: [What we leave behind — runbooks, training, support window]

Deliverables (standard)

Artifacts: [tracking plan, schemas, prompts, eval sets, reports]
Automations / services: [agents, workflows, integrations]
Dashboards / reports: [KPI board, adoption report, experiment readout]
Runbooks: [ops, rollback, admin console]
Handoffs: [training, office hours, video walk-throughs]

Success metrics

Execution KPIs: [latency, match rate, test coverage, incident MTTR]
Adoption KPIs: [weekly active users, task coverage, deflection rate]
Business KPIs (brief): [2–3 tied to outcomes — link to Offer for full outcome language]

Risks & mitigations

Risk	Mitigation
[e.g. Scope creep]	[e.g. Change order process, weekly scope check]
[e.g. Access delays]	[e.g. Kickoff checklist, escalation path]
[e.g. Experiment puts app in danger]	[e.g. Doomsday thresholds, revert plan]

Delivery notes (internal)

Squad & roles: [Who does what — e.g. GTM lead, data engineer, PM]
RACI / ceremonies: [Key decisions, who’s accountable, cadence]
Tooling: [Required tools, env, credentials]
Definition of done: [How we know the engagement is complete]
Compliance flags: [PII, SOC 2, HIPAA — if applicable]
Change log / version: [e.g. v1.0, date]

SOW copy block (pasteable)

Use this to paste into SOWs so scope and expectations stay consistent.

Scope: [Paste standard scope language]

Outcomes: [Paste standard outcomes]

Inclusions: [Paste what’s included]

Exclusions: [Paste what’s not included]

Client responsibilities: [Paste client obligations]

Acceptance criteria: [Paste how we define success / sign-off]

Brainforge Knowledge

Explorer

sop-template