paulasilvatech ASP Agentic SDLC Personas v1.0.0
EN PT ES
02 product · Planning

Business Manager

Translates business into KPIs.

Updated: 2026-04-24 14 sections Download .zip

The Business Manager is the persona that binds strategy to measurement. In an AI-native SDLC, the Business Manager operates a stack of validated primitives that turn business intent into machine-readable KPIs and outcome hypotheses.

Executive summary

The Business Manager translates enterprise strategy and OKRs into product-level KPIs, value stories, and investment cases that the rest of the SDLC can optimize against. In an AI-native SDLC the Business Manager operates inside the Planning phase with a fixed set of primitives: one KPI translation agent, four slash prompts, scoped instructions, schema-validated hooks, and a curated list of validated MCPs. Primary outputs are OKR trees, KPI specifications, business cases, and value stories linked to spec anchors.

Role and responsibilities

Think of the Business Manager like the conductor of an orchestra reading the score the composer wrote. The conductor does not play any instrument, but they are accountable for the fact that every section plays in time and on key, and that what the audience hears matches the composer’s intent. In an AI-native SDLC the OKR tree is the score, the KPI specs are the tempo markings, and the Business Manager is responsible for the performance across product, architecture, and build.

Primary responsibilities:

  • Author and maintain the OKR tree in docs/okrs/ with measurable, time-bounded key results
  • Translate every strategic objective into a product-level KPI with baseline, target, and measurement method
  • Publish the value story for every initiative, linked to the SPECIFICATION.md owned by the Product Owner
  • Maintain the business case with cost, benefit, and risk for every feature above a defined threshold
  • Operate the KPI Translator agent and the /okrs, /kpi-map, /biz-case, /value-story prompts
  • Govern the portfolio review in GitHub Projects or Azure Boards
  • Close the loop with Application Insights and Azure Monitor telemetry to verify KPI achievement

Jobs to be done

  1. As a Business Manager, I want to convert a strategic objective into an OKR tree within a day, so that the portfolio is aligned at the start of every quarter.
  2. As a Business Manager, I want every KPI to have a baseline, target, and measurement method, so that outcomes are auditable, not anecdotal.
  3. As a Business Manager, I want the value story to be linked to the spec, so that engineering decisions trace back to business intent.
  4. As a Business Manager, I want business cases generated from templated inputs, so that the cycle time from idea to funding is measured in days, not weeks.
  5. As a Business Manager, I want live telemetry on KPI achievement from Application Insights, so that I can intervene before the quarter ends.
  6. As a Business Manager, I want a monthly portfolio health report generated automatically, so that leadership decisions are grounded in current data.

Pain points before AI-native

  1. OKRs written as slides, not data. Slide decks cannot be queried. Mid-quarter drift is invisible until the review meeting.
  2. KPIs without baselines. A target without a baseline is a wish. Teams optimize what is easy to measure, not what matters.
  3. Value stories disconnected from specs. Leadership hears one narrative, engineering delivers another, and the gap only surfaces at launch.
  4. Business cases authored in spreadsheets. Cost, benefit, and risk live in static files with no link to execution artifacts.
  5. Telemetry ignored after launch. Feature success is declared on ship date. Actual impact is never measured against the original target.

AI-native daily workflow

The Business Manager operates a fixed loop each day. The loop uses GitHub Copilot primitives inside Visual Studio Code and Claude Code at the terminal, plus a small catalog of validated MCPs for external context.

Morning setup

  1. Open the portfolio repository in Visual Studio Code. GitHub Copilot Chat loads AGENTS.md and the scoped OKR instructions.
  2. Pull the latest KPI telemetry from Application Insights via the Azure MCP Server and refresh the KPI dashboards.
  3. Review overnight stakeholder input captured in Teams and Outlook through the Microsoft 365 Agents SDK MCP.
  4. Run /kpi-map to confirm every active initiative is mapped to at least one KR.

Midday execution

  1. OKR authoring. Invoke /okrs on the quarter’s strategic theme. The KPI Translator agent produces an OKR tree with numbered objectives and measurable key results, and refuses to emit a KR without a baseline and target.
  2. KPI translation. Invoke /kpi-map to bind each KR to a product-level KPI. The agent checks that the measurement method references a concrete data source, typically Application Insights, Azure Monitor, or GitHub Projects.
  3. Business case. Invoke /biz-case for any initiative above the funding threshold. The agent fills cost, benefit, risk, and assumption sections against the template.
  4. Value story. Invoke /value-story to bind the business case to the SPECIFICATION.md owned by the Product Owner. The agent produces a one-page narrative with measurable outcomes.

Afternoon review

  1. Run a portfolio health sweep across all active OKRs. Flag any KR without fresh telemetry, any KPI below trajectory, and any initiative without a linked spec.
  2. Open a pull request on the OKR tree. GitHub Copilot Code Review comments on measurability; leadership reviewers approve content.
  3. Publish the daily portfolio digest to the executive Teams channel via the Microsoft 365 Agents SDK, summarizing progress, risks, and decisions required.
  4. Sync the backlog in GitHub Projects or Azure Boards so every item carries the OKR and KPI tags.

Agent

AgentFilePurpose
kpi-translator.github/agents/kpi-translator.agent.mdTranslate objectives into KPIs, author OKR trees, generate business cases and value stories

The KPI Translator uses claude-sonnet-4-6 by default. Tools: read, edit, search, grep, glob. No bash access. Extended thinking is enabled for /biz-case only, where benefit and risk modeling benefits from deeper reasoning.

Slash prompts

CommandFilePurpose
/okrs.github/prompts/okrs.prompt.mdAuthor or revise the OKR tree for a strategic theme
/kpi-map.github/prompts/kpi-map.prompt.mdBind each KR to a product-level KPI with baseline, target, and source
/biz-case.github/prompts/biz-case.prompt.mdGenerate a structured business case with cost, benefit, risk
/value-story.github/prompts/value-story.prompt.mdProduce a one-page narrative linking business case to spec

Instructions scoped

Scoped applyTo reduces token cost by approximately 68 percent compared to global instructions.

Scope (applyTo)FilePurpose
docs/okrs/**/*.md.github/instructions/okrs.instructions.mdOKR tree format, measurability rules, banned vague verbs
docs/kpis/**/*.md.github/instructions/kpis.instructions.mdKPI spec schema: baseline, target, source, cadence
docs/biz/**/*.md.github/instructions/biz-case.instructions.mdBusiness case template and evidence requirements

Hooks

Hooks cost zero LLM tokens. They are the strongest governance layer for business artifacts.

  • pre-commit: reject any OKR without a measurable KR and any KPI without a baseline or source
  • post-commit: refresh the portfolio dashboard from the latest OKR and KPI files
  • pre-merge: block merge of any business case that lacks a linked spec anchor

Validated MCPs

MCPPurposeOwner
Azure MCP ServerQuery Application Insights and Azure Monitor for live KPI telemetryMicrosoft (official)
GitHub MCP ServerRead and update GitHub Projects for portfolio governance and OKR-tagged backlogGitHub (official)
Azure DevOps MCP ServerSync OKRs and KPIs with Azure Boards when the team uses Azure DevOpsMicrosoft (official)
Microsoft 365 Agents SDK MCPPublish digests to Teams and ingest decisions from OutlookMicrosoft (official)
Microsoft Learn Docs MCPGround business cases in current Microsoft and Azure pricing and capability documentationMicrosoft (official)

Real examples

Example 1: quarterly OKR authoring

Input: A strategic theme from the annual plan, “Increase self-service contract renewal rate from 22 percent to 55 percent by Q4.”

Invocation: /okrs with the theme and last year’s telemetry pulled from the Azure MCP.

Expected output:

  1. A docs/okrs/2026-q3.md with one objective and three KRs, each carrying a baseline, target, and Application Insights query as the measurement source.
  2. Three issues in GitHub Projects tagged with the OKR and assigned to the owning squads.
  3. A digest posted to the leadership Teams channel via the Microsoft 365 Agents SDK.

Example 2: business case for a funding gate

Input: A proposal for a new partner API integration above the funding threshold.

Invocation: /biz-case followed by /value-story.

Expected output:

  1. A docs/biz/partner-api-2026.md with cost, benefit, risk, and assumption sections filled against the template.
  2. A one-page value story in docs/biz/partner-api-2026-value.md that links to the SPECIFICATION.md anchor and names the KPI the initiative will move.
  3. A pull request that triggers the funding gate review, with Copilot Code Review commenting on measurability and scope.

Anti-patterns

  1. KRs without baselines. A target of “increase adoption” is unmeasurable. Mitigation: the pre-commit hook rejects any KR without a baseline and source.
  2. Slide-first OKRs. If the authoritative copy lives in a deck, it cannot be queried or diffed. Mitigation: the OKR tree is authored in markdown under version control.
  3. Value stories disconnected from specs. When the narrative does not link to a spec anchor, engineering optimizes something else. Mitigation: /value-story refuses to close without a spec link.
  4. Telemetry ignored post-launch. Declaring success on ship date is a habit, not a measurement. Mitigation: the portfolio dashboard flags any KR with stale telemetry.
  5. Business cases in spreadsheets. Static files drift from reality. Mitigation: business cases live in markdown with hooks that validate template completeness.

KPIs and impact metrics

KPIBaselineTargetMeasurement
OKR coverage, initiatives mapped to a KR55 percent100 percentPortfolio dashboard query
KPIs with baseline and measurement source40 percent100 percentKPI spec linter in GitHub Actions
Time from strategy update to OKR tree3 weeks< 3 daysGitHub PR timestamps
Funding cycle time, proposal to gate6 weeks< 2 weeksBusiness case PR timestamps
KR on-track rate at mid-quarterUnknown> 75 percentAzure MCP telemetry query
Post-launch outcome verified against target20 percent100 percentApplication Insights vs target

Maturity in four levels

LevelNameMarkers
L1ManualOKRs in slides, KPIs in spreadsheets, value stories verbal
L2AssistedCopilot used to polish OKR prose, still no machine-readable artifacts
L3AugmentedKPI Translator agent, four slash prompts, scoped instructions, Azure MCP for telemetry, KPIs with baselines
L4AutonomousFull primitives kit, hooks enforced, portfolio dashboard live, leadership digest automated, post-launch verification standard

Integration with other personas

  • From Enterprise Architect: capability scan and constitution principles that constrain the OKR tree
  • To Product Owner: approved OKRs and KPIs that the spec must hook into via /link-acceptance
  • To Requirements Engineer: KPI-bound outcomes that inform gap scans and traceability
  • To Engineering Manager: OKR tree that drives capacity planning and squad assignment
  • To SRE: SLOs aligned with customer-facing KPIs for reliability investment
  • To Release Manager: value story as the backbone of release narrative and stakeholder comms

Glossary

  • OKR: Objectives and Key Results. A quarterly goal-setting framework with measurable outcomes.
  • KPI: Key Performance Indicator. A product-level metric with baseline, target, source, and cadence.
  • KR: Key Result. A measurable component of an objective, bound to one or more KPIs.
  • Value story: a one-page narrative linking a business case to a spec anchor and a KPI.
  • Business case: structured document with cost, benefit, risk, and assumption sections used at funding gates.
  • Portfolio dashboard: generated view aggregating all OKRs, KPIs, and initiative statuses.

References