# M365Marketing — Completed Exemplar BRD

> **What this is**: A fully authored BRD for the M365Marketing campaign-allocation dashboard. Every section is filled in to the level a real Director of Product Marketing would sign off on. It is the *finished state* the [M365Marketing starter](m365marketing-starter.md) reaches when the discipline has been applied end to end.
>
> **How to use it**: Read alongside the other exemplars ([AIRS](airs-exemplar.md) for research, [CloudRevenue](cloudrevenue-completed.md) for sales operations) to see the same eight-section shape survive across three domains. When you complete the [M365Marketing starter](m365marketing-starter.md) yourself, compare your Sections 4–8 against this exemplar — but not field for field. Check whether your version passes the three validity checks from Chapter 2 (single-sentence, disagreement, reversal). If it does, your BRD is defensible even if it differs from this one.

---

## Business Requirements Document

### M365Marketing Campaign Allocation Dashboard

- **Project**: Marketing-spend allocation dashboard supporting weekly campaign decisions across the M365 product-marketing portfolio
- **Stakeholder**: Director of Product Marketing, M365
- **Date**: 2026-06-10 (draft) / 2026-06-24 (sign-off)
- **Priority**: High

### 1. Executive Summary

The Director of Product Marketing for M365 manages a marketing budget across twelve concurrent campaigns spanning paid search, paid social, display, content syndication, and field-event channels. The decision the dashboard supports is the weekly Monday-morning reallocation: which campaigns get a budget increase this week, which get a budget cut, and which stay flat. The dashboard replaces a Friday-afternoon analyst-built spreadsheet that has consistently arrived after the Monday meeting starts; the timing mismatch has led to reallocation decisions based on stale data three of the last six weeks. The dashboard's primary job is to be reliably available by Monday 8 AM; its secondary job is to reclaim the marketing analyst's one-day-per-week reconciliation cost for higher-value work on the attribution model itself.

### 2. Business Context

The marketing team uses a combination of the platform-native reporting from each ad-spend platform (Microsoft Advertising, LinkedIn Campaign Manager, Meta Ads Manager, etc.), the CRM lead-to-opportunity reporting, and the marketing-attribution model the team runs in a separate analytics warehouse. The Friday spreadsheet pulls from all three sources and reconciles them by hand. The pull-and-reconcile work takes the marketing analyst the equivalent of one working day each week, which is the bottleneck behind the timing mismatch. The Director wants the spreadsheet's content available by Monday 8 AM in a self-refreshing dashboard, with the analyst's hours reallocated to attribution-model improvements and campaign post-mortems.

The dashboard does not replace the Monday meeting or the campaign-lead conversations that follow it. Campaign leads continue to own their campaigns; the dashboard's job is to give the Director and the twelve campaign leads a single agreed set of numbers to argue *from*, not a set of numbers to argue *about*.

### 3. Key Business Questions

***Channel performance***

1. Which paid-channel campaigns are pacing above or below their planned weekly cost-per-qualified-lead and by how much?
2. Which content-syndication campaigns are driving the highest velocity from initial download to qualified pipeline?
3. Which field-event campaigns have produced the most qualified pipeline relative to their per-attendee spend, trailing six events?

***Segment targeting***

1. Which customer segments (industry, company size, geography) are driving the highest qualified-lead volume per dollar across each channel?
2. Are any segments oversaturated — receiving impressions or contacts at a frequency that historical analysis shows depresses response?
3. Which segments are underweighted relative to their share of the total addressable market the M365 sales team is targeting this quarter?

***Spend allocation***

1. If the Monday meeting decides to reallocate 15% of the weekly budget, which campaign-to-campaign moves produce the largest expected lift in qualified pipeline?
2. Which campaigns have hit their planned budget cap for the quarter and need either a cap increase or a planned ramp-down?

### 4. Success Criteria

| Criterion | Measurement |
| --- | --- |
| **Monday-meeting availability** | The dashboard is fully refreshed and green-checked by 8:00 AM Pacific every Monday. Confirmed by an automated pipeline-health check that emails the Director's chief of staff if any of the four upstream refreshes has failed or is more than 60 minutes late. Baseline: 100% on-time across the first eight consecutive Mondays post-deployment. |
| **Cross-platform reconciliation consensus** | Attribution-window differences across Microsoft Advertising (7-day click), LinkedIn (30-day view), Meta (7-day click / 1-day view), and the attribution warehouse (custom multi-touch) are reconciled with a single documented rule per metric. Rules signed off by the marketing analyst, the attribution-model owner, and the two channel leads whose campaigns are most affected (paid search + LinkedIn). Sign-off logged as an appendix to this BRD. |
| **Analyst-hours reclaimed** | The marketing analyst logs less than 2 hours per week on reconciliation work (down from ~8 hours/week under the Friday-spreadsheet baseline). Measured over the first four post-deployment weeks. The reclaimed hours are visibly reassigned to attribution-model improvements or campaign post-mortems, tracked in the analyst's weekly summary. |
| **Reallocation decision speed** | The Monday meeting completes its reallocation decisions in under 45 minutes (down from ~75 minutes under the current state, where the first 30 minutes was consumed by number reconciliation). Confirmed by meeting-length tracking over four consecutive Mondays. |
| **Anomaly surfacing** | Platform-outage weeks (where a channel returns zero impressions and zero spend due to an ad-platform incident) are surfaced with an in-dashboard callout rather than being averaged into the trailing metrics. Tested against the three known platform-outage weeks in the historical data. |

### 5. Stakeholder Requirements

#### Primary Users

- **Director of Product Marketing (M365)**: Weekly use every Monday morning to drive the reallocation meeting; occasional use throughout the week to sanity-check campaign-lead requests for mid-week budget shifts. Requires a *decision-mode view* (one metric per panel, high contrast, minimal chrome) optimized for a 12-person conference-room screen; also requires a *phone-friendly view* for pre-meeting Sunday-evening or Monday-morning skims.
- **Marketing Analyst (Friday spreadsheet's current owner)**: Daily use during business hours. Primary operator for exception handling — when the automated pipeline surfaces a metric that looks wrong, this analyst is the person who investigates and either fixes the source-system extract or files an errata note against the reconciliation rule. Requires drill-through to the raw ad-platform reporting on any metric.
- **Twelve Campaign Leads (one per campaign)**: Weekly use ahead of the Monday meeting; on-demand use throughout the week. Each campaign lead primarily views their own campaign but must be able to see the relative performance of the other eleven for reallocation-conversation context. Requires filter persistence per user (a campaign lead should not need to re-filter every visit).

#### Secondary Users

- **Attribution-Model Owner (in the analytics warehouse team)**: Occasional use, primarily when reconciliation rules need updating (a new channel joins, an existing channel changes its attribution window). Requires drill-through to the attribution warehouse for source-of-truth checks.
- **VP Marketing Operations (Director's manager)**: Weekly skim; deeper use during quarterly business reviews. References the dashboard's trailing-quarter roll-ups.
- **Sales Segment Owners (three roles corresponding to the top three industry segments)**: Occasional use, primarily to reconcile marketing-generated pipeline against their own sales-pipeline reports. Requires read-only access to the segment-targeting surfaces.

### 6. Data Requirements

| Requirement | Detail |
| --- | --- |
| **Data Sources** | (1) Microsoft Advertising Reporting API (spend + impressions + clicks + conversions, hourly); (2) LinkedIn Campaign Manager API (same metrics, hourly); (3) Meta Ads Manager API (same metrics, hourly); (4) Content-syndication vendor extract (weekly); (5) Field-events CRM (per-event, updated within 48 hours of event close); (6) M365 CRM lead-to-opportunity data (nightly); (7) Marketing-attribution warehouse view (nightly, produced by the attribution-model team). |
| **Refresh Frequency** | All API sources refresh into the semantic model at 02:00 Pacific every day (Sunday–Friday). Attribution-warehouse and CRM refresh at 03:00 Pacific. The dashboard is fully consistent by 04:00 Pacific, giving four hours of buffer before the 8:00 AM Monday availability commitment. The content-syndication weekly extract lands Friday afternoon; a late arrival triggers a documented fallback (use last week's data with a visible "content-syndication data one week stale" callout). |
| **Historical Horizon** | 24 months. Twelve months is the minimum for seasonal-pattern detection; twenty-four months lets the year-over-year comparisons the Director cares about during quarterly reviews. The trailing six field events (a rolling window, not calendar-quarter-bound) must always be available for the field-event panel. |
| **Data Sensitivity** | Internal-confidential. No customer PII (the analytics-warehouse layer strips it before extraction). Campaign spend numbers are considered competitively sensitive and are excluded from any export or screen-share to non-M365 audiences by default; the dashboard requires an explicit "share-safe" toggle before exports to external audiences. |
| **Known Quality Issues** | (a) Attribution-window differences across the four paid channels (see Success Criterion 2) — must be reconciled per documented rule per metric before deployment. (b) The `note` column in the performance data flags three types of week: `paused` (campaign intentionally off, zero-values expected), `platform-outage` (ad platform incident, zero-values suspect), and empty (normal week). The dashboard must display each type distinctly and must exclude platform-outage weeks from trailing averages. (c) Historical Meta data before 2025-04 uses the legacy attribution model and is not directly comparable to post-2025-04 data — flag the discontinuity with a visible marker on any panel that crosses the boundary. |

### 7. Deliverables

| Item | Detail |
| --- | --- |
| **Format** | Power BI workspace with three reports: (1) *Monday Reallocation* (the executive view; ten panels max, tuned for the Director's 45-minute meeting); (2) *Campaign Deep-Dive* (per-campaign drill, one page per active campaign, owned by the individual campaign lead); (3) *Segment & Attribution* (segment-targeting and cross-channel attribution surfaces, primarily used mid-week rather than at the Monday meeting). |
| **Access Method** | Power BI Service, deployed to the M365 Marketing workspace. Director, marketing analyst, and attribution-model owner access via direct workspace membership. Twelve campaign leads access via a published app; the app hides the source datasets and exposes only the three reports. Sales segment owners access via a separate published app with read-only permissions on the segment-targeting surfaces only. Mobile access is required for the Director's phone-friendly view. |
| **Mobile Required** | Yes, for the Director's *Monday Reallocation* view (phone-friendly layout of the top four panels). Not required for the other two reports. |
| **Target Go-Live** | 2026-08-03 (a Monday). The two Mondays prior (2026-07-20 and 2026-07-27) will run the dashboard in trial alongside the existing Friday spreadsheet; the Director signs off on cutover only after both trial runs pass all five success criteria. |
| **Review Cadence** | Weekly cadence review during the first month post-deployment (assess Monday-morning availability and reallocation-decision speed against baseline). Quarterly thereafter, coinciding with the marketing operations quarterly business review. |

### 8. Constraints & Assumptions

- **Constraint**: Monday 8:00 AM Pacific availability is a hard deadline. Any pipeline failure detected after 07:00 must trigger a documented rollback to the last-known-good snapshot rather than a delayed refresh. The Director will not accept "just wait ten minutes" as a workflow.
- **Constraint**: The four attribution-window reconciliation rules must be signed off by the marketing analyst, the attribution-model owner, and the two most-affected channel leads before deployment. If any of the four signatures is missing, the dashboard cannot ship the disputed metric — it must show either the reconciled version or omit the metric entirely with a visible note. "Ship the analyst's rule and hope no one notices" recreates the exact problem this dashboard is meant to eliminate.
- **Constraint**: Ad-platform API rate limits cap the freshness of the paid-channel data at hourly granularity. Any sub-hourly analysis is out of scope; requests for near-real-time spend adjustment must go to the platform-native tools, not the dashboard.
- **Constraint**: The published-app access model for the twelve campaign leads must not expose competitor-relevant spend data across campaigns. Campaign leads see their own campaign's spend numbers and other campaigns' *relative* performance (indexed to a shared baseline), not other campaigns' absolute spend.
- **Assumption**: The attribution-model warehouse view will remain available and stable throughout the dashboard's planned 24-month lifetime. The attribution-model owner has committed to a 30-day change notice for any breaking change; the dashboard's fallback plan for a broken attribution feed is to fall back to platform-native conversions with a visible "attribution-model unavailable" callout.
- **Assumption**: The marketing analyst will continue in role through at least Q1 2027. If turnover occurs, the exception-handling responsibilities transfer with the role; the analyst's runbook (kept in the workspace documentation section) captures the handoff-critical knowledge.
- **Assumption**: The 12-campaign portfolio size is stable. Campaign turnover happens naturally (ended campaigns replaced by new campaigns) but the total active-count stays near 12. If the count grows to 16 or more, the *Monday Reallocation* panel design needs revisiting; that is a scope change, not a maintenance update.

### Approval

| Role | Name | Date | Signature |
| --- | --- | --- | --- |
| Stakeholder (Director of Product Marketing, M365) | | 2026-06-24 | |
| Data Owner (Marketing Analytics Warehouse) | | 2026-06-24 | |
| Marketing Analyst (day-to-day operator) | | 2026-06-24 | |
| Attribution-Model Owner | | 2026-06-24 | |
| Analyst (author) | | 2026-06-24 | |

Channel-lead sign-off on attribution-window reconciliation rules (Success Criterion 2) is logged as an appendix separate from this BRD; sign-off date for the reconciliation round is 2026-07-20.

---

**Pattern notes.** Compare this BRD's Section 4 Success Criteria against the [CloudRevenue completed exemplar](cloudrevenue-completed.md)'s Section 4. Both name a specific hard deadline (Monday-meeting availability for M365Marketing; the September offsite for CloudRevenue) as the load-bearing criterion. Both also include a cross-stakeholder consensus criterion (attribution-window reconciliation here; metric-definition consensus there). The pattern — one deadline-driven criterion plus one consensus-driven criterion plus three operational criteria — is not accidental; it is what a Section 4 that actually gates deployment looks like across domains.