Flagship program Dow Jones · Feb 2026 – present

Six teams. One front door. No new headcount.

If you read one program to decide whether I can do this job, read this one. A central brand function was absorbing work from six teams with no shared front door and no way to state what it was already carrying. I didn't ask for headcount. I built Relay, the AI intake and planning tool, then built the Asana capacity model that made the answer defensible. Plan in Relay. Execute in Asana.

The execution playbook Full JD → proof map

6 teams triaged ~8 events · 6 launches in 5 months 6 of 8 properties migrated ~80% faster builds

Decision log

What was actually decided, and why.

A DPM's real output is a trail of decisions someone else can audit. Here is the short version of this one.

Decision 01

Build the front door instead of asking for headcount

What changedRequests from six teams (subscription strategy, product design, product management, performance marketing, CRM, event marketing) now arrive through one structured intake instead of six inboxes and three hallway conversations.

Why that wayHeadcount takes a quarter to approve and a quarter to hire. A tool that makes a good request easier than a bad one changes behavior in a week, and it survives whoever is on vacation.

Decision 02

Keep a human review gate on AI triage

What changedRelay proposes a classification and a route. A person confirms it before anything becomes committed work in the queue.

Why that wayThis is a news company. An automated misroute is not a productivity story, it's a trust story. The gate is what lets the tool be aggressive everywhere else.

Decision 03

Make capacity a data question, not an opinion

What changedI own the brand marketing team's Asana implementation for reporting, capacity and resource management, supporting ~8 events and 6 launches over the past five months.

Why that wayWithout capacity data, "we're at capacity" is a feeling and gets overruled. With it, the tradeoff conversation happens once, in the open, with a number in it.

Decision 04

Sequence the migration against the event calendar

What changedThe design system migration reached 6 of 8 core properties and cut build time ~80% without pausing the launch schedule.

Why that wayFoundational work that competes with the calendar always loses. Foundational work that rides the calendar wins quietly: migrate the property whose next launch pays for the migration.

The execution playbook

How I run a program, in eight moves.

Not a methodology. A sequence I have actually run, with the program that proves each step. This is what I would bring to a design program at Meta on day one, adapted to whatever already works.

Move one

Intake: build one front door before improving anything

Find every path work currently takes into the team, including the ones that bypass the process. Then make the good path faster than the favor, so people choose it rather than being told to. Nothing downstream can be fixed while demand is still arriving invisibly.

Done before: Dow Jones · Relay, triaging requests from 6 teams

Move two

Roadmap under shifting requirements

Plan in phases with decision points rather than deliverable dates, and write down what each phase is allowed to change. A roadmap that can't absorb a research finding will be abandoned the first time one arrives, and then nobody has a plan at all.

Done before: Google Search Ads · a year-long research and design phase whose conclusions kept moving

Move three

Stakeholder map and decision log

Name who decides, who consults, and who only needs to know. Then keep a short written log of what was decided and why. The log is the artifact that stops a decision from being relitigated in month four by someone who wasn't in the room in month one.

Done before: Dow Jones · a central brand function serving 6 requesting teams across three mastheads

Move four

Risk register and flag cadence

Write the risks down before they're urgent, with likelihood, impact, a mitigation and an owner. Then make raising a flag the cheap, boring, expected behavior, because the expensive risks are the ones somebody noticed early and didn't want to be dramatic about.

Done before: The Row · platform testing with Monaverse ahead of a single launch date

Move five

Review gates, including responsible-AI and accessibility checks

Put reviewers in the kickoff, not in the last week. Sequence the roadmap around the gate. For AI, that means consent, provenance and accuracy are build requirements with a human review step. For accessibility, it means standards live in the design system so conformance is inherited rather than remembered.

Done before: GE Healthcare review gates · Relay's human review gate · NYU Langone design-system audit

Move six

Measurement: one qualitative and one quantitative measure, at kickoff

Both halves, defined before the work starts, reported on a schedule the team doesn't maintain by hand. And then actually used: the Dow Jones migration order was resequenced because the adoption data said which properties were blocking each other.

Done before: Asana reporting · 6 of 8 migrated, ~80% faster builds · GE executive visualizations

Move seven

Scale and delegate with written expectations

A program that only runs when the DPM is in the room hasn't been scaled, it's been supervised. Make the standard structural, hand over the ceremonies, and write the expectations down so delegation is a handoff rather than a hope.

Done before: Everyrealm · sourced and built the team, then handed them the methodology

Move eight

Vendor and budget operations

Own the vendor relationship personally through implementation and testing, not just at signature. Tie milestones to demonstrated integration rather than delivered files. Scope creep is quiet; the invoice is loud; the gap between them is the DPM's job.

Done before: Monaverse through platform testing to launch · Infosys on APS · agency-side at Moving Brands and Prophet

One caveat, stated up front. This playbook is what I have run in brand, product-marketing and enterprise-delivery contexts. I have not run it inside a trust-&-safety org, and I have not run it as a dedicated accessibility program. I would expect to adapt moves five and eight substantially with the people who already own those domains at Meta. I'd rather say that now than discover it in month two.

What I would deliberately not do in the first 90 days. Reorganize the team. Replace the tooling. Introduce a framework with a name. Those moves look like leadership from the outside, and they spend the one thing a new DPM can't buy back: the team's trust that I learned how the work actually flows before deciding how it should.

The rest of the bench

More programs

Open any card for Program details / Risk & mitigation

Scale & delegate

Dow Jones · design system

Eight event properties into one system

6 of 8 core properties migrated, build time cut ~80%, cross-property inconsistencies eliminated. Standards inherited rather than remembered.

StandardsAdoption

Content + localization

Google · embedded at Huge

android.com, tv.google, wearos on one calendar

Roadmap planning plus creative, content, deployment and localization across three properties, including the Android 12 launch, against immovable public dates.

Shifting requirementsContent ◐

Arizona Public Service transactional website and app design

Regulated · 28+ engineers

Arizona Public Service · Vertic

APS + Infosys: a system into production

A comprehensive design system after 3 months of research and 8 months of design, stewarded into production with Infosys and 28+ engineers, with skeptical executives inside the process.

Review cadenceExec buy-in

Inclusive design ◐

NYU Langone · via Moving Brands

UX/IA audit strengthening a design system

A structural audit of a health system's design system. This is the standards-and-inclusive-design work that is the honest half of the accessibility requirement, without claiming it was an accessibility program.

AuditStandardsHealth

The Row metaverse 3D architectural landmarks

Vendors & contracts

Everyrealm · The Row

The Row / Monaverse: vendor ownership to launch

Owned the Monaverse relationship through implementation and platform testing to a single coordinated launch, plus motion, 3D-rendering and promotional-video vendors.

Vendor mgmtPlatform testing

Scale from zero

Everyrealm · Director of Creative Operations

Creative ops from zero, then delegated

Sourced and built a team of 3D, UI/UX, motion and visual designers at a $50M+ (a16z) start-up, then defined the plans and methodology they ran across 12 metaverse platforms.

DelegationProcess design

The whole job description

Every responsibility. Every qualification. Every honest ◐.

Ten responsibilities, five minimum qualifications, seven preferred qualifications: all of them, in the JD's own order, each with the specific program behind it. Four rows are marked ◐ partial. Those four are the reason the other twenty-two are worth believing.

22JD lines met ✓

4named gaps ◔

26JD lines answered, in the JD’s own order

JD → proof Design Program Manager · Meta ✓ met · ◐ partial

Responsibilities

R1 · ResponsibilityIdentify and drive large, org-level product initiatives with multiple stakeholders, competing deadlines, dependencies and moving parts.✓ met

ProofDow Jones: a central brand function serving 6 requesting teams across WSJ.com, Barron's.com and MarketWatch.com, while migrating 8 event properties into one design system (6 of 8 done) and shipping ~8 events and 6 launches over the past five months. Competing deadlines are the default state, not the exception.

R2 · ResponsibilityPrioritize, assess tradeoffs, propose clear solutions and communicate strategically.✓ met

ProofGoogle web properties: continuously adjusted priorities, plans and resources across android.com (Android 12), tv.google and wearos.google.com against immovable public dates. Dow Jones: "migrate the next property" vs. "the event that ships Friday," decided on stated principles rather than volume.

R3 · ResponsibilityOwn and facilitate meetings that drive progress against goals, establish clear next steps and owners, and resolve blocks.✓ met

ProofAPS: structured the review cadence (workshops, deep dives, research readouts) so misalignment surfaced before build, across a 3-month research phase and an 8-month design phase. Everyrealm: introduced review ceremonies and production methodology into a team I had just hired.

R4 · ResponsibilityProactively assess execution risk, raise flags, and resolve issues before they become problems for larger teams.✓ met

ProofAPS: brought skeptical energy-utility executives into the process rather than around it, so objections landed at readout instead of at launch. The Row: platform testing with Monaverse ahead of a single coordinated launch. This is why the whole site is built on a risk register.

R5 · ResponsibilityBe the operations subject-matter expert to the product team; mentor and share expertise.✓ met

ProofDow Jones: built Relay so the messy requests six teams send me become one clean, prioritized plan without anyone having to be taught how to ask. Everyrealm: sourced, hired and trained the creative team, then gave them the process rather than the answers.

R6 · ResponsibilityCommunicate persuasively to the right audience to get buy-in and feedback.✓ met

ProofGoogle: sustained working relationships with product marketing managers, strategists and designers; a director-level testimonial is on file. GE Healthcare: built executive-level visualizations that kept a global campaign's leadership aligned while volume flowed underneath.

R7 · ResponsibilityProactively problem solve and distribute information effectively.✓ met

ProofDow Jones: I own the brand marketing team's Asana implementation for reporting, capacity and resource management. Information distribution built as infrastructure, so nobody depends on a weekly email arriving.

R8 · ResponsibilityDefine and measure program impact qualitatively and quantitatively, iterating to ensure usage and adoption.✓ met

ProofQuantitative: 6 of 8 properties migrated, build time cut ~80%, ~8 events and 6 launches over the past five months, requests from 6 teams through one front door. Qualitative: cross-property inconsistencies eliminated; requesting teams stopped routing around the queue. Both halves for six programs are published in How I measure impact.

R9 · ResponsibilityScale programs and delegate with clear expectations.✓ met

ProofEveryrealm: built the creative-ops function from zero at a $50M+ (a16z) start-up, sourcing a team of 3D, UI/UX, motion and visual designers, then defining the plans and methodology they ran without me. Dow Jones: the design system exists precisely so the next property doesn't need me.

R10 · ResponsibilitySource and manage operational budgets, vendor relationships and contracts for programs like Accessibility.◐ partial

ProofVendors and contracts: proven. The Row / Monaverse owned through implementation and platform testing to launch, plus external motion, 3D-rendering and promotional-video vendors; the Infosys partnership on APS with 28+ engineers; team-building at Everyrealm; years agency-side at Moving Brands and Prophet. Not proven: I have never owned a dedicated Accessibility program's budget, vendor roster or conformance targets. Half of this row is real and half would be new.

Minimum qualifications

M1 · Minimum7+ years of program management / project management experience.✓ met

Proof13+ years of end-to-end program and project management, from agencies and start-ups through Fortune 500, B2C and B2B, plus a design-program-shaped role right now at Dow Jones. Nearly double the minimum.

M2 · MinimumManaging cross-functional projects with shifting requirements, from concept and roadmapping through execution and delivery.✓ met

ProofGoogle Search Ads: a year-long UI/UX research and design phase exploring an unbuilt future, leading 4–6 designers with Google's Search UX Design Director and his team: requirements shifted by design, not by accident. APS: concept → research → design → production with Infosys and 28+ engineers.

M3 · MinimumWorking directly with product designers and content designers.◐ partial

ProofProduct design: met. I partner daily with commerce & acquisition product design teams at Dow Jones; I led 4–6 designers on Google Search Ads with Google's Search UX Design Director. Content: partial. I oversaw creative, content and localization for android.com, tv.google and wearos.google.com, and run brand copy across 10+ Dow Jones event properties. But that is content as an owned workstream, not a partnership with a dedicated content-design org with its own craft standards and backlog. See the first-90-days plan.

M4 · MinimumIndependently managing timelines, setting priorities and escalating blockers across multiple workstreams.✓ met

ProofGoogle: one launch calendar across three properties with independent release dates. Dow Jones: 10+ event sites, a migrating design system and a six-team intake queue, concurrently. Everyrealm: 12 metaverse platforms plus web2 properties and brand work, with no inherited process.

M5 · MinimumUsing data to identify problems and communicate solutions.✓ met

ProofDow Jones: the Asana implementation exists specifically to produce capacity, resourcing and progress data, and the ~80% build-time reduction is the measured outcome of a system decision. Relay converts unstructured requests into structured, countable demand. GE Healthcare: executive-level visualizations plus supporting data-analysis delivery on a global campaign.

Preferred qualifications

P1 · PreferredCompliance, trust & safety, or responsible AI product development.◐ partial

ProofRegulated delivery: real. GE Healthcare (global health campaign), Merck and Novartis (pharma), the Endo–Mallinckrodt merger work at Prophet, NYU Langone (health), APS (a regulated utility). Review gates were never optional. Responsible AI, real: Relay (in production at Dow Jones, a human review gate on AI triage) and an agreements system I built, which turns what people commit to in a working session into a reviewed, signed record. Trust & safety: not claimed. I have never written a content policy, run enforcement, or sat in a T&S review.

P2 · PreferredDemonstrated ongoing AI skill development: prompt/context engineering, agent orchestration.✓ met

ProofRelay in production at Dow Jones, plus an agreements system and a resourcing tool built solo, directing AI. Self-taught in agent orchestration, prompt and context engineering, and the production foundations of scalable AI products. These are operations systems built for problems I ran into in the workplace, not ventures and not products I am selling: several tools integrated, a custom database underneath, permissions designed in, and infrastructure I set up and keep running. Finished systems in real use, not demos. Detailed on the AI fluency section.

P3 · PreferredResponsible/ethical AI practices: risk assessment, bias mitigation, quality and accuracy reviews.◐ partial

ProofRelay, in production at Dow Jones, keeps a human review gate on AI output before triage becomes work. An agreements system I built is the consent-and-provenance artifact: it turns what people commit to in a working session into a reviewed, signed record, so a scope or a hand-off cannot quietly evaporate. Consent and provenance were design decisions, not additions. A resourcing tool I built for my own team stages every decision for a one-click yes, so nothing changes unless a human confirms it. Not claimed: I have not run a formal bias-audit program.

P4 · PreferredAccessibility programs or inclusive design practices.◐ partial

ProofInclusive design and standards: real. NYU Langone: a UX/IA audit strengthening their design system via Moving Brands. APS: a comprehensive design system across public, account and transactional experiences for a regulated utility, where standards conformance was not optional. Program ownership: new. Owning a dedicated accessibility program (its budget, its vendors, its conformance targets, its remediation backlog) would be new to me. The vendor and budget half of this requirement is proven (see R10, P7); the program half is not.

P5 · PreferredIntegrating AI tools to redesign workflows and drive measurable impact: efficiency gains, quality improvements.✓ met

ProofDow Jones: built Relay, receiving and triaging requests from 6 teams. This is the JD bullet, in production, at a company with real editorial and legal review. Paired with the Asana implementation (planning and prioritization happen in Relay; execution, reporting, capacity and resourcing live in Asana), the redesign is measurable in capacity and cycle time rather than in enthusiasm. It is the flagship program at the top of this page.

P6 · PreferredLeading product or business strategy in a time-sensitive environment.✓ met

ProofGoogle web properties: reprioritized against immovable public launch dates including Android 12. Prophet: an MVP website redesign for a pharma client during the Endo–Mallinckrodt merger, which is about as time-sensitive as strategy work gets. Beyond: Google's "Google for Work" website under a tight deadline with a team across time zones.

P7 · PreferredManaging vendors and contractors.✓ met

ProofThe Row / Monaverse: owned the vendor relationship through implementation and platform testing to a single coordinated launch, plus external vendors across motion, 3D rendering and promotional video. APS: the Infosys partnership and 28+ engineers. Everyrealm: sourced and built a team of contractors and specialists from zero. Plus agency-side at Moving Brands and Prophet, which is the fastest way to learn how vendors are really scoped and billed.

Four partials, on purpose. R10 (accessibility program budget), M3 (content designers), P1 (trust & safety) and P4 (accessibility programs). Each is marked in the same table as the strengths, in the same typeface, without a softening adverb. A candidate who marks nothing partial is telling you they haven't read the JD carefully, or that they have, and decided to hope.

Send me how work reaches your design org today and I’ll tell you where the leaks are.

Back to the risk register Want this register with your programs in it?