A worked example

What a finished packet looks like.

Jordan Lee is invented, and so are the company and every number. What's real is the shape: a rubric rating with evidence, impact written as numbers someone can check, and a one-page case your manager can read in 90 seconds. No signup needed to read it.

Illustrative example · Jordan and every number are invented

The work behind it

How Jordan got to that page.

Jordan has 7 years of experience at a B2B payments company. This is the packet at the end of Step 11. Four of five rubric dimensions came back with real evidence, so the Runway Plan branch is skipped. Notice how each number says where it came from (finance, analytics, support). That's what makes a claim hold up when someone asks.

Step 2

Rate yourself against the bar

Dimension	Rating	Evidence
Scope	Staff	Checkout redesign spanned billing, payments, and onboarding. None of the three teams owned the end-to-end flow, which carries about $120M a year in invoice payments.
Ambiguity	Staff	Named the "invoice abandonment" problem from support tickets before it was on any roadmap.
Influence	Staff	Got three team leads to agree on one flow before design started. No reporting line to any of them.
Systems / Leverage	Staff	Design system v2 adopted by three product teams. New screens on-system went from 60% to 91%, saving about 28 engineer-weeks in a year.
Craft depth	Thin	Raises my own bar, but I haven't set one other people are measured against. Started a weekly critique that two other teams joined. Too early to call it a standard.

Step 3

Write the thesis

I operate at Staff because I find and frame problems across team boundaries before anyone assigns them, and I get other teams to adopt what I build, shown by a checkout redesign worth about $370K a year in revenue and a self-serve launch that brought in $1.1M in new ARR in two quarters.

Step 4

Build the impact inventory

Checkout and invoicing redesign

Context: Customers were abandoning invoice payment. Three teams each owned a piece of the flow and nobody owned the whole thing.
Action: Mapped the full journey, ran 12 customer sessions, and wrote the case for treating it as one problem. Got the three team leads to agree on a single flow before any design started.
Impact: Payment completion rose from 61% to 74% over two quarters. On roughly $120M a year in attempted invoice volume at a 2.4% take rate, finance estimates about $370K in added annual net revenue. Failed-payment tickets fell 35% (about 490 fewer a month), roughly $50K a year in support cost. Average days to payment dropped from 21 to 14.
Level-Proof: Defined a cross-team problem nobody assigned to me, then got three teams to build against one answer.

Design system v2

Context: Three product teams ran separate component libraries. About 40% of new screens used off-system components, and each team was rebuilding the same patterns.
Action: Proposed shared tokens and a contribution model. Built the first 12 components with two engineers. Ran weekly office hours for six months.
Impact: On-system screens went from 60% to 91% in three quarters. Engineering leads estimate about a sprint saved per new feature, roughly 28 engineer-weeks over the year, or about $170K in loaded cost. Defects filed against shared UI fell 52%.
Level-Proof: Three teams adopted a standard I created, with no authority over any of them.

Self-serve onboarding

Context: The company was sales-led only. A self-serve tier had to launch in six months, with no research behind it.
Action: Led discovery, partnered with the PM and two engineers, and recommended cutting v1 scope by a third to hit the date. Set up the first experiment framework for the team.
Impact: Launched in five months. Signup-to-activation conversion rose from 22% to 41%, and time to first value fell from 9 days to 2. Self-serve brought in about $1.1M in new ARR in its first two quarters, 18% of new accounts, at roughly a third of the acquisition cost of sales-led deals.
Level-Proof: Shaped the plan beyond the design brief, including the call on what not to build.

Step 5

Establish scope and ownership

My work in the last year touched three product teams (billing, payments, onboarding) and the shared design system used by all of them. The checkout flow alone carries about $120M a year in invoice payments. None of those teams reported to my manager. I worked across boundaries nobody assigned me to cross.

Step 6

Surface influence without authority

In the checkout project, billing wanted to fix their step in isolation and payments wanted to wait for a platform migration. I brought both teams a side-by-side of where customers dropped off, showing that neither fix alone moved the number. Both leads agreed to one shared flow and a joint release. That plan didn't exist until I proposed it, and the joint release is what produced the 13-point lift in completion.

Step 7

Document craft and systems contribution

Design system v2 raised the floor on consistency for three teams, and engineering now builds from it by default. That's where the 28 engineer-weeks came from. On craft depth, I started a weekly critique that two other teams joined. I'm not claiming a quality bar yet. I'm documenting one this quarter.

Step 8

Capture mentorship

Dani, a mid-level designer, couldn't run stakeholder reviews on her own 18 months ago. I paired with her on three reviews, then stepped back and gave feedback afterward. She now runs reviews independently, owns the billing flows, and was promoted to Senior this cycle. Her last two launches shipped on schedule.

Step 9

Collect peer evidence

Two aimed at craft, the thinnest dimension. One from support on the checkout result.

"Jordan's critiques changed how our team reviews work. We stopped debating taste and started asking what would move activation. We went from 22% to 41% activation, and that habit is a big reason we kept iterating instead of calling it done."PM, onboarding team

"I've adopted the system's component spec as my default. It's saved my team about a sprint per feature, and defects on shared components are down by about half. Reviews with Jordan's team are the only ones where design and eng arrive already aligned."Engineering lead, payments

"Failed-payment tickets dropped by about 490 a month after the checkout redesign. That's roughly $50K a year in support cost we're not spending."Support operations lead

Step 10

Pre-empt the hardest question

The question. "Completion went up 13 points, but the platform migration shipped the same quarter and three teams touched this flow. How much of that lift is your design, and how much is just a better quarter?"

The answer. That's the right thing to ask, so I checked it before putting the number in. We launched the new flow as a 50/50 test for six weeks before going to 100%, and the migration hit both groups. The old flow stayed at 62% while the new one reached 74%, so the migration didn't move completion on its own. Finance's $370K uses the full 13 points for that reason. Two limits: the test ran six weeks, so $370K is an annualized estimate, not booked revenue, and billing's step fix shipped alongside my flow, so I can't split the lift between the two. What I own is getting both teams into one flow, and that joint release is what made it work. The craft gap is the other thing I expect you to push on, and it's already in my one-pager.

Step 11

Hand it to your manager

Hi [Manager], attached is my draft case for Staff. I think it shows I work across team boundaries, that other teams adopt what I build, and that it moves numbers finance cares about. Could you read it before it's due and tell me honestly where it's weakest? I know craft depth is thin and I'd rather fix it with you than have it come up cold.

What a finished packet looks like.

Jordan Lee, Senior Product Designer. Case for Staff.

Thesis

Projects

Scope

Mentorship

Known gap