What it does
Map your real PM skills against concrete behaviors at the Self, Team, and Org levels, get a personalized read on what to work on next, and walk into a review, promo case, or mentor conversation with evidence instead of vibes.
Currently
Live MVP, free, no signup, your data stays in local storage. A behavioral-anchor flow rates you across 21 skills in 7 domains and ends on a personalized growth plan in my voice. It is also nearly uninstrumented: only one completion event fires today, so I am steering on hunches until the funnel events land. That is the next thing in.
Roadmap
/now-next-later-roadmapNow/Next/Later Roadmap- Core instrumentation + lock the acted-on definition
- Intake question: what made you take this?
- Kate's named-coach commentary library on common patterns
- Behavioral anchor clarity pass across all 21 skills
- Role-selection reframe at the flagged drop point
- Manager-ready PDF + save CTA + bring-this-to-your-review framing
- Cross-surface coaching-inquiry attribution
- Plan-perceived-as-personal micro-survey
- Deepen the pattern-based growth-plan logic
- Accessibility pass: keyboard, screen reader, text-equivalent results
- Why this rating micro-explanations
- Pick-your-focus-skill interaction (affirm-intent signal)
- Minimal weekly metrics read
- Scope the manual shared-rubric facilitation kit for leaders
- Share loop (held against its RICE until demand has a home)
- Soft coaching bridge (gated behind a capacity-light home)
- Productized async deep-plan tier (paid, no live hour)
- Group / cohort coaching offer
- Real-distribution percentiles from anonymous data
- Return / re-measure: growth since last time
- Multi-user team mode (parked until leader demand is proven)
Recently shipped
- Jun 9, 2026Restyled with a per-tool palette and recaptured the result thumbnail so the shareable artifact looks like something a PM is proud to hand over.
- Jun 2, 2026Shipped the MVP: 21 PM skills across 7 domains, behavioral anchors at Self / Team / Org levels, and a personalized growth plan with my commentary.
How I know it's working
instrumented · awaiting trafficThe count of completed assessments where the user signals the personalized growth plan landed and they intend to act on it: they save or export the result, share it, or affirm intent in-tool (this is useful / I'll work on this / picked a focus skill). A raw completion with none of these does not count. It is honest user value and the cleanest free-side leading indicator of a coaching inquiry, and it cannot be faked with a harder CTA.
The events are wired into a privacy-first store. Counts populate here as people use it. This is the honest part of the work: instrument first, claim later.
How it's built
How I built it
Skill Matrix started as the top-of-funnel surface for my coaching practice, but it had to earn its keep as a real product first. A PM rates themselves across 21 skills in 7 domains, every rating anchored to a concrete behavior at the Self, Team, and Org levels, and walks away with a personalized growth plan in my voice. So I ran the full product process on it, the same one I would run for a client.
What the discovery surfaced
A jobs-to-be-done pass made the core wedge sharp: this is not "rate yourself," it is "replace a vague feeling about where you stand with a clear, behaviorally-grounded read and one or two next moves you trust." Maya, the plateaued mid-level IC with no manager coaching the craft, became the primary persona. Devin, the aspiring or nontraditional-path PM, is who the judgment-free, tenure-free philosophy was written for. Priya, the PM lead who wants one shared language, is a real adjacent market the single-user design cannot fully serve yet. And Sam is the coaching-curious follower whose job is satisfied as a side effect of the others landing well.
That framing forced a hard call on the North Star. The tool is free and its business purpose is feeding my coaching funnel, which pulls toward a business metric. But the judgment-free promise is load-bearing and fragile: the moment the experience optimizes for selling coaching, the emotional jobs collapse and the funnel dies with them. So the North Star is a user-value metric engineered to predict the business outcome, acted-on growth plans, with coaching inquiries per 100 completions sitting one step downstream as the guardrail, not the compass.
What shipped, and what got cut
The MVP is deliberately browser-only with local storage, so there is no signup and nothing to trust me with, at the cost of cross-device sync and durable measurement. RICE scored the backlog, but I read it through a second lens, value per unit of my own time, because the binding constraint is my hours, not market demand. That correction pushed the share loop and the soft coaching bridge down below their RICE rank: a working share loop just manufactures inquiries into a waitlist for a practice that is already at capacity.
The pre-mortem named the real danger plainly. The whole business case rests on one unvalidated causal chain (acted-on plan to trust to inquiry), and the quarter can complete in full while leaving it unproven. So the bet is discipline over surface area: instrument first, run the convert interviews in week one to test the thesis events cannot, put more of my judgment into the auto-plan, and scope a capacity-light home for demand before manufacturing any. The honest current state is that almost nothing is instrumented yet, so until those events land I am steering on conviction, not data.
The strategy behind it
A snapshot of the product process this app was built from. Each artifact comes from a k8 skill you can run yourself. It reflects the original plan, not the live state above.
Jobs to be done
/jtbd-analysisJTBD Analysis| When I… I want to… | Importance | Frequency |
|---|---|---|
| When I sense I have plateaued or I cannot name my own strengths and gaps, I want to map my real PM skills against concrete behaviors, so I can replace a vague feeling with a clear picture. | high | medium |
| When I want to grow but do not have a manager or coach giving me useful direction, I want a personalized read on what to work on next, so I can move forward without guessing. | high | medium |
| When I am prepping for a performance review, promotion case, or 1:1 with a mentor, I want a structured, credible read on my skills, so I can argue for myself with specifics instead of vibes. | high | low |
| When I am earlier in my career or came in through a nontraditional path, I want to be measured on what I can actually do rather than years served or title, so I can feel that the bar is fair and see a real path up. | high | low |
| When my PMs grade themselves on different scales and I cannot coach consistently, I want one shared vocabulary and rubric for PM skill, so I can have the same conversation with everyone and calibrate fairly. | medium | low |
| When I am curious about coaching but not ready to commit or pay, I want a free, judgment-free way to feel understood and get a taste of the approach, so I can decide whether to go deeper. | medium | low |
Persona
/persona-createPersona Create- Get an honest read on where she stands across the whole PM craft, not just delivery
- Find one or two specific things to work on next that actually move her forward
- Build a case she can take into her next review, since her manager will not build it for her
- Generic PM quizzes give a score with no behavioral grounding and nothing to act on
- Her growth depends on a manager who does not coach the craft
- She conflates busy-and-shipping with growing, and has no mirror to check it
Persona
/persona-createPersona Create- Find out whether their actual skills add up to PM, independent of title or tenure
- See a fair, behavior-based path in rather than a tenure wall
- Quiet the impostor feeling with a bar that is real and reachable
- Every measure they meet rewards tenure and title, which they do not have
- They cannot tell which adjacent skills transfer and which gaps are real
- Impostor feelings, with nowhere safe to test reality
Persona
/persona-createPersona Create- Establish one shared vocabulary for PM skill every PM understands the same way
- Calibrate and coach fairly across the team
- See the spread of skills so she can staff and develop deliberately
- Her PMs self-assess on different scales, so coaching does not transfer
- Building a competency matrix herself is heavy and drifts the moment she ships it
- No easy team view of who is strong where
Persona
/persona-createPersona Create- Satisfy curiosity with zero commitment
- Feel understood, get a small useful read, and decide whether to go deeper
- Decide privately, without pressure, whether coaching is for them
- The gap between curious and book-a-paid-call is too big to cross cold
- Skeptical of anything that smells like a sales funnel
- Does not want to be sold to while feeling vulnerable about their skills
Prioritization and risks(the internal detail)
RICE prioritization
/rice-scoreRICE Score| Feature | Reach | Impact | Conf. | Effort | Score |
|---|---|---|---|---|---|
| Convert + Maya interviews | 560 | 3 | 70% | 0.25 | 4704 |
| Intake: what made you take this? | 560 | 1 | 70% | 0.1 | 3920 |
| Core instrumentation + lock acted-on definition | 560 | 2 | 80% | 0.25 | 3584 |
| Role-selection reframe | 560 | 2 | 55% | 0.25 | 2464 |
| Kate's named-coach commentary library | 450 | 3 | 60% | 0.75 | 1080 |
| Behavioral anchor clarity pass | 560 | 2 | 65% | 0.75 | 971 |
| Manager-ready PDF + save CTA + review framing | 560 | 2 | 55% | 0.75 | 821 |
Pre-mortem risks
/pre-mortemPre-mortemTrust-leak from a share loop plus a soft coaching bridge: completions rise while inquiries stay flat, the result screen starts to feel like a sales funnel at the moment a PM is deciding whether to trust the read, and the judgment-free promise quietly breaks.
highMitigation: Ship the weekly trust-leak read and an inquiry guardrail before or with the bridge, with a hard auto-pause rule if inquiries-per-100 drop while completions rise across two weekly reads. A/B the bridge against a no-bridge control.
The core causal chain (acted-on plan to trust to coaching inquiry) never gets validated because the convert interviews do not happen at depth, so a busy, well-instrumented quarter still cannot answer whether the tool makes money.
highMitigation: Treat convert interviews as a named week-one deliverable run in parallel with instrumentation, target 5 to 8. If too few past converts exist to interview, that is itself a first-class finding and the fallback (prospect and IC interviews) activates immediately.
The email gate on the PDF leaks trust at the highest-intent moment: it sits on the strongest acted-on signal and contradicts the free, no-signup promise, converting emails while suppressing the qualifying event.
highMitigation: Deliver the PDF ungated and make email an optional send-me-a-copy ask after the artifact is in hand. If the gate stays, A/B it on save rate and downstream inquiry rate, and stop claiming no signup in the hero.
Growth-bound effort in a capacity-bound practice manufactures a waitlist, not revenue: my hours go to a share loop and a bridge that grow demand I cannot serve, since the practice is capped by my time, not the market.
mediumMitigation: Scope at least one capacity-light home for demand this cycle (productized async tier or facilitation kit) before widening the funnel. Gate the share loop's expansion on a scoped home existing, and add a capacity-utilization watch to the weekly read.
Solo-operator key-person risk: build, design, content, and all coaching delivery are one person, so a multi-week gap halts the plan and the revenue at once.
mediumMitigation: Sequence the cheapest, highest-leverage write-once items first (instrumentation, commentary library, anchor pass) so a compressed quarter still banks durable value, and prefer write-once judgment artifacts over recurring-hour commitments.