# Scoring frameworks: impact effort matrix, RICE and more > How an impact effort matrix sorts quick wins, big bets, fill-ins and money pits, how a value vs effort scorecard and RICE turn criteria into scores, how to keep confidence honest and read a ranking, where cost of delay, WSJF and MoSCoW fit, and why a score only starts the conversation. - Course: [Outcome-Based Product Roadmaps](https://productovi.com/courses/outcome-based-roadmaps.md) - Module: Prioritizing with evidence - Lesson 11/15, ~46 min - Source: https://productovi.com/courses/outcome-based-roadmaps/scoring-frameworks ## From the founders' scores to an impact effort matrix **Alex Tutor:** Ian stops by your desk on CatChow's product team with news from the founders. **Ian:** The founders scored the spreadsheet themselves over the weekend. Loyalty points: ten out of ten. Can we just use their scores? **Sofia:** Ten out of ten for what, though? Nobody said what a ten means. **Alex Tutor:** That's the gap. A **prioritization framework** is an agreed, repeatable way to rank items against the same criteria, so anyone can see how a score was reached. **Sofia:** Where do I start? There are dozens of them. **Alex Tutor:** With the simplest: an **impact effort matrix**, or value vs effort matrix. It's a 2×2 grid, value on one axis and effort on the other, with four quadrants. High value and low effort are **quick wins**; high value and high effort are **big bets**. **Alex Tutor:** Low value and low effort are **fill-ins** for spare time. Low value and high effort are **money pits**: avoid them, however exciting they sound. ![High value and low effort are quick wins, high value and high effort big bets, low value and low effort fill-ins for spare time, low value and high effort money pits.](https://productovi.com/courses/outcome-based-roadmaps/diagrams/m4-l2-value-effort-grid.png) *The simplest framework: value against effort.* **Sofia:** So each corner of the grid comes with its own instruction. > **Exercise:** Match each corner of the value vs effort grid to what a team should usually do with items there. ## A value / effort scorecard **Sofia:** But “high value” is still a feeling. How do I make it less of one? **Alex Tutor:** Score value against your objectives: for each one, give a candidate 0 to 2, meaning no effect, some or a lot. Add them up. That's the value side of a **value vs effort scorecard**. **Lucy:** And effort? Please don't ask me for days. I can tell you small, medium or large. **Alex Tutor:** That's all a scorecard needs. **T-shirt sizing**: XS, S, M, L and XL, counted as 1 to 5. Relative sizes, with no calendar attached. ![A row of plain T-shirts on hangers grows from tiny to huge, and a developer points at one in the middle.](https://productovi.com/courses/outcome-based-roadmaps/inline/scoring-frameworks-illustration-2.webp) **Alex Tutor:** Divide value by effort, then multiply by confidence as a risk discount. The online order form: 2 for clinics, 0 for subscribers, size S, 80% sure. That's 2 ÷ 2 × 0.8 = 0.8. **Brandon:** Here's a cheaper idea: let clinic orders skip the warehouse queue. Clinics would get their boxes a day sooner. **Alex Tutor:** Good for clinics, but subscriber boxes would wait longer, and running out is their top complaint. A scorecard allows negatives for exactly that: −1 for subscribers. Score it. > **Exercise:** Score Brandon's idea with the value vs effort scorecard: clinic orders skip the warehouse queue. Clinics: **+2**. Subscribers: **−1**. Size: **S (2)**. Confidence: **80%**. Value is the sum of the objective scores; the score is value ÷ effort × confidence. ## RICE scoring: reach, impact, confidence, effort **Sofia:** The scorecard still mixes both objectives into one value. Is there something more specific? **Alex Tutor:** **RICE**: Reach × Impact × Confidence ÷ Effort. Reach is how many customers an item touches in a quarter. Impact is how far it moves each of them towards one objective's key results. **Alex Tutor:** Impact uses fixed steps: 3 massive, 2 high, 1 medium, 0.5 low, 0.25 minimal. Confidence: 100% for tested results, 80% for a clear pattern in data, 50% for a few requests or a hunch. **Alex Tutor:** That's the core. Our [RICE framework guide](/blog/rice-prioritization) has a longer worked example and the usual traps, and the [RICE calculator](/tools/rice-calculator) does the arithmetic for a whole list. Today, the numbers are CatChow's. **Sofia:** Sofia, Lucy and I filled in the nine candidates. Effort is in person-months, for everyone involved. **Alex Tutor:** Good. And the pilot has already moved one number: both clinics placed their last two weekly orders through the online form, without a call. That's why the form has 80%. **Dataset (CSV):** [CatChow candidate solutions, scored in November](https://productovi.com/datasets/outcome-based-roadmaps/scoring-frameworks/catchow-candidates.csv) Nine candidates from CatChow's roadmap with the RICE inputs Sofia and Lucy agreed in mid-November. Reach is per quarter; effort is in person-months for everyone involved. Synthetic data made for this course. - `id`: Candidate ID - `candidate`: Candidate solution - `theme`: The roadmap theme it serves (none = no theme) - `objective`: 1 = grow subscription revenue by keeping subscribers longer; 2 = grow the vet clinic channel - `reach_per_quarter`: How many customers it touches in a quarter - `reach_unit`: Who is counted in reach - `impact`: Effect on each customer towards the objective's key results: 3, 2, 1, 0.5 or 0.25 - `confidence`: How sure we are about reach and impact: 1.0, 0.8 or 0.5 - `effort_person_months`: Total work for everyone involved, in person-months **Alex Tutor:** Before reading the whole table, work out one row by hand: flexible delivery dates. > **Exercise:** Using the dataset **CatChow candidate solutions, scored in November**, calculate the RICE score of **flexible delivery dates** (C2). ## Reading the ranking **Sofia:** All nine are scored. Top of the list: Apple Pay and Google Pay at checkout, with 1,200. Something's off. **Alex Tutor:** The math is right. It's cheap, reaches many buyers and barely moves your key results. A quick win can top a ranking without being strategic. **Sofia:** And every clinic item scores below 50. The portal gets 9. **Alex Tutor:** Reach counts 54 clinics against thousands of subscribers, so the numbers don't compare. Rank within each objective. How to split capacity between objectives is a strategy decision, not a score. **Alex Tutor:** Check one clinic row yourself before you look at the bottom of the table. > **Exercise:** Using the dataset **CatChow candidate solutions, scored in November**, calculate the RICE score of the **online order form for clinics** (C6). ## Honest confidence **Ian:** I asked the founders to fill in your RICE columns. For loyalty points: impact 2, 100% confidence, two months of work. Why is yours so much lower? **Sofia:** Our evidence is a few comments from long-term subscribers. That's 50%, not 100%. **Alex Tutor:** And impact 1, because points also reward people who were staying anyway. Much of the effect lands on subscribers you weren't going to lose. **Lucy:** And two months is only my team's part. Add the points rules, finance, emails, support training and marketing, and it's eight person-months. **Alex Tutor:** That's why effort counts everyone. Same formula, honest inputs: work out the score from the table and compare it with theirs. > **Exercise:** Using the dataset, calculate the honest RICE score of **loyalty points** (C4). Then compare it with the founders' version: reach 5,000, impact 2, confidence 100%, effort 2. ## False precision and gaming **Sofia:** So loyalty points get 312.5 and the quiz gets 450. Is the quiz clearly better? **Alex Tutor:** Not clearly. Each score multiplies four estimates. Writing 312.5 is **false precision**: round the scores, treat close ones as ties, or group them into high, medium and low. ![A tall tower of four wobbly stacked blocks leans to one side, with a tiny neat flag on top.](https://productovi.com/courses/outcome-based-roadmaps/inline/scoring-frameworks-illustration-3.webp) *Each score multiplies estimates; round it and argue about the inputs.* **Sofia:** And the founders' ten out of ten? **Alex Tutor:** That's gaming: whoever wants an item rates it high. The guards: define each scale step with a real example, score together, ask for evidence behind anything above 50%, and later compare estimates with results. **Sofia:** So the score isn't the decision. **Alex Tutor:** No. A score starts the conversation; it doesn't end it. It shows which estimates are worth arguing about. > **Exercise:** Which practices keep a team's scores honest? Select all that apply. **Sofia:** Back to the ranking: the warehouse is dead last, at 1.8. **Alex Tutor:** Of course: on its own, it changes little for clinics. It's the enabling work from last time. Judge it by what it unlocks; in time, it goes first anyway. > **Exercise:** Apple Pay and Google Pay at checkout tops objective 1 with a RICE score of 1,200, but its impact is 0.25 and it serves no theme. What should Sofia do with it? ## Urgency and scope: cost of delay, WSJF, MoSCoW **Brandon:** One thing RICE ignores: the conference on 23 May. If clinics could order online by then, I could sign new clinics right at our stand. **Alex Tutor:** True, RICE has no clock. **Cost of delay** asks what we lose for every week an item isn't done. For the clinic solution, the conference makes that loss concrete. ![An hourglass stands next to a stack of coins that slowly slide away through a gap in the table as the sand runs.](https://productovi.com/courses/outcome-based-roadmaps/inline/scoring-frameworks-illustration-4.webp) **Alex Tutor:** **WSJF**, weighted shortest job first, divides cost of delay by job size and takes the highest ratio first. Short, urgent work goes ahead of long work of similar value. **Alex Tutor:** Try WSJF on two clinic options, in relative points. > **Exercise:** In relative points: the online order form has a cost of delay of **8** and a size of **2**; the clinic portal has a cost of delay of **9** and a size of **6**. Under **WSJF**, which goes first? **Sofia:** And MoSCoW? The founders' sheet had “Must have” written next to loyalty points. **Alex Tutor:** The **MoSCoW method** sorts a release's scope into Must, Should, Could and Won't have this time. It communicates priorities you've already set; it doesn't set them. **Alex Tutor:** So what should the founders' “Must have” mean? Pick the right use. > **Exercise:** Two ways to use the **MoSCoW method** at CatChow. Which one fits what MoSCoW is for? **Sofia:** Brandon will read the table and ask whether his portal is out. **Alex Tutor:** It isn't out; it's ranked. It scores low because it's big and untested, not because it's a bad idea. Explain that to him in a way he can live with. > **Exercise:** Brandon asks whether the clinic ordering portal is out. Write Sofia's reply (2–4 sentences): why it ranks below the online order form and automatic reorders, and what would move it up. Don't dismiss it, and don't promise a date. ## Which estimate would change the order? **Sofia:** If a score only starts the conversation, what do I bring to it? **Alex Tutor:** The one estimate that would flip the order. For each close pair, ask: which single number, if it's wrong, would swap them? **Sofia:** Automatic reorders for clinics: at one person-month instead of three, they'd score 54 and pass the online form. **Alex Tutor:** Exactly, so that estimate is worth checking before anyone commits. Often a cheap test that raises confidence is worth more than another meeting. **Sofia:** Then let me find one like that in objective 1. **Alex Tutor:** Find it, and say how you'd check it cheaply. > **Exercise:** In objective 1 of the dataset, find one estimate that, if it were wrong, would change the order of two candidates. Say what the new score would be and how Sofia could check that estimate cheaply (2–4 sentences). ## Case: Chapterly's RICE sheet **Sofia:** Does anyone get RICE right the first time? **Alex Tutor:** Rarely. Chapterly's head of e-commerce sent me his RICE sheet for next quarter. He wants to build the top item first. It looks very rigorous. **Sofia:** “Looks rigorous.” I've stopped trusting that phrase. **Alex Tutor:** Check the units, the scales, the confidence and the effort. Then say what order would make more sense, and why. > **Exercise:** Critique Chapterly's RICE sheet and suggest a better order. ## What to remember **Sofia:** Scoring makes the reasoning visible. Ian sees why loyalty points rank low, Brandon sees what would lift the portal, and Lucy estimates in sizes, not days. **Alex Tutor:** And a score starts the conversation. The decision, and the reasons behind it, are still yours to write down. > **Key ideas** > > - A **prioritization framework** ranks items against the same criteria, so anyone can see how a score was reached. The value vs effort grid, or **impact effort matrix**, sorts quick wins, big bets, fill-ins and money pits. > - A **value vs effort scorecard** adds up each item's effect on your objectives (negatives allowed for trade-offs), divides by effort, often in **T-shirt sizes**, and multiplies by confidence. > - **RICE** = Reach × Impact × Confidence ÷ Effort. Use one reach period, fixed steps for impact and confidence, effort for everyone involved, and compare items within one objective. > - **Cost of delay** and **WSJF** add urgency; **MoSCoW** communicates release scope once priorities are set. > - Scores multiply estimates: avoid **false precision** and gaming, find the estimate that would change the order, and let the score start the conversation, not end it. **Alex Tutor:** Next time: the first monthly review. Lucy's migration test is done, Brandon wants more clinics, and the founders aren't finished with loyalty points. Time for trade-offs, and for saying no. - Previous lesson: [Feature prioritization: why gut feeling fails](https://productovi.com/courses/outcome-based-roadmaps/why-gut-feeling-fails.md) - Next lesson: [Trade-offs, the iron triangle and saying no](https://productovi.com/courses/outcome-based-roadmaps/trade-offs-and-saying-no.md) --- Productovi Learning. Free, no sign-up needed to read.