The framework in one page

#Why you need a method

Product sense questions sound wide open: "Design a product for people who travel with pets." The interviewer is scoring how you get to an answer: whether you understand people, choose with reasons and land on something specific enough to build. The full list of signals is in what interviewers actually score.

Most candidates who struggle here are not short of ideas. They are short of shape. They jump to features, wander between users and run out of time before measuring success. A method tells you what to say next when your mind goes blank.

This module uses one method for every product design question: the Design Ladder.

Its order (goal, user, problem, solutions, choice, measure) is the backbone most product-design methods share. What it adds is a sentence you commit to at every rung and a summary that reads them back.

#The Design Ladder

The ladder has six rungs, climbed in order, each resting on the one below: you can't pick a solution before you know whose pain you are solving, or the pain before you know the goal.

The times assume a 30-minute question. Treat them as a budget, and follow the interviewer if they steer you.

Key idea Every rung ends with one sentence you commit to out loud. The walk down, your summary, is just those six sentences read back in order.

#Two habits that make the ladder work

Commit at every rung. The rung sentence forces a decision ("I'll design for first-time owners") instead of a survey of possibilities, and hands the interviewer a clean place to redirect you. After two or three of the rung sentences, usually Person and Bet, a quick check such as "Does that work, or would you like me to go another way?" costs three seconds and often saves five minutes.

Write the sentences down as you go. Jot each rung sentence in a few words on a notepad, or a blank doc on a video call. At the end you read your notes instead of composing a summary under pressure.

#What sits inside each rung

Each rung has one small tool, taught properly in the next five lessons.

• Frame. Ask two or three questions whose answers would change your design, and state everything else as an assumption. The goal has two halves: what the company wants and what the user wants. Then announce the route in one sentence: "I'll pick a user, find their biggest pain, give a few options, choose one and say how I'd measure it."
• Person. Group users by what they are trying to do, not by age alone. Pick one group using three tests: they are poorly served today, there are enough of them to matter, and helping them serves the goal.
• Pain. Walk through before, during and after the task. Rank each pain on how often it happens and how much it hurts.
• Options. Give a near option (the obvious improvement), a far option (ambitious, maybe new technology) and a sideways option (a different angle, often not software at all).
• Bet. Put the options in a small table scored on value to your chosen person, effort, and risk. Pick one, say the trade-off out loud, and describe the first version you would ship.
• Proof. Name a headline metric tied to the goal, two helpers that explain why the headline moved, and a guardrail you refuse to harm.

#A short worked example

Here is the ladder on a small prompt, compressed so you see the whole shape. The worked answers later in the module show it at full length.

Interviewer Design a product to help people keep houseplants alive.

You Two quick questions. Am I a startup or part of an existing company, such as a garden retailer? And is hardware in scope?

Interviewer You're a startup. Anything is in scope.

You Then I'll assume the US, and that as a startup the goal is one group who will pay and keep using it. So: we're building a plant-care product for a group I'll pick next, and success means plants that survive and customers who come back. From here I'll pick a user, find their biggest pain, give a few options, choose one and say how I'd measure it.

Person. Groups: first-time owners in apartments, collectors, people given a plant as a gift, and offices. Collectors know what they are doing and offices buy differently, while first-time owners lose the most plants and get the least help. Rung sentence: "I'll design for first-time owners in apartments, because they lose the most plants and have nobody to ask."

Pain. Before: they buy plants that don't suit the light in their home. During: they water on a guess and don't notice trouble until leaves drop. After: the plant dies and they stop buying plants. Watering on a guess happens every week and is the most common way the plant dies. Rung sentence: "The pain worth solving is that they don't know when to water, and guessing wrong kills the plant."

Options.

• Near: an app that sends watering reminders based on the plant type.
• Far: a stick you push into the soil that is set for the plant type and simply lights up when it needs water. Basic moisture meters exist, but they show a dial people don't know how to read.
• Sideways: sell each plant already potted in a self-watering pot matched to the light in your window, so there is nothing to remember.

Bet.

Rung sentence: "I'd build the soil stick first, accepting hardware risk instead of the cheaper app, because a reminder guesses and the stick knows." The first version is one stick, one light, no app.

Proof. Headline: the share of buyers who say their plant is still alive 90 days after purchase, from a short follow-up survey. Helpers: the share who buy a second stick within six months, and the share who say they watered when the light came on. Guardrail: the return rate, because a stick with false signals gets sent back. Rung sentence: "We'll know it works if 90-day plant survival rises without returns climbing."

Walk down.

You To summarize: as a startup, we want a group who'll pay and stay. I picked first-time apartment owners because they lose the most plants. Their real pain is watering on a guess. I looked at a reminder app, a soil stick and a self-watering bundle, and I'd build the soil stick first, accepting hardware risk because it measures instead of guessing. We'll know it works if plants survive past 90 days without returns climbing.

The summary adds nothing new: it is the six rung sentences in order.

#What the weak version sounds like

"I'd build an app with watering reminders, a plant identifier using the camera, a community forum, and maybe augmented reality for placing plants."

Four features, no user, no pain, no choice and no measure. Some ideas might be good, but nothing was chosen, so there is no evidence of judgment.

#How the rest of this module uses the ladder

The next five lessons take the rungs one or two at a time, from Clarify the prompt and set the goal to Close with how you would measure it. The favorite product question runs the ladder in a different order, twelve full answers start at Worked answers 1 to 4, and Common traps and how to recover gives you lines for when an answer goes sideways.

Try it For "Design a product for people who commute by bike", give yourself five minutes to write only the six rung sentences, then read them aloud as a walk down. Any sentence without a "because" is not finished.

#Recap

• The Design Ladder has six rungs in a fixed order: Frame, Person, Pain, Options, Bet, Proof.
• Every rung ends with one sentence you commit to out loud, and you jot it down.
• The 30-second summary is the walk down: your six sentences read back in order.
• Options come in three kinds (near, far, sideways), and the Bet always names what you give up.
• Proof is one headline metric, two helpers and one guardrail.