[ ]productinterview
Run a free mock

product sense · standard

Design a product for the elderly

Design a product for the elderly.

Updated Jun 2026 Calibrated to the strong-hire bar

The most common failure on this question is treating “elderly” as a disability category rather than a human segment with goals and identity. Candidates who open with “I’d design a phone with bigger buttons and simpler navigation” are describing what iOS and Android already do natively with accessibility settings. Interviewers are not measuring your knowledge of accessible UI patterns. They are measuring whether you can find a real, underserved problem, anchor to a specific person, and connect the product to a job that person actually cares about.

Scope it before you solve it

“Elderly” spans 30 years and three radically different realities. Name the split out loud, then pick one and defend it.

  • 65-74, active retirees. Ninety percent own smartphones (AARP 2026, up from 55% in 2016). Seventy-one percent bought new tech in 2025. The constraint is not capability or access. The problem is design contempt: 60% of older adults say technology lacks age-conscious design despite being active users.
  • 75-84, transitioning. Digital fluency is inconsistent. This is the highest-stakes moment: independence is still possible but fragile. The margin between “managing fine” and “needing more help” is often one missed medication or appointment. Currently underserved by both consumer tech and formal care infrastructure.
  • 85+, high dependency. The buyer is usually an adult child or care facility, not the user. The user and the economic decision-maker are different people. Consumer tech is not designed for this handoff and regularly fails both.

Pick the 75-84 transitioning segment and say why explicitly: it is the largest unmet gap, independence is still real and worth protecting, digital adoption is high enough that software can reach them, and the business has a clear buyer (adult children) plus a motivated user.

The real job to be done

The job is not “take a pill at 8am.” It is: stay in control of my own day without having to ask my daughter if I remembered something.

That reframe changes the entire design. The surface problem is missed reminders or complicated interfaces. The emotional job is identity: I am still someone who manages my own life. A 2025 systematic review of 132 studies (PMC) confirmed four validated design pillars for older adults: simplified navigation, enlarged touch targets, voice interaction, and error-tolerant interfaces. Strong candidates name these and then go further. These are table stakes for usability. Lovable requires that the product actively reinforces the user’s sense of agency, not just their ability to complete tasks. A product that makes users feel monitored or cognitively diminished will be powered off within a week regardless of how accessible it is.

The 2026 feasibility shift

In 2026, voice AI, ambient sensing, and LLM-driven personalization make nearly any interaction pattern buildable. The constraint is no longer “can we do this technically.” The hard problems are lovability and viability.

For this segment, voice-first is the highest-return interaction modality. AARP 2026 data shows 80% of older adults who used AI voice assistants say the devices help them live independently and safely at home. AI usage among adults 65+ nearly doubled from 18% in 2024 to 30% in 2025. The silver economy is projected at $15 trillion by 2030.

A strong 2026 answer names what AI makes newly possible: proactive, context-aware reminders that learn the user’s actual schedule from behavior rather than requiring manual entry. “You usually take your morning pill at 7:45. It’s 8:10 now. Do you want me to remind you again in 10 minutes?” That is not a feature list. That is the difference between a tool and a product that earns daily trust.

The Microsoft alarm clock cluster asks about elderly, blind, deaf, and deaf-blind variants. These are distinct prompts that require different reasoning. The elderly alarm clock question is not a sensory substitution problem. It is a cognitive load and independence maintenance problem. Candidates who conflate the two fail on empathy depth, which is exactly what Microsoft’s interviewers probe for in this cluster.

The buyer/user split: the viability trap most candidates miss

For adults 75+, the economic buyer is often an adult child or care facility, not the user. This creates a tension that will collapse the product if unaddressed.

The buyer wants visibility, reassurance, and evidence that the parent is okay. The user wants privacy, independence, and not to feel surveilled. A product that optimizes for only the buyer feels like a monitoring device. The user stops engaging. The buyer stops paying. The product fails both sides.

A strong answer names this explicitly and proposes a design that gives each constituency what they actually need: the user gets a calm, non-condescending daily anchor that respects their autonomy; the family gets gentle, consent-based notifications only when something is genuinely missed twice, not once. The product earns trust from both sides, or it earns trust from neither.

strong

"Before I dive in, let me scope this. Elderly covers 65-year-olds who just retired and 90-year-olds in assisted living. Those are completely different product problems. I'll focus on adults 75-84 living independently: AARP 2026 data shows 90% of adults 65 and over own smartphones, and 60% say technology doesn't account for their age. For the 75-84 cohort specifically, independence is real but increasingly fragile. That's where the design problem actually lives.

The job I'm designing for is not 'set a medication reminder.' It is: stay in control of my own day without needing to ask someone else for confirmation. Missed medication is emotionally loaded not primarily because it's dangerous. It signals a loss of autonomy. The product has to address that identity dimension, not just the scheduling one. Usable means they can complete the task. Lovable means they feel like themselves while doing it.

The product I'd design is a voice-first daily anchor device: an AI assistant that learns the user's actual routine from behavior rather than requiring setup, uses proactive reminders in natural language, and escalates to a named family contact only on a second missed confirmation. The voice persona is peer-level and calm, not childlike. No app required for core function. In 2026, 80% of older adults who've used AI voice assistants say it helps them live independently. That's the modality to build on.

I'd address the buyer/user split directly: family members receive a weekly digest, not a live activity feed. The user controls what the family sees, set up once during onboarding. This is the only design that sustains adoption from both sides. The family member is often the economic buyer, and they'll cancel if they feel the product isn't giving them peace of mind. But the user will stop engaging if they feel watched.

Primary metric: daily active confirmation rate (are users engaging with reminders or dismissing them?). That's the clearest signal the product is calibrated correctly. Secondary: independence duration, measured as days between family escalations. Lovability metric: NPS measured separately from user and family buyer, both must score above 50. Viable go-to-market: direct to adult children as a gift purchase at $99-149 device plus $12/month subscription, targeting the $15 trillion silver economy with distribution through AARP partnerships and primary care referrals."

weak

"I'd design a phone for elderly people with bigger buttons, larger text, high contrast, and simpler navigation. I'd add voice control, an emergency SOS button, and a medication reminder. For metrics I'd track daily active users and app store ratings." This treats 65 and 90 as the same user. It lists accessibility features that iOS and Android already ship natively, with no analysis of what gap is actually unmet. It picks no specific segment and names no specific person, which means the solution can't be grounded in a real unmet need. It ignores the emotional job entirely: bigger text is usable, but a product that helps someone feel like they haven't lost control of their own life is lovable. Most critically, it ignores the buyer/user split: for the 75+ segment the economic buyer is often an adult child, and a product that doesn't serve both sides will fail commercially. Interviewers at Microsoft, Google, and Meta have been probing the usable/lovable distinction since 2024. An answer that stops at accessibility compliance doesn't clear that bar.

Metrics that signal product thinking

Weak answers list DAU and ratings. Strong answers instrument the actual job:

  • Daily confirmation rate. Are users engaging with the reminder, or muting it? Muting signals the product is overbearing, not helpful.
  • Independence duration. Days between family escalations. If this is increasing over time, the product is working.
  • Dual NPS. Measured separately from user and family buyer. Both must score high or you have a churn problem on one side.
  • Time-to-first-confirmation. If users confirm within a few minutes, timing is well-calibrated. If it consistently takes 45 minutes, the schedule is wrong and the reminder is noise.

What the interviewer is actually checking

At Microsoft specifically, which asks this question in APM and PM loops alongside the alarm clock variants, the signal is empathy depth combined with product viability. The interviewer distinguishes candidates who understand the emotional job (maintaining autonomy) from those who stop at the functional job (completing a task). Feature lists without segment reasoning and prioritization rationale signal feature-shipping instincts rather than product thinking.

The broader skill: resist the temptation to solve for the label (“elderly”) and instead identify the specific person, the specific job, and the specific gap in the current landscape.

The jobs-to-be-done framework is the right lens for reframing “design a product for elderly users” as “design a product that maintains someone’s sense of being in control of their own day.” For the 2026 feasibility shift and why lovability is now the hard constraint, see feasibility is free and lovable, not just usable. The design an alarm clock for the blind question is a related but distinct Microsoft cluster prompt: the elderly variant is about cognitive load and identity, not sensory substitution.

Asked at

Related