glance Product Manager Interview: Questions, Experience & Prep (2026)

Glance is InMobi's lock screen content platform, running on hundreds of millions of Android phones across India and Southeast Asia. Its product sits at the crossroads of AI-driven personalisation, short-form content, gaming, and mobile advertising. Every PM at Glance is, in effect, shaping what users see the instant they lift their phone.

As of July 2026, Glance has 39 open Product Manager roles in India. For context on the wider PM market, Bangalore leads hiring with 271 open roles, Delhi follows with 177, and Pune, Mumbai, Hyderabad, and Chennai add further opportunities across a total of 2,009 PM openings tracked nationally.

Candidates report a process that typically runs three to five rounds. You will generally encounter a take-home product case or written brief, one or two structured PM rounds covering product thinking and metrics, a cross-functional or design discussion, and a final round with senior leadership. The full process typically takes three to six weeks, though team urgency and level vary.

Salary bands for PM roles in India broadly sit at 12-20 LPA for associate PMs, 24-40 LPA for PMs with three to six years of experience, 40-60 LPA at the senior PM level, and 55-90+ LPA for group or principal PMs, based on industry surveys and platforms like Glassdoor.

Glance interviews tend to concentrate on three themes: product sense for a mobile-first, content-heavy platform; data and metrics thinking; and how you handle stakeholder complexity, given that telco partners, content creators, advertisers, and users all share the same surface.

1. How would you redesign the Glance lock screen feed to improve daily active engagement without increasing battery drain or data usage?
2. Glance serves users across multiple Indian languages. How do you decide which language to prioritise for a new content feature?
3. Walk us through how you would define and track the North Star metric for Glance Games.
4. A major telco partner wants to preload a customised content skin that conflicts with your personalisation engine. How do you evaluate and handle this request as a PM?
5. How would you solve the cold-start problem for a brand-new user who has never interacted with Glance?
6. A key content engagement metric drops sharply over two consecutive weeks with no known product release. How do you diagnose the cause and decide on a response?
7. How would you think about introducing a paid or premium content tier on Glance without damaging the free user experience?
8. Glance is considering deeper expansion into Southeast Asia. What product changes would you prioritise for an Indonesian audience versus an Indian one?
9. Describe a time you had to shut down or deprioritise a feature you had personally championed. What did you learn?
10. How do you balance the interests of content creators, advertisers, and end users when all three groups compete for the same lock screen real estate?
11. Short-video apps and OEM home screens compete for the same user attention. How would you sharpen Glance's product differentiation?
12. How would you build a roadmap for a new AI-powered content recommendation engine, from discovery to launch?

Q: How would you solve the cold-start problem for a brand-new Glance user?

*Situation:* At my previous company, we ran a short-form video feed and noticed that users who had no interaction history in their first few sessions rarely returned after the first week.

*Task:* I was asked to reduce early drop-off by improving the relevance of what new users saw before any engagement data was collected.

*Action:* I split the cold-start gap into two cases: fresh installs with no signal, and users with some browsing but no taps. For the first group, I worked with data science to use device language, SIM region, and time-of-day as proxy signals, surfacing region-popular content rather than generic trending items. For the second group, I introduced a lightweight onboarding prompt of three questions at most, feeding explicit interest tags into the ranker. I ran an A/B test across both approaches simultaneously.

*Result:* The region-proxy approach improved second-session retention per our internal OKR tracking. The onboarding prompt performed well for users who completed it, but completion rates were low, confirming that added friction on a passive surface hurts more than it helps.

---

Q: A key engagement metric drops sharply over two weeks. Walk us through your diagnosis.

*Situation:* At a previous role, the daily active rate on a content module declined across two consecutive weeks without any known product release driving the change.

*Task:* My job was to identify the root cause and recommend a response before the next sprint planning session.

*Action:* I broke the metric down by segment: platform version, OEM, geography, content category, and acquisition source. The drop was concentrated in one OEM variant following a system update that had changed how a lock screen permission was granted. I confirmed this with engineering through event logs, then designed a targeted re-permission nudge for the affected device segment.

*Result:* The nudge recovered engagement among the affected cohort within a week. More importantly, I added OEM firmware releases to our standard monitoring checklist, which caught two similar issues in the following quarter.

---

Q: Describe a time you had to kill a feature you personally championed.

*Situation:* I had pushed hard for a 'trending stories' carousel on a mobile content product, convinced it would drive exploration and increase session depth.

*Task:* After launch, I owned the call on whether to keep, iterate, or remove it.

*Action:* Three weeks post-launch, data showed users were scrolling past the carousel at a high rate. User research sessions surfaced consistent feedback that the home screen felt 'cluttered.' I ran a brief internal survey with a small sample and found users preferred deeper personalisation over editorial trending picks. I presented the data honestly to the team, acknowledged my original hypothesis was wrong, and recommended removing the carousel and redirecting that engineering capacity to the personalisation model.

*Result:* Home screen engagement recovered in the following two weeks. The team appreciated the transparency, and it became a reference point for how we handle failed bets without assigning blame.

CIRCLES for product design questions

CIRCLES (Comprehend, Identify, Report, Cut, List, Evaluate, Summarise) gives you a structured way to tackle 'design a product' or 'improve this feature' questions. For Glance, anchor your user definition tightly to the Indian mobile context. A daily commuter in a Tier 2 city who picks up their phone many times a day is a very different user from an urban professional who mostly encounters Glance passively.

North Star plus counter-metrics

For any Glance metrics question, name one North Star (for example, 'quality sessions per day'), then immediately pair it with a counter-metric (for example, 'accidental tap rate' or 'content dismissal rate'). Interviewers at content platforms want to see that you understand how engagement metrics can be gamed or inflated if left unguarded.

The three-stakeholder lens

Almost every Glance PM question involves at least three parties: the end user, the advertiser or content partner, and the telco or OEM. When answering stakeholder questions, name all three explicitly and show how you weigh trade-offs between them. A common ordering candidates use is: user trust first, partner value second, revenue third. Be prepared to defend your prioritisation.

Jobs To Be Done (JTBD) for the lock screen

The lock screen is not a destination product. Users are not choosing Glance; they are picking up their phone for a different job (check a message, see the time, open an app). JTBD framing helps you articulate why content relevance and zero-friction dismissal matter more on the lock screen than anywhere else in the product.

RICE for prioritisation questions

For roadmap and prioritisation questions, use RICE (Reach, Impact, Confidence, Effort) or ICE (Impact, Confidence, Ease). Show your scoring out loud. Glance interviewers tend to push back on impact estimates, so be ready to explain how you would measure impact before committing to build.

Mobile-first, India-first product instinct

Glance operates in a market where users may have low-end devices, intermittent connectivity, and consume content primarily in Hindi or a regional language. Interviewers want to see that you treat these as default constraints, not edge cases to handle later.

Comfort with content and personalisation trade-offs

Lock screen real estate is extremely limited and the interaction context is passive. PMs who think about engagement only through clicks and session length often miss the nuance that irrelevant content on the lock screen actively erodes user trust. Show that you treat relevance and trust as first-class metrics alongside activity numbers.

Data fluency without data dependency

Glance interviewers typically want PMs who can move without perfect data, especially in new markets or new feature areas. Demonstrate that you can form a clear hypothesis, define a minimum viable measurement, and make a call rather than waiting for full statistical certainty before every decision.

Cross-functional clarity

Glance teams work closely with data science, content operations, and partner (OEM and telco) teams. Interviewers look for candidates who can translate technical constraints for non-technical partners and commercial trade-offs for engineers, without losing either audience in the process.

A bias toward surface simplicity

The lock screen cannot be overloaded. Candidates who propose complex feature sets without considering cognitive load, battery cost, or dismissal rates tend to score lower. Show that you know when to cut, not just what to add.

Week 1: Understand the product deeply

Spend time using Glance daily on an Android device. Note what content surfaces in your feed, how quickly it loads, and what happens when you dismiss a card. Read publicly available interviews with Glance and InMobi leadership about their content strategy and approach to the Indian market. Check their official blog and investor-facing materials for stated product priorities.

Week 2: Build your metrics vocabulary

For every Glance surface you can identify (lock screen feed, games, sports scores, shopping), write down three metrics to measure health: one engagement metric, one quality or trust metric, and one business metric. Practice explaining why you chose each one and what the counter-metric would be.

Week 3: Practice product cases

Work through at least five product design or metrics questions using the CIRCLES or JTBD frameworks described in the frameworks section. Time each answer. Review your recording for vague claims, skipped trade-offs, and unstated assumptions.

Week 4: Prepare your stories

Write down five to seven product experiences from your career covering: a feature you built and shipped, a metric you meaningfully moved, a stakeholder conflict you resolved, a feature you shut down, and a time you changed your mind based on data. Practice each in STAR format (Situation, Task, Action, Result) and aim to deliver each one concisely without trailing off.

Before the interview

Review recent news about Glance and InMobi from 2024 to 2026. Prepare two to three questions for your interviewer about current team challenges or roadmap priorities. Confirm the format of your round with the recruiter so you can pace yourself appropriately.

Treating the lock screen like an app

The most common mistake candidates make is designing Glance features as though users open the product intentionally. The lock screen is an interruptive surface. Proposals that add navigation, multi-step flows, or dense information layouts tend to signal a lack of platform understanding to interviewers.

Ignoring India-specific constraints

Candidates trained on global product frameworks often propose solutions that assume fast internet, high-end devices, and English-primary content. At Glance, the opposite is frequently true for a large share of the user base. Failing to acknowledge data costs, device diversity, or regional language preferences is a quick way to lose credibility in the room.

Metrics without counter-metrics

Saying 'I would track daily active users' without pairing it with a quality or trust counter-metric signals shallow metrics thinking. Always name what you are optimising for and what you are protecting against. This is especially critical on a passive surface where engagement can be inflated by accidental interactions.

Over-indexing on features, under-indexing on distribution

Glance's reach is built through OEM and telco partnerships, not an app store listing. Candidates who propose features without considering how they interact with the partner relationship or device pre-load constraints miss a fundamental part of Glance's operating model.

Vague answers to stakeholder conflict questions

For questions about handling conflicting stakeholder interests, saying 'I would align everyone' without naming the specific parties, the trade-off being made, and how you would communicate the decision tends to score poorly. Interviewers want concrete detail about who you talked to, what you said, and what you traded away.