Senior product designer. Monetization and growth at Dainik Bhaskar. I design the moment people decide to pay.

Anirudh Sharma • Peerlist

The problemAndroid 13 made notification permission opt-in. Overnight, the app's primary traffic channel depended on readers actively saying yes, and our opt-in rate collapsed.Why the standard playbook failedBest practice says ask during onboarding, while the user is fresh and moving through setup. We were doing exactly that, correctly, and it was not working.The reason is what the question actually asks. At onboarding, a permission dialog asks someone to accept an abstract risk of being interrupted forever, from an app they have used for ninety seconds. Most people say no, and on Android that answer is expensive to reverse.What I didI broke the rule and added friction on purpose.<ul><li>A rehearsal before the ask. Readers saw a real news alert exactly as it would appear on their lock screen, then were asked for permission. The decision changed from an abstract risk to a concrete judgement about something they had just seen.</li><li>News-triggered re-permission for existing users. Rather than nagging on app open, the prompt appeared alongside a genuinely major story, when the value of an alert was self-evident.</li></ul>Results<ul><li>70% opt-in at onboarding, after the collapse</li><li>The app's main traffic channel survived the platform change</li></ul>What I took from itBest practices are someone else's context, generalised. Now when I follow one, I want to know what it was protecting. Asking honestly outperformed asking earlier.

Notification permissions after Android 13

Why it existedNobody asked me to build this. My job was subscription and monetization.But teams kept rebuilding the same components slightly differently, and it was worse than ordinary inconsistency because we publish in Devanagari and Gujarati. Those scripts have different line heights and character widths, so a component that was technically consistent across both still looked wrong in one of them. Every designer was solving that problem alone, under deadline.What I built<ul><li>Tokens for colour, type and spacing, with light and dark themes</li><li>Type scales per script, so Devanagari and Gujarati each render correctly rather than sharing one compromise</li><li>Component variants covering the real states teams needed, not the ideal set</li><li>Documentation written for people shipping that week</li></ul>The biggest single win was collapsing roughly ten article-card variants into one component.How I rolled it outI did not mandate it. I went team to team and helped people adopt it, because a system people are forced into gets worked around.I also set one governance rule: a component has to ship in a real product before it enters the system. No designing in the abstract.Results<ul><li>Teams stopped rebuilding basics</li><li>Handoff time and UI errors both dropped</li><li>Our products started looking like one product, which for a news brand is trust</li></ul>What I took from itConsistency is a tool, not the goal. A component that follows every rule and still feels wrong in one script is not consistent, it is just uniform.

DB Design Language System

The problemEvery previous attempt at surfacing new features or secondary content in the app landed at 1 to 2% CTR. Readers came for the news, scrolled the feed, and left. Anything that looked like a promotional slot was ignored.The insightThe failures had one thing in common. They looked like ads. So instead of asking for attention, I borrowed formats readers already trusted and used constantly.What I did<ul><li>Story rings at the top of the feed, using a pattern readers already knew from social apps, for time-sensitive and short-form content</li><li>Native-format feed widgets built to sit in the reading flow rather than interrupt it, matching the article cards around them in type, spacing and behaviour</li></ul>Results<ul><li>9% CTR on story rings</li><li>15 to 25% CTR on native-format widgets</li><li>Now the default launch surface for new features across the company</li></ul>What I took from itAttention is not won by standing out. It is won by earning a place in a pattern people already trust.Full case study at <a target="_blank" rel="noopener noreferrer nofollow ugc" class="break-all link-break text-gray-500 hover:underline" href="http://voleshads.com">voleshads.com</a>

Discovery system — Dainik Bhaskar

The problemDainik Bhaskar reaches millions of daily readers, mostly in tier-2 and tier-3 cities on mid-range Android phones. The brand was large and trusted, but subscription conversion sat well below potential. Analytics told us it was failing and never why.The researchI ran a field study in Bhopal and sat with real readers. Most of them did not know paid plans existed at all. It was an awareness problem, not a willingness-to-pay problem.The lock screen was the clearest example. It told readers they had run out of free articles and should come back later. A dead end where the highest intent in the product was sitting.What I did<ul><li>Turned the lock screen into a storefront. Two plans with prices, an anchor of less than ₹1 a day, a pre-selected default, one call to action. Two plans, not five, because more choice meant more friction for this audience.</li><li>Moved plans into the home feed through a membership banner and a plan widget, so pricing was visible during normal browsing rather than only at the wall.</li><li>Shipped a ₹1 first-month trial with the following month's price stated plainly.</li><li>Compressed checkout onto the paywall itself. Select a plan, pick UPI or PhonePe, pay without leaving the article.</li></ul>The tradeoffsEditorial owns the home feed. Rather than argue for territory, I designed the widget so editorial controlled placement and presentation, which turned a territory fight into a scheduling decision.I took the paywall concept to the CEO before detailed design, which got leadership alignment early instead of at review.I kept the design restrained on purpose. Two plans, one anchor, one action. Adding benefits and tiers would have felt more generous and converted worse.Results<ul><li>24.7% lift in plan-page views</li><li>9.8% of those additional readers converted to paid</li><li>19.5% further lift in the first quarter after intro offers</li></ul>What I would do differentlySet up baseline instrumentation and attribution before designing, not alongside it. And design the renewal decision from the start, since the subscription lifecycle does not end at purchase.