Fixing Trust and Transparency Issues in Yandex Go Rides

Phase 01 · App teardown

✳The problem they went after

Yandex Go bundles ride-hailing, food delivery and groceries for urban commuters across Russia and the CIS. This write-up stays inside the ride-booking flow, where the friction lives: drivers marking themselves arrived before reaching the pickup spot and starting paid wait time, GPS pins that miss the real pickup and drop-off points, no final price once a stop is added mid-ride, and drivers overriding the payment method the rider picked.

✳What they actually did

The teardown gives credit before it complains, naming car availability and pricing shown before you commit, live driver ETA tracking, and support for multiple payment methods. It then isolates the friction and reverse-engineers three features, GPS auto-detection, most-used destinations and the closest-class selector, tracing each from rider convenience through to business outcomes like reduced activation time and higher transaction volume. The problem hypotheses land on unfair wait-time charges and navigation accuracy, backed by personal use and user reviews.

✳What you can take from it

Worth stealing

Three features traced from user value to business outcome
Strengths listed before friction, keeping the critique credible
Premature wait-time activation named as a trust break
Hypotheses grounded in user reviews, not just opinion

Read this if

Writing your first app teardown and unsure how to structure it

The transferable bitReverse-engineering a feature all the way to the business outcome is the transferable move here. Listing what works before listing what breaks keeps the critique credible rather than reading as a complaint log. And the hypotheses that come out at the end are narrow enough to actually test, which is what a teardown is for.

✳The submission itself

Open the PDF →

Reading on a phone? The embedded viewer is cramped — open the PDF in a new tab instead.

Phase 02 · User interviews

Fixing Comfort Class Experience Gaps in Yandex Taxi

✳The problem they went after

Riders in Tashkent pay the comfort class premium expecting a better car and a better driver, and keep receiving vehicles and driver behaviour associated with standard class. Interviews put specifics on it: air conditioning, clean interiors and professional drivers are what people expect from the tier and none of it arrives reliably. When a paid tier stops holding its promise, the whole class system loses meaning.

✳What they actually did

The interviews surface more than the headline gap. Drivers coming in from out-of-town regions do not know the city and follow app-suggested routes blindly, which makes rides longer and irritates people who know a better way. Waiting mode sometimes gets switched on before arrival, so charges start early. Cash payment trouble, drivers without change, and difficulty reaching support push riders toward card payments and, past a point, toward competitor apps.

✳What you can take from it

Worth stealing

Interviews expand one complaint into a cluster of gaps
Tier promise tested against what riders actually receive
Routing failure traced to out-of-town driver unfamiliarity
Payment friction linked directly to competitor switching

Read this if

Running discovery interviews and wondering how far to follow a tangent

The transferable bitA clean example of interviews widening a problem instead of confirming it. The starting complaint was tier quality; what came back was a cluster covering service standards, routing intelligence and payment. Each strand points at a different owner inside the company, which is exactly the map you want before deciding what to fix first.

✳The submission itself

Open the PDF →

Reading on a phone? The embedded viewer is cramped — open the PDF in a new tab instead.

Phase 03 · Solution prioritisation

Using Gamification to Fix Driver Behavior on Yandex Taxi

✳The problem they went after

Drivers activate paid wait time before reaching the correct pickup location, riders get billed for time the driver was not there, and trust goes with it. Root-cause analysis and issue tree mapping pin the causes on driver misconduct, app performance limits on low-network devices, and the absence of any real accountability for repeat offenders.

✳What they actually did

Thirteen solutions were generated and sorted into confidence tiers, spanning a gamified leaderboard, milestone badge rewards, intelligent pickup spot tracking, app optimisation for low-network conditions, and temporary access blocking for repeat rule-breakers. RICE put the leaderboard on top with a score of 13.5, followed by a learning progress tracker and the access blocking mechanism. All three were wireframed as interactive prototypes: the leaderboard links ratings to reduced commissions and bonus earnings, the tracker teaches platform rules before drivers hit edge cases, and blocking supplies the consequence.

✳What you can take from it

Worth stealing

Issue tree separates misconduct, app performance and accountability
Thirteen ideas sorted into confidence tiers
RICE picks the leaderboard at 13.5 as top initiative
Winners split across incentive, education and enforcement

Read this if

Learning to run brainstorm-to-prioritisation without the shortlist looking arbitrary

The transferable bitThe three winners attack the same problem from three angles: incentive, education and enforcement. That combination is far more defensible than three variations on one idea, and it maps back one-to-one onto the root causes the issue tree produced, so the shortlist never looks arbitrary.

✳The submission itself

Open the PDF →

Reading on a phone? The embedded viewer is cramped — open the PDF in a new tab instead.

Phase 04 · The PRD

Using Driver Leaderboards to Fix Yandex Taxi Wait-Time Charges

✳The problem they went after

Riders on Yandex Taxi keep getting billed for wait time that started before the driver actually reached the pickup location. It reads as a small charge on the receipt and behaves like a trust problem in the account: people move to competitors with more transparent pricing, and rider retention metrics take the hit directly.

✳What they actually did

The system gives drivers points and levels for good reviews, low cancellations and following platform guidelines, with higher placement unlocking better bonuses so good behaviour pays. Repeat ride rate is the primary metric, supported by average rides per active user and an NPS-based rider satisfaction score. The pitfalls section is the substantial part: full-time drivers gaming the board, and route cherry-picking to chase shorter rides. Mitigations cover individual performance bonuses, point multipliers for long-distance rides, and suspension criteria for drivers with repeated complaints.

✳What you can take from it

Worth stealing

Repeat ride rate chosen as the primary metric
Gaming risks documented before launch, not after
Point multipliers counter short-route cherry-picking
Bonus earnings tied to leaderboard placement

Read this if

Writing a PRD for an incentive or gamification feature

The transferable bitAny incentive system produces behaviour nobody asked for, and this PRD writes down the two most likely distortions before launch, then attaches a specific mitigation to each. Doing that work upfront is a large part of what separates a PRD from a feature pitch, and the pitfalls section alone is worth opening.