Research typeQualitative user research
MethodSemi-structured interviews
Participants5 urban navigators
FieldworkJune 9–11, 2026
RoleResearcher (solo)

Google Maps serves over a billion people a day, yet riders and drivers alike still meet unreliable real-time data, cluttered screens, and missing road information. This study asks where, exactly, the experience breaks down for people who depend on it to get around their city — and what the evidence says we should do about it.

01 Research goals

At which points do urban users most frequently meet confusion, anxiety, or navigational error inside Google Maps?
  • Understand people’s real navigation habits and contexts.
  • Identify the moments that trigger negative emotion.
  • Surface the workarounds users invent to patch the product’s gaps.
  • Propose improvements backed by user evidence.

02 Method

I ran five ~30-minute semi-structured interviews, chosen over a survey because exploratory questions need genuine emotion, behavioural specifics, and motivation — not just multiple-choice. Each conversation walked through usage context, a recent navigation session step by step, pain points and feelings, and self-made workarounds.

IDLocationTime therePrimary mode
P1Canada2 yrsBus · subway · walking
P2Canada2 yrsBus · subway · walking
P3Canada7 yrsBus · subway · driving · cycling
P4Hong Kong28 yrsSubway · driving · walking
P5Canada6 yrsDriving (daily)

03 Findings

I extracted key quotes from all five interviews and grouped them through affinity mapping. Six themes recurred — one of them raised, independently, by three of five participants.

Four core pain points

01

Unreliable transit times push users to build their own buffer systems

3 / 5 · P1 · P2 · P3

Google Maps’ estimated bus arrival time is just not reliable — I always cross-check with the Transit app.

P1

I always show up 15 minutes early to account for the error. It’s just become a habit.

P1

Google Maps showed the bus had already left, but it was actually just running late.

P3

Impact. Users have internalised distrust of the transit data. Instead of relying on the app, they arrive significantly early or open a third-party app to verify — adding cognitive load and wasted time to every commute.

AnxietyUncertaintyDistrust
02

Driving navigation fails at complex intersections and highways

2 / 5 · both drivers · P3 · P5

In Toronto, with all those overpasses and on-ramps, I can’t tell which road is which on Google Maps. I keep taking wrong turns.

P3

Some intersections ban left turns at certain times, but Maps doesn’t warn you — it still tells you to turn. You only find out when you’re already there.

P5

Because it’s 2D, the directions for getting on and off highways are unclear. I get it wrong all the time — a 3D view for bridges and ramps would make a huge difference.

P5

Impact. Drivers can’t process 2D navigation fast enough through complex structures, leading to missed exits, wrong turns, and — with time-restricted turns — unsafe last-minute lane changes.

StressConfusionSafety concern
03

A cluttered interface buries the one thing the user needs

2 / 5 · P1 · P3

The direction indicator for the bus isn’t prominent enough. The interface feels messy — I can’t grab the key info at a glance.

P1

Big Toronto stations have many stops, and buses sometimes share a platform with streetcars. Maps doesn’t show this clearly — first-time visitors get lost.

P3

Impact. At multi-line hubs users can’t quickly tell where to wait, raising the risk of missing their vehicle — especially for people new to the local transit system.

ConfusionDisorientationFrustration
04

Stale business info and missing closure alerts erode trust

3 / 5 · P2 · P4 · P5

Restaurant hours are often seriously wrong. It relies entirely on community contributions, and small businesses often aren’t listed at all.

P2

The most common issue for me is store hours that have changed but haven’t been updated.

P4

Road closures and construction zones just aren’t flagged.

P5

Impact. Users arrive to find places closed, or hit unannounced road closures mid-route. These failures accumulate and quietly undermine trust in the whole platform.

DisappointmentDistrust

Users are already patching the product themselves

A consistent pattern cut across every interview: people have quietly built multi-app habits to cover the gaps.

ScenarioWhere Maps falls shortThe workaround
Real-time transit arrivalInaccurate predictionsSwitch to the Transit appP1 · P3
Driving in complex areas2D interface too confusingSwitch to Apple MapsP3
Business detailsOutdated informationCheck Google Reviews separatelyP1
Low visual clarity while driving2D hard to followTurn on voice guidanceP5

Users aren’t abandoning Google Maps — they’re patching it. Every workaround marks a real, unmet need, and that makes them the highest-priority places to improve.

04 Recommendations

R1Pain point 01

Make transit-data reliability transparent

Label each arrival time by source — a green “Live” vs. a grey “Estimated” indicator — and proactively tell users when live data is unavailable instead of silently showing a guess. Shift the experience from false confidence to informed uncertainty.

Measure Time in transit view · rate of switching to third-party apps

R2Pain point 02

Add a 3D / AR view for complex driving

Auto-trigger a 3D or AR view at complex nodes — ramps, overpasses, restricted intersections. Warn about rules like time-restricted left turns with audio + visual cues before the intersection, not at it.

Measure Route-deviation rate · frequency of mid-route replanning

R3Pain point 03

Fix the information hierarchy at transit hubs

Surface the user’s specific boarding location — not just the station name. Give the next action, time remaining, and exact platform far more visual weight than secondary detail. (Citymapper’s “where to stand” is a good model.)

Measure Completion rate at multi-platform stations · mid-journey exits

R4Pain point 04

Show information freshness, widen real-time coverage

Flag listings that haven’t been updated with a “may be outdated” label, integrate municipal road-closure data, and make community corrections one tap — with an incentive to keep data fresh.

Measure Business-info accuracy · closure-alert coverage · correction rate

05 Limitations & next steps

Five participants give exploratory, not statistically representative, findings, and retrospective interviews carry recall bias; the recommendations are not yet validated. Next I’d widen the sample across more cities and commute patterns, run usability testing to watch real navigation behaviour, and prototype and test the proposed changes.

06 Reflection

Underneath “the app has bugs” is really a trust problem. When people can’t tell whether a piece of information is reliable, they compensate — arriving early, switching apps, turning on voice — rather than relying on the product. Good design isn’t only about showing information; it’s about managing expectations and earning confidence.

The most revealing moments in every interview were the workarounds. Users don’t always articulate what’s broken — but their behaviour tells you exactly where the product is failing.

Independent portfolio research · Jingyan Jiang · June 2026