Commissioned research · 2026
Real Sports
Understanding friction through public evidence.
Contribution: research synthesis and product strategy recommendations. This commissioned product research produced a research handoff; shipped improvements and production outcomes are unmeasured.
The research question
Where does understanding break down?
Finding the moments when people struggle to understand Real Sports or complete a task—and turning those signals into testable product priorities.
- Public records
- 418
- Issue groups
- 8
- Research window
- May 10–August 8, 2026
01 / Corpus & method
Public evidence, carefully bounded.
Public posts, comments, and app-store reviews were collected, screened, deduplicated, and grouped with keyword rules. The collection script targets 418 records. This is a purposive research corpus, not a random sample.
Figures retain the original corpus coding. Labels can include positive reviews and false matches; they describe how records were grouped, not confirmed defects or unique people.
| Source | Records | Share |
|---|---|---|
| Reddit post | 257 | 61.5% |
| Reddit comment | 79 | 18.9% |
| Apple App Store | 66 | 15.8% |
| Google Play | 16 | 3.8% |
02 / Original corpus coding
Two groups hold 67.2% of the records.
Cards & marketplace and onboarding & clarity account for 281 records. Volume helps frame follow-up questions; it does not establish severity.
| Theme | Records | Share |
|---|---|---|
| Cards & marketplace | 147 | 35.2% |
| Onboarding & clarity | 134 | 32.1% |
| Monetization & access | 39 | 9.3% |
| Coverage & feature gaps | 33 | 7.9% |
| Community & moderation | 33 | 7.9% |
| Performance & data reliability | 19 | 4.5% |
| Notifications & control | 7 | 1.7% |
| Platform parity & updates | 6 | 1.4% |
03 / Findings to investigate
From a signal to a question.
Make the next step visible
234 task / how-to records · 56.0% of corpus
“Is there a way to tell how much base rax a team card has earned?”Reddit post · 2026-08-07 ↗
People ask the community to locate information needed for a specific task. This suggests a discoverability problem worth testing, rather than a lack of interest.
Proposed response. Test task-level help and clearer routes to card earnings, balances, and history.
Explain marketplace rules in context
147 marketplace · 39 access records
“All cards that aren’t from player packs will go to auction if offered on within 24 hours of opening.”Reddit comment · 2026-08-03 ↗
A peer explains a hidden condition. The account is evidence of explanation work in the community, not independent verification of the product rule.
Proposed response. Show eligibility, fees, and the result of an offer before someone commits.
Make trust observable
19 reliability · 33 moderation records
“not updating the game scores times and player stats”Google Play · 2026-06-28 ↗
A positive review also describes stale information. Reliability and moderation records warrant separate follow-up: source labels alone do not establish failure rates.
Proposed response. Test freshness indicators and recovery states; evaluate report status and escalation in a separate moderation task.
04 / Product priorities
Five starting points.
Recommendations from the original handoff. These are hypotheses for validation, not shipped outcomes.
| Priority | Proposed scope |
|---|---|
| 01Sport-first information architecture | Home, Search, Watchlist, Community, You. Keep Cards and Marketplace contextual. |
| 02Global AI search — Ask Real | One global input returning sourced answers, entities, games, plays, cards, and direct actions. |
| 03Contextual docs and help | Index current documentation in Ask Real and open help inside the active task. |
| 04Rax hub | Balance, history, earn forecast, next-best use, goals, and tested non-card utilities. |
| 05Guided Marketplace | Expose eligibility, Pro gates, fees, offer-to-auction behavior, transfer state, comps, and history. |
Eight further opportunities
Card portfolio intelligence
NextCost basis, value, Rax yield, upgrade path, comparable sales, and alerts.
Interactive sports graphics
TestPrototype one shared vector renderer for pitch, shot, drive, and formation views.
Personalized Home and Watchlist
NextFollow teams, players, games, and cards; rank live modules; save searches; respect alert settings.
Profile reorganization
NextPreserve personality; organize Activity, Stats, Cards, and Achievements under a clearer header.
Moderation and support
ProtectReport status, block and anti-alt controls, filters, appeals, and visible escalation.
Live-data confidence
ProtectFreshness timestamps, delayed-data states, correction history, and service status.
Coverage finder
NextSearch unsupported entities, request or vote, show status, and notify on launch.
Platform parity
LaterAvailability labels, parity-aware release notes, and tablet or web fallback.
05 / Limits & next validation
What this study can—and cannot—tell us.
Read the distribution with care.
Reddit contributes 336 of 418 records (80.4%). People who post publicly may differ from quieter users. Posts and replies can share context, and multiple records may come from one person.
The fixed collection target, relevance screening, and keyword coding shape the result. There is no population denominator, inter-rater reliability estimate, causal finding, or validated predictive model.
Validate the interpretation next.
Audit a stratified sample with a second coder, review false-positive theme matches, then test core tasks with new and experienced users. Observe completion, errors, and understanding before prioritizing implementation.
Revisit reliability and moderation separately with task-specific evidence. No uplift or production impact is claimed here.