Explore Problems

A Chase customer describes being passed between vaguely-named internal teams like escalation, research, and the Executive Office without resolution, eventually filing formal complaints with the OCC and CFPB. The account illustrates how large banks internal escalation structures can function as a maze that absorbs complaints without producing outcomes.

A long-time T-Mobile customer describes being left on hold for an hour with no answer, alongside a name misspelling on the account that has gone uncorrected despite five separate requests. The combination points to both support responsiveness gaps and unreliable account data correction processes.

A founder describes their landing page receiving 4.5 million views after an investor shared a screenshot, yet the surge converted only a single paying subscriber. It highlights the gap between attention and virality versus actual monetized conversion for early-stage products.

Developers who expose an MCP server have no way to see the full agent session behind incoming tool calls -- what the user originally asked, what the agent reasoned, or where it got stuck -- because that conversation lives entirely inside the end user's AI client. This makes it nearly impossible to know which use cases are popular or where users are frustrated.

People who make consequential calls, founders, parents, planners, investors, have no easy way to log what they predicted and how confident they were before an outcome resolves; by the time the result is known, hindsight has rewritten the memory of the original uncertainty, making it impossible to later see where their judgment was actually sharp or overconfident.

A customer needed a chandelier installed at a tall ceiling height, but two different contractors dispatched through the marketplace each arrived with only a 10-foot ladder when 15 feet was needed, despite the requirement being communicated in advance. After repeated failed visits and lost work time, the customer returned the fixture for a refund but was still charged $250 for an installation that never happened.

GEICO's DriveEasy telematics app reportedly misidentifies driving as walking and flags routine downhill braking as hard braking, using force thresholds far below what the driver's own in-vehicle accelerometer registers as normal. Since these readings feed directly into insurance discount eligibility and premiums, inaccurate sensor interpretation can unfairly penalize safe drivers.

Developers running many concurrent Claude/Codex sessions for both engineering and go-to-market work struggle to keep track of what's finished, what's abandoned, and how to get back into old sessions. The best current workaround is asking the assistant itself to search for a past conversation, which is unreliable and unstructured. This points to a missing session-management layer for people who work across many parallel AI agent threads.

A note-taking app user was frustrated they could not access their notes without an internet connection while out shopping, and separately dislikes AI features being pushed into the search experience when they just want to search their own notes. This reflects a broader tension between offline reliability and unwanted AI-first redesigns in note-taking tools.

A customer switching a family plan from Verizon to T-Mobile spent over six hours across three days dealing with repeated agent handoffs and phone-number porting mistakes, including two numbers ported to the same device. The account illustrates how fragile and error-prone carrier onboarding and number-porting workflows can be for customers switching providers.

A renter documented with photos that a rental truck's actual return odometer reading did not match the mileage recorded on the billing receipt, resulting in a $59.54 overcharge. This highlights a billing-integrity gap where customers have no independent way to verify vehicle usage records against rental charges.

Investors buying property outside their home state need a reliable general contractor to perform walkthroughs and assessments on their behalf, since they cannot inspect the property in person. Finding a trustworthy, vetted contractor willing to do one-off remote walkthroughs is a recurring challenge with no standard sourcing channel. This gap creates risk of relying on unverified referrals or skipping proper due diligence altogether.

Property rehabbers must decide whether to reveal their target renovation budget to contractors before bidding, fearing that disclosure invites bids anchored to the budget rather than actual project cost. The high engagement around this question signals a persistent trust and pricing-transparency gap between investors and contractors.

Real estate investors commonly target tax-delinquent homeowners as motivated sellers, but delinquency status frequently does not correlate with actual willingness to sell, producing a high rate of false-positive leads. This wastes marketing spend and outreach time across the direct-mail and cold-calling lead generation workflows investors rely on.

Families in Kerala seeking postnatal care struggle to find reliable, comparable information on centers, since pricing, room types, amenities, and reviews are scattered across sources. This makes choosing a trustworthy provider slow and uncertain during an already stressful period, leaving families reliant on word-of-mouth or incomplete listings.

Makers who launch products on discovery platforms like Product Hunt get a single day of traffic before their listing disappears into an unbrowsed archive. Weeks of launch preparation yield only short-lived visibility, leaving no lasting channel for continued discovery or backlink value.

Developers building on LLMs typically judge prompt quality by manual spot-checking rather than running it against a structured test dataset, leaving them without a pass rate, per-case failure reasoning, or a way to catch regressions when a prompt is edited. This makes prompt iteration largely guesswork instead of a measurable, repeatable process.

Fans of women's professional sports leagues like the WNBA, NWSL, and PWHL must switch between multiple apps to follow scores, schedules, and standings, and often miss games because coverage is scattered or absent from mainstream sports apps. This fragmentation makes it harder to stay engaged with the growing women's sports landscape.

A consumer describes a marketplace-referred contractor abandoning a job mid-demolition after being denied permits, leaving property damaged, while the platform's only offered remedy - a personal settlement payment - would jeopardize the user's need-based income eligibility. This highlights a gap in how service marketplaces handle contractor-caused damage and financial liability for vulnerable users.

Users of AI-powered support chatbots report the AI sometimes repeats itself or struggles to steer conversations back to the relevant topic, requiring better prompt engineering to get useful answers. This points to a broader reliability gap in conversational steering for AI customer support agents.