Razorpay

Public discussion around Razorpay over the past four weeks carried a predominantly negative tone, with complaint-oriented mentions outweighing praise by a wide margin. Bug reports were the single most prominent thread running through conversations, appearing in roughly half of all mentions, and reliability concerns followed closely behind. Together these two themes shaped the bulk of discourse, leaving commenters with an overall impression of a product they felt was technically unstable in real-world deployments. Specific frustrations cited in sample mentions included intermittent 500 errors on payment callback endpoints and a perceived lack of separation between test and live modes in admin settings, both of which drew pointed criticism.

On the sentiment trajectory, discussion opened the window at a noticeably higher tone than it would sustain. The score dropped sharply in the second week as mention volume more than doubled, suggesting that a surge of new voices entering the conversation skewed negative. The weeks that followed showed the score oscillating in a narrow band without recovering to its earlier level, pointing to a sentiment plateau rather than a recovery. The final recorded week saw another slight dip alongside shrinking mention volume, which some observers might read as disengagement rather than resolution.

Praise did surface, most consistently around integration quality, where a cluster of commenters expressed satisfaction with Razorpay fitting into existing development workflows. Feature requests and UPI-related integration asks also appeared, reflecting genuine enthusiasm for expanding the product rather than abandoning it. However, this positive thread was consistently outnumbered.

Opinion was divided most visibly on integration breadth. While some commenters praised the integrations that existed, a separate and roughly equally sized group flagged gaps, particularly around payout methods available to Indian developers. Chargeback handling and ledger reconciliation also surfaced as contested areas, with some treating them as missing features and others framing them as reliability failures, suggesting no clear community consensus on how to characterize those shortcomings.