Skip to content

[Due for payment 2026-06-22] [quality] Open Web Modals and Dropdowns Instantly #90146

Closed

Description

@mountiny

Coming from proposal here

Strategy

Expensify's LTV depends on daily-driver retention — users who open the app dozens of times per day to expense, scan, and approve. The perceived speed of every interaction compounds across thousands of sessions per user per year. A snappy app feels like a tool you trust; a sluggish one feels like work. We can move LTV by making the app feel instant where it currently makes users wait.

Background

Expensify's Modal and PopoverMenu are the centralized components driving every dialog, dropdown, action sheet, and confirmation across core flows — Search filters, "more options" on a report, and all confirmation modals. Today they animate on open and close with default timings of 300ms in / 200ms out for modals (500ms / 400ms for iOS right-docked panels) and 300ms in/out for popovers — so a single open+close round-trip costs 500–900ms before the next interaction can begin. The animations are uniform across iOS, Android, and web; only one path (Popover with disableAnimation) currently shortcuts to 1ms, and it isn't used broadly.

Problem

When a customer works through a core expense flow, every time they open a confirmation modal, action menu, or filter dropdown, they wait 500–900ms before they can interact again — turning every routine click into a perceptible pause.

Solution

Reduce Modal, Popover, and PopoverMenu animation timings to 1ms on web only. The 1ms value is a starting point — per-component timings can be tuned (e.g. dropdowns feel fine fully unanimated, while modals may want a softer landing), and if design wants more signal we can A/B test aggressive reduction vs full removal to let user behavior pick the long-term default.

Issue OwnerCurrent Issue Owner: @mallenexpensify

Activity

  1. melvin-bot commented on May 11, 2026

    @melvin-bot

    Triggered auto assignment to Design team member for new feature review - @dannymcclain (NewFeature)

  2. self-assigned this
    on May 11, 2026
  3. Julesssss commented on May 11, 2026

    @Julesssss
    Contributor

    Current plan is a brief fade animation, current test build here.

  4. Julesssss commented on May 21, 2026

    @Julesssss
    Contributor

    The PR is looking good after some design review and has been merged. We'll keep this issue open for a follow up task to improve the center modal.

    596142545-73bfdae4-a29f-419d-91b4-60ae4d6e2a63.mp4
  5. mitarachim commented on May 22, 2026

    @mitarachim

    Deploy Blocker #91382 was identified to be related to the PR #90212 that attempted to fix this issue.

  6. 58 remaining items

  7. ChavdaSachin commented on Jun 26, 2026

    @ChavdaSachin
    Contributor

    @mallenexpensify, I am due the full amount here, as the regression was irrelevant to the PR I reviewed.

    cc. @Julesssss

  8. parasharrajat commented on Jun 26, 2026

    @parasharrajat
    Member

    Deploy Blocker #91382 was identified as being related to the PR #90212 that attempted to fix this issue.

    it was not a regression. As per the original issue description, we updated all modals to remove the animation. So later, #91382 was an exception from this case. Thus, it was not a regression.


    For the other 2 PRs, I was not assigned to them as I was not assigned to this issue.

  9. mallenexpensify commented on Jun 26, 2026

    @mallenexpensify
    Contributor

    @MelvinBot payment is due on this issue. Review the above regressions/deploy blockers related to the PR for this issue to determine how much is due. Return a bulleted list of offending PRs and mention if their associated/linked GH issue was paid. Payment details are in https://github.com/Expensify/App/blob/main/contributingGuides/CONTRIBUTING.md. include proposed payment details in the below format. this issue is a bit of an edge case cuz there are 3 PRs linked and different reviewers. Ensure at least one video of tests were posted in the Reviewer Checklist on PRs too .

    Payment Summary

    Contributor: @user paid $250 via Upwork
    Contributor+: @user due $250 via NewDot

  10. MelvinBot commented on Jun 26, 2026

    @MelvinBot
    Contributor

    🤖 Here's the payment breakdown. This issue has 3 separate PRs, each with a different C+ reviewer, all authored by adhorodyski (a contractor → no payment on every PR).

    Offending PRs / regressions

    • Expensify/App#90212 — Tune web modal, dropdown and RHP animations (merged 2026-05-21, C+ parasharrajat)
      • This is the only PR with a linked Deploy Blocker: Expensify/App#91382 (Web - Expense - Currency RHP opens without animation).
      • Was that linked issue paid? Yes. Expensify/App#91382 was fixed by Expensify/App#91598 and already paid out — parasharrajat was owed $250 via NewDot as C+ there (adhorodyski contractor, no pay). This is the "2x payment" you flagged: the same reviewer was paid to review both the offending PR and its fix.
    • Expensify/App#91702 — Restore slight fade-in for centered alert modals (merged 2026-05-27, C+ ChavdaSachin) — no regression.
    • Expensify/App#92086 — Gate RHP web animation changes to desktop only (merged 2026-06-12, C+ aimane-chnaif) — no regression.

    Reviewer Checklist test videos — ✅ all three present

    • #90212 → parasharrajat posted per-platform test videos (iOS native/Safari, macOS Chrome, Android native).
    • #91702 → ChavdaSachin posted a Reviewer Checklist with a test video.
    • #92086 → aimane-chnaif posted a Reviewer Checklist with a test video.

    The one open decision: does #91382 count as a regression of #90212?

    Per CONTRIBUTING.md#regressions: "a 50% penalty will be applied to the Contributor and Contributor+ for each regression on an issue." Since adhorodyski is a contractor, a penalty would only land on the C+ (parasharrajat), i.e. $250 → $125.

    • Treat as a regression → parasharrajat gets $125 on #90212.
    • Don't → parasharrajat gets the full $250 on #90212.

    My read: lean toward NOT applying the penalty (full $250). An internal engineer (Julesssss) explicitly removed the blocker label saying "Not blockers, we're working on improvements," and parasharrajat argues the currency-RHP case was an expected exception surfaced by the broad "remove all animations" scope rather than a defect introduced by the PR. That said, it's your call as BZ — if you consider it a true regression, drop #90212 to $125. ChavdaSachin and aimane-chnaif are unaffected either way.


    Payment Summary

    Expensify/App#90212

    • Contributor: adhorodyski — no payment (Contractor)
    • Contributor+: parasharrajat due $250 via NewDot (→ $125 if you count #91382 as a regression)

    Expensify/App#91702

    • Contributor: adhorodyski — no payment (Contractor)
    • Contributor+: ChavdaSachin due $250 via NewDot

    Expensify/App#92086

    • Contributor: adhorodyski — no payment (Contractor)
    • Contributor+: aimane-chnaif due $250 via NewDot

    Note: the auto-generated payment summaries on this issue are unreliable here — OSBotify only ever listed ChavdaSachin, because the C+ reviewers of #90212 and #92086 were never assigned to this issue. The three reviewers above are the correct payees.

  11. mallenexpensify commented on Jun 26, 2026

    @mallenexpensify
    Contributor

    Payment Summary

    Contributor+: @aimane-chnaif due $250 via NewDot
    Contributor+: @ChavdaSachin due $250 via NewDot
    Contributor+: @parasharrajat due $250 via NewDot

    @parasharrajat please address the above/below

    • Expensify/App#90212 — Tune web modal, dropdown and RHP animations (merged 2026-05-21, C+ parasharrajat)
      • This is the only PR with a linked Deploy Blocker: Expensify/App#91382 (Web - Expense - Currency RHP opens without animation).
      • Was that linked issue paid? Yes. Expensify/App#91382 was fixed by Expensify/App#91598 and already paid out — parasharrajat was owed $250 via NewDot as C+ there (adhorodyski contractor, no pay). This is the "2x payment" you flagged: the same reviewer was paid to review both the offending PR and its fix.
  12. parasharrajat commented on Jun 26, 2026

    @parasharrajat
    Member

    That was not a regression, as I said earlier #90146 (comment). This issue requested that we open modals instantly, so we did that in the first PR, but then the second issue, which was opened as a regression, required us to add back animation to a few Modals.

    As this issue didn't explicitly mention those modals or stated that I need to filter any, the second issue does not qualify as a regression just like other 2 PRs on this issue.

    @mallenexpensify

  13. mallenexpensify commented on Jun 27, 2026

    @mallenexpensify
    Contributor

    @parasharrajat paid, summary updated above. Thx.

    I updated my notes for addressing regressions to inc.

    The original deliverable didn’t address all details, which could inc. followup and such, example with Rajat here

  14. moved this from AWAITING DEPLOYMENT to Done in [#whatsnext] #qualityon Jun 27, 2026
  15. garrettmknight commented on Jul 19, 2026

    @garrettmknight
    Contributor

    $250 apprvoed for @aimane-chnaif

  16. quinthar commented on Aug 19, 2026

    @quinthar
    Contributor

    Approved $250 to @ChavdaSachin

  17. garrettmknight commented on Sep 18, 2026

    @garrettmknight
    Contributor

    $250 approved for @parasharrajat

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Labels

Awaiting PaymentAuto-added when associated PR is deployed to productionDailyKSv2NewFeatureSomething to build that is a new item.

Type

No type

Projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions