FA2 dashboard — UX/UI audit

Headline metrics read as body text, not KPIs

The strip $1.6B+ raised · 59 teams  |  $528.2B+ AUM/TVL · 23 teams  |  8.9M followers · 64 verified  |  74 active repos  |  18.2 avg score · 19 top tier compresses six high-value figures into one line of small uppercase-grey text. These are arguably the single most important takeaways on the page (total deployed capital, total AUM, audience reach, ecosystem health). They deserve the same .stats > .stat > .big-number treatment used elsewhere in this report’s style guide.

Recommendation. Promote to a 4-tile or 6-tile KPI strip with large-format numbers (28-32px) and a one-word label, sitting directly below the category tab bar. The current strip is data-correct but visually deflating; first-time visitors miss the scale of what they’re looking at.

The filter bar carries 25+ controls in one block

Below the metrics strip, the filter region holds: 6 sort controls (A–Z, Funded, Warnings, Followers, GitHub, Score) + a CSV link + a grid/table toggle + 7 tag pills + 3 GitHub-activity pills + 3 score-tier pills + 2 funding/twitter toggles + 1 warnings toggle + 4 opportunity-space pills. Twenty-seven interactive elements before the first card row. The grouping labels (SORT, GITHUB, SCORE, FUNDING, TWITTER, WARNINGS, OPPORTUNITY SPACE) help, but the visual weight of every chip is identical. Cognitive load on first arrival is high.

Recommendation (low effort). Collapse anything beyond Sort + the 7 FA2-tag pills behind a single + Filters (5) chip that opens a popover or expands inline. Show active-filter count next to the chip. Keep the most-used filters (sort + tag) always visible.

Recommendation (higher effort). Inspect dashboard analytics: which filters are actually used? Suspect Sort and tag pills account for >90% of clicks; the rest could become an opt-in “advanced” tray.

Category tab bar is clipped at the right edge

At 1280 px the rightmost tab (Standards Bo…) is visibly truncated, and Research & Academic Labs 30 is offscreen entirely. There’s no scroll affordance and no overflow indicator, so on this width users with no other prompt may not realise three categories exist past the visible edge. Vertical scrolling cannot reach them either.

Recommendation. Either (a) tighten the per-tab horizontal padding so 9 tabs fit at 1280 px, (b) collapse low-volume categories into a More ▾ dropdown, or (c) add a fade gradient + arrow affordance so the bar is clearly horizontally scrollable.

Top-right CTAs use 3 different button shapes for 3 actions

FA2 News (blue solid pill), Advice (black solid pill), Feedback (green solid pill) sit side-by-side but use different palette colours and the third is an <a> while the first two are <button>. The visual hierarchy implies these have different importance levels, but they’re really three peer surfaces. The colour palette also doesn’t reuse anywhere else in the page (the brand blue is for primary tabs and source links; black is for body text; green is otherwise unused).

Recommendation. Use one neutral chrome (e.g. light-grey background, dark text) for all three, with a single primary CTA visually differentiated only if you have an actual hierarchy you want to enforce (likely Feedback = primary blue, News + Advice = secondary).

Filter bar should auto-collapse in table view

Table view exists for users who want data density. Yet the same chunky filter bar sits above, eating ~340 px of vertical space before any table row is visible. On 600px-tall viewports (laptops, half-screen browsers) zero data rows appear above the fold.

Recommendation. When viewMode === 'table', collapse the filter bar by default behind a + Filters chip. Keep Sort visible (the table headers handle sort anyway, so it could be hidden too). Even shaving 200 px brings 4 rows above the fold.

PL Teams and field-mapping pages look like two different products

This page (Builders & Projects) uses a serif H1, inline labels on the filter bar (TIER / GEOGRAPHY / etc.) at the leftmost position of each row, and a different stats treatment (23.0 avg score · 4 Tier 1 · 42 Tier 2  |  14 high PL relevance  |  24 production / mature…). PL Teams uses a sans H2-ish heading, uppercase grouping labels above chip rows, and the inline-text stat strip flagged in §1.

Both layouts have merit. They are not consistent with each other. A user toggling between PL Teams and Builders feels like they’re crossing a layout boundary, not switching a filter.

Recommendation. Pick one filter-bar pattern and apply to both. The inline-label-per-row pattern (used here) is denser and more legible than the uppercase-label-above-row pattern (used in PL Teams), and it scales better as filter groups grow. Migrate PL Teams to match.

Last updated 2026-03-05 is ambiguous

Does this mean the team last updated their PL Directory profile, the research report was last edited, or this dashboard’s data file was rebuilt? All three are plausible interpretations and they answer different reader questions.

Recommendation. Disambiguate the label: Profile updated / Report written / Last activity. If multiple of these exist, surface them all in the key-value strip.

Three different active-state colours for one filter row

In the multi-filter screenshot the active Apr 2026 pill is solid blue, the active Funding pill is a green outline with green icon background, and the active Verified pill is green outline. The user toggles them all the same way (click a pill, count updates) but the “active” affordance shouts three different things.

Recommendation. Pick one active treatment for the whole filter region. Suggest: filled tinted background, no border, accent-coloured ink. The filter-type can be hinted via the static icon/dot to the left of the label, not via the active-state colour. Keeps the active state predictable across rows.

Date-pill row wraps to two lines, isolating “All time”

At 880 px panel width, the time-range row holds 8 pills (Last 14 days, Apr 2026, Mar 2026, Feb 2026, Jan 2026, Dec 2025, Nov 2025, All time). The first 7 fit on line one; All time wraps alone to line two on its own. Hand-of-mouse symmetry breaks; the row also competes vertically with the second filter row below.

Recommendation. Either (a) tighten per-pill padding by ~2 px so 8 fit, (b) drop oldest months automatically (Nov 2025 has 162 items but is 6 months stale — collapse anything > 6 months into Older  ▾), or (c) move All time to be the rightmost option after a divider, with the implication that it’s an escape hatch.

Score and Type pills mix two different mental models

The second filter row stitches together type pills (Funding / Launches / Partnerships / …) + a Verified toggle + significance pills (★ High / Notable). These are conceptually different filters — type is a single-select tag, Verified is a boolean, significance is a numeric threshold — but they share a row and styling. New users have to learn that ★ High means ≥15 score, not has the High tag.

Recommendation. Either group them with an explicit divider + caption (TYPE — pills — QUALITY — verified+score), or move significance into the Sort row as a sort modifier (Latest · Top · High-only). Reduces the second row to: type pills + verified, which is more uniform.