Legal technology earns its keep in quiet moments: when an anxious client finally understands what happens next, when a paralegal uploads the right document to the right matter without a phone call, when a self-represented litigant completes a form that would otherwise have defeated them. Those moments are products of design. And yet the legal sector still treats user experience and accessibility as polish applied at the end, rather than as the structural decision they really are, one that determines who can use the system at all.
The stakes are unusually high. A confusing workflow does not merely annoy; it produces missed filing dates, privilege errors, billing disputes, and broken trust. An inaccessible portal does worse: it locks out the people who most need help. The question that should govern every legal interface is blunt and practical, can the intended person complete the task accurately, confidently, and without unnecessary assistance? For a large share of legal software, the honest answer is still no.
The justice gap is partly a usability gap
Begin with the scale of unmet need. The Legal Services Corporation's 2022 Justice Gap study found that low-income Americans received no or inadequate help for 92% of the civil legal problems that substantially affected them, with about 74% of low-income households facing at least one such problem in a single year (Legal Services Corporation). Cost is the most cited barrier, roughly 46% of those who skipped seeking help pointed to affordability (LSC via Prairie State Legal Services), but money is not the only wall. Interfaces that demand specialist vocabulary, assume a desktop, or break for assistive technology quietly add their own toll.
Better design will not close the justice gap on its own. It can, however, remove the avoidable friction layered on top of genuine scarcity: unreadable instructions, mobile-hostile forms, and language written for other lawyers rather than for the human being trying to keep their home or their children. Usability is not a soft concern in this context. It is an access-to-justice mechanism.
The civil justice gap, by the numbers
Share of low-income Americans affected, 2022 Justice Gap study
Accessibility is the foundation, not the finishing coat
If you want a single statistic to puncture complacency, use this one: in the 2026 WebAIM Million analysis of the top one million home pages, 95.9% had detectable failures against the Web Content Accessibility Guidelines, and the average page carried 56.1 distinct errors (WebAIM). These are not edge cases; they are the norm. Legal platforms, which handle high-stakes tasks for a public that includes people with visual, motor, auditory, cognitive, and language-related needs, cannot afford to sit inside that 95.9%.
The errors are also depressingly repetitive. Low-contrast text appeared on 83.9% of pages, missing alternative text on 53.1%, and missing form input labels on 51%, the last especially damaging in legal software, where forms are the product (WebAIM). Empty links and empty buttons, found on 46.3% and 30.6% of pages respectively, leave screen-reader users guessing what a control does (WebAIM). The good news is that a handful of fixes, applied at the design-system level, cascade across an entire product.
The recurring failures of the web
Share of one million home pages with each WCAG error type, 2026
Source: The WebAIM Million, 2026 report.
The standard to build against is settled. The W3C's Web Accessibility Initiative treats WCAG as the shared international benchmark, and the current version, WCAG 2.2, published in its recommendation form, is backward compatible with 2.1 and 2.0 at the same conformance level (W3C). For new and redesigned legal experiences, Level AA is the right floor.
The numbers that govern legibility
Two WCAG thresholds do more day-to-day work than any others, and every legal-product team should know them by heart. Normal-size text must reach a contrast ratio of at least 4.5:1 against its background to meet Level AA, while large text, roughly 24px regular or 18.66px bold and above, may use a lower 3:1 ratio (W3C WCAG 2.2). Non-text elements that carry meaning, such as form borders and focus indicators, need 3:1 as well. These are minimums, not targets; a 4.5:1 ratio is calibrated to compensate for the contrast-sensitivity loss typical of a person with roughly 20/40 vision.
| Requirement | Threshold (Level AA) | Why it matters in a legal portal |
|---|---|---|
| Text contrast | 4.5:1 normal, 3:1 large | Gray metadata and disclaimers stay readable in sunlight and for low-vision users |
| Non-text contrast | 3:1 | Form fields and the active "Submit filing" focus ring are visible |
| Keyboard operability | All functions | A user can open a matter, review, comment, and file without a mouse |
| Target size | 24 × 24 CSS px | Controls remain tappable one-handed on a phone |
| Color independence | Not color alone | Overdue tasks show a label, not just a red dot |
| Captions | Live synchronized media | A CLE webinar or client briefing is usable without sound |
Accessibility's quiet payoff is that it improves the product for everyone. Captions help people in noisy waiting rooms; strong contrast helps any user squinting at a phone outdoors; clear headings speed up scanning for the busiest partner. Inclusive design is not a parallel track bolted onto the roadmap. It is simply good legal product design with the edge cases taken seriously.
Plain language is a trust strategy, not a downgrade
Legal documents too often read as if addressed to opposing counsel rather than to the client who must act on them. When an engagement letter, fee arrangement, or scope limitation is incomprehensible, the firm has not communicated, it has merely transmitted. Plain language fixes this without sacrificing precision. The U.S. government's plain-language guidance frames the goal as writing for the reader: organizing information clearly and making public-facing content genuinely usable (PlainLanguage.gov).
A practical target for client-facing material, intake instructions, payment policies, portal notices, is roughly an eighth- to ninth-grade reading level, unless a statutory notice or jurisdictional requirement demands otherwise. The discipline that target enforces is familiar: shorter sentences, defined terms, concrete examples, and a direct statement of what the reader must do next.
| Conventional legal wording | Plain-language rewrite |
|---|---|
| "Client shall remit payment within ten days of presentment of invoice." | "Please pay each invoice within 10 days after you receive it." |
| "Failure to provide requested documentation may adversely affect representation." | "If we don't get these documents, we may not be able to move your case forward." |
| "The scope of representation excludes appellate proceedings." | "This agreement does not cover an appeal after the trial court's decision." |
The best workflow is rarely the shortest. It is the one that makes the right action easy and the dangerous action hard.
Design for confidence, not click counts
The old "three-click rule", that anything important should be reachable within three clicks, has the appeal of a slogan and the rigor of none. The Nielsen Norman Group has long argued that click counting alone fails to predict satisfaction or success, and that users will happily take more steps when each one feels purposeful (Nielsen Norman Group). In legal software, the better metric is confidence per step.
Consider a document upload. A six-screen flow that confirms the client, the matter, the privilege designation, and the recipient is safer, and ultimately less stressful, than a two-click shortcut that leaves the user wondering whether they just disclosed confidential material to the wrong party. The right pattern makes the destination obvious, validates early, shows progress, and offers a correction path. A confirmation screen on a risky action is not friction; it is a guardrail.
Measure experience like a business function
Opinion is a poor substitute for evidence, and mature teams instrument their portals accordingly. The most portable benchmark is the System Usability Scale, a ten-item questionnaire whose database average sits at 68, according to MeasuringU; scores above 80.3 map to an A-grade result, while anything below 51 is considered poor (MeasuringU). Because SUS is not a percentage, results need careful translation when they reach executives, a 68 is average, not a failing grade of 68%.
Reading a System Usability Scale score
Benchmark thresholds on the 0 to 100 SUS scale
Source: MeasuringU, interpreting SUS scores.
SUS is only one instrument in a balanced scorecard. Task success rate, task-completion time, user error rate, support contacts per task, and feature adoption together reveal where friction lives and what it costs. Quantitative analytics show where users struggle; moderated usability tests, accessibility audits with real assistive technology, and client interviews explain why. The strongest UX programs connect those findings to legal-business outcomes, fewer support tickets, faster matter setup, more complete timesheets, and higher client satisfaction.
| Metric | What it reveals | How to act on it |
|---|---|---|
| Task success rate | Whether users finish without help | Test upload, intake, payment, messaging, deadline acknowledgment |
| Completion time | Effort per successful task | Confirm redesigns shorten common lawyer and client flows |
| Error rate | Where mistakes cluster | Fix confusing labels, weak defaults, missing validation |
| Support contacts | Where humans must intervene | Prioritize the highest-volume support drivers |
| SUS | Perceived usability | Benchmark after major releases against the 68 average |
Mobile is now a legal-workflow requirement
Legal work no longer lives at a desk. It happens in court corridors, at client sites, in airports, and for many clients, a phone is the primary, sometimes only, route online. A desktop-only portal is therefore not merely dated; it is exclusionary. Mobile-first design means building the essential task for the smallest screen first and enhancing upward, which forces teams to decide what truly matters. The same target-size and contrast rules that satisfy WCAG also happen to make a phone usable with one thumb in poor light.
How clients reach a legal portal, by primary device
Illustrative access mix for a public-facing legal services platform
Illustrative composition reflecting the mobile-primary reality documented by access-to-justice research; figures are directional, not survey-precise. Context: Legal Services Corporation.
What good looks like
The operating principles cohere into a single discipline. Make every workflow legible, so users always know where they are and what happens next. Write for comprehension, treating plain language as compatible with, not opposed to, legal precision. Build accessibility into the definition of done, so no feature ships unless it works with a keyboard, a screen reader, sufficient contrast, and clear error handling. Design the core task for a phone first. And measure continuously, because UX is never finished; each release should be tested against task success, error rates, accessibility, and user confidence.
The next generation of legal technology will not win on feature lists. It will win on whether a frightened client, a hurried associate, and a self-represented litigant can each accomplish what they came to do. The platforms that internalize this, that treat usability and accessibility as the structure rather than the trim, will not just digitize old legal workflows. They will make justice-related services navigable for the people who have been locked out the longest.
Sources
- Legal Services Corporation, The Justice Gap: The Unmet Civil Legal Needs of Low-income Americans (2022)
- Prairie State Legal Services, 2022 Justice Gap Report key findings
- WebAIM, The WebAIM Million, 2026 report on the accessibility of the top one million home pages
- W3C, Web Content Accessibility Guidelines (WCAG) 2.2
- W3C WAI, Understanding Success Criterion 2.5.8: Target Size (Minimum)
- W3C Web Accessibility Initiative, WCAG overview
- MeasuringU, 5 Ways to Interpret a SUS Score
- MeasuringU, Measuring Usability with the System Usability Scale
- Nielsen Norman Group, The 3-Click Rule for Navigation Is False
- PlainLanguage.gov, Federal Plain Language Guidelines
- AudioEye, Takeaways from the WebAIM Million report