UX Designer Tools and Tech Stack: The Honest 2026 Audit (Real Prices, No Filler)

Most UX stacks I've audited aren't a stack. They're Figma plus six plugins someone installed in 2023, two research seats nobody has logged into since Q1, and a Zeplin renewal that auto-billed for the third year running.

The real cost of your design stack isn't the $15-per-editor Figma bill. It's the three tools you're paying for that nobody opens, the Hotjar account that runs in parallel with FullStory because two PMs each bought their own, and the UserTesting contract that locked you into 50 sessions when you used 11.

I'll keep this practical. Real 2026 prices. The trade-offs I've actually run into. And a 30-day audit plan you can start Monday and finish before next quarter's budget conversation.

Why this matters now (and why finance is going to ask)

Design budgets got cut in 2024, again in 2025, and the people approving renewals in 2026 are not the same people who approved them in 2022. They have a finance partner, a SaaS spend dashboard, and a list of every tool with under-50% utilization.

"We use Figma" used to be a complete answer. Now the follow-up is: which seats, who logged in last month, why does research need a separate tool, and can we consolidate the two session-replay vendors before renewal.

If you can't answer those questions in a meeting, somebody else will answer them for you, and you won't like the answer. The designers I know who survived the last two budget rounds had one thing in common: they walked into the meeting with the tool list, the per-seat cost, and a kill list already drafted. Defense beats panic.

The Core Six (what every UX stack actually needs)

Six categories. Most teams need one tool in each. A few need two. Almost nobody needs three.

1. Design — Figma is the default, and it's fine

Figma at $15 per editor per month on the Professional plan is the boring, correct answer for 90% of teams. The Organization plan jumps to $45 per editor per month for SSO, design system analytics, and branching, which matters if you're past 30 designers or in regulated land.

The honest alternatives:

• Sketch: $12 per editor per month, Mac-only. If your whole team is on Mac and you've been on Sketch since 2017, the migration cost to Figma probably isn't worth it. If you're starting fresh in 2026, you're picking Figma. I won't pretend otherwise.
• Penpot: open source, self-hostable, $0 for the community edition. Genuinely usable in 2026. I've seen two startups run on it because their security team wouldn't allow Figma cloud. If you have that constraint, it's a real option. If you don't, you're paying yourself in time saved by using the tool everyone else uses.

Don't buy Adobe XD. It's been on life support for two years.

2. Prototyping — the trap is buying Framer when Figma was fine

Figma's prototyping is good enough for 80% of what teams actually ship. Click-through flows, basic transitions, smart animate, the variables update in 2024. All of it lands in your existing $15 seat. You don't need a second tool for most user testing.

Where you actually need more:

• Framer: $15 per user per month on the Pro plan, with publishing on the Mini and Basic tiers below that. The pitch is real interactions and a publishable site, which matters if you're prototyping marketing experiments or your prototype IS the product. If you're prototyping app screens for usability tests, you don't need Framer. I've watched two teams buy Framer for "richer prototypes" and then never use the publishing layer, which is the entire point.
• ProtoPie: starts around $13 per editor per month on Pro, more for Enterprise. Real reason to buy it: complex sensor states, hardware prototypes, conditional logic that Figma can't fake. If you're designing for a tablet that talks to a Bluetooth scanner, ProtoPie pays for itself the first time you avoid building a real prototype in Swift. If you're not, skip it.

Rule of thumb I use: if your prototype needs to fool the user for 90 seconds in a usability test, Figma is fine. If it needs to fool the user for 10 minutes or simulate a real device, look at ProtoPie. If it needs to be the actual marketing site, Framer.

3. Research — pick one, then resist buying a second

Research is where stacks bloat fastest because every PM has a favorite tool and every designer has a different one.

• Maze: $99 per month on the Starter plan, $200+ on Professional. Strong on unmoderated tasks, tree tests, first-click tests, and prototype tests piped straight from Figma. If your team runs lots of small async tests, Maze is the right default. The reporting is good enough to share with non-researchers without translation.
• UserTesting: enterprise pricing, usually north of $30K per year for any contract that has the recruit panel attached. Strong if you need recruited participants in specific segments, moderated sessions at scale, and your research team has the bandwidth to run them. Weak if a generalist designer is going to run two studies a quarter, because you'll burn the seat and the session count.
• dscout: built for diary studies and longitudinal research. Niche but the right tool when the alternative is a months-long mess of WhatsApp screenshots. If you're not running diary studies, you don't need it.
• Lookback: moderated remote sessions, cheaper than UserTesting, around $25 per user per month on the lower tier. Good middle ground if you want to actually talk to users without committing to UserTesting's panel.

The mistake I see most: a team buys UserTesting Enterprise, the researcher leaves, and everyone defaults back to Maze for unmoderated work. Two years later the UserTesting renewal still goes through because nobody's name is on it. Name an owner per tool, every tool, every renewal.

4. Session replay — one is enough, two is a budget mistake

• Hotjar: Observe Plus from $39 per month, Business from $99. Heatmaps, session recordings, on-page surveys. The right default for product teams under 100 people who want session replay without building a data team around it.
• FullStory: enterprise pricing, talk-to-sales territory, but the funnel-and-event analysis is genuinely deeper than Hotjar. Right call if your PMs need to slice sessions by 12 properties and run cohort analysis without exporting to a warehouse.

I've never seen both belong on the same team. If you have both, one of them was bought by a PM who didn't know about the other one. Pick the cheaper one unless the analytics depth is paying for itself in shipped insights, then cancel the other before renewal.

5. Accessibility — Stark plus axe is enough for most teams

• Stark: Figma plugin, Individual plan around $5 per month, team plans scale up. Contrast checks, color-blind simulation, focus order, alt text suggestions. The minimum viable accessibility tool inside Figma. Any team shipping anything customer-facing should have this on every editor seat.
• axe DevTools: Pro starts around $40 per user per month, free tier for the basic browser extension. Audits the live site in DevTools. Right answer for catching what design tooling can't catch (actual rendered DOM issues, ARIA mistakes, screen-reader hierarchy).

If you're regulated, audited, or shipping public-sector work, you'll need more than these two. If you're a normal product team trying to not ship inaccessible work, Stark in Figma plus axe in browser is the floor, and the floor is fine.

6. Handoff — Dev Mode killed Zeplin for most teams

This is the category where most teams are still paying for something they don't need.

• Figma Dev Mode: built into Figma, included in your Professional or Organization seat. Dev-friendly inspect, code snippets, design tokens, status flags, deep links to Storybook. For 80% of teams, this killed Zeplin in 2023 and they're just not done unwinding the contract yet.
• Storybook: free, open source. The actual source of truth for components if you have a design system. Storybook plus Dev Mode is the modern handoff stack.
• Zeplin: Team plan around $7 per editor per month, Organization higher. There's a real case for Zeplin only if you're handing off to a fully external dev shop that doesn't have Figma seats. For most internal teams, Zeplin is paying twice for what Dev Mode already does.

I've personally killed Zeplin twice at two different companies. Both times the renewal was already in the budget for the next year. Both times the eng team had stopped opening it nine months earlier and nobody noticed.

Common pitfalls (things I've watched teams burn money on)

• Buying Framer because someone wanted "richer prototypes" and never using the publish layer. If you're not publishing, Figma was fine.
• Renewing UserTesting without checking session count. A 50-session contract used at 11 sessions is $30K for $7K of value. Negotiate down or move to Maze.
• Running Hotjar and FullStory in parallel. Two session-replay tools is a budget unforced error. Pick one.
• Paying for Zeplin after Dev Mode shipped. Two years late on this kill is normal. Three years late is on you.
• Twelve plugins per Figma file, three of which duplicate native features added in 2024. Every plugin is an audit, performance, and security item. Prune annually.
• Per-IC research seats nobody uses. If only one designer per quarter runs research, buy one rotating seat, not five.
• Forgetting to cancel. Annual contracts auto-renew. Calendar the renewal date 60 days out, not 7.

The 30-day audit (run this Monday)

This is the version I've run myself, twice. It works. It also makes you the person finance actually wants to talk to instead of avoiding.

Week 1 — inventory

List every tool. Every. One. Include the free ones because the free ones become paid ones at scale.

For each tool, capture:

• Tool name and category (one of the Core Six)
• Plan and monthly cost
• Number of seats
• Per-seat owner (a name, not a team)
• Last login per seat, pulled from the admin panel
• Annual renewal date

If you can't get last-login data from the admin panel, that's a finding by itself. Tools that don't tell you who's using them are tools you can't defend.

Week 2 — interview

Talk to three designers and one PM. Ask the same four questions:

1. Which tools do you open weekly?
2. Which tools do you open monthly or less?
3. Which tools have you never opened?
4. If we cut one tool tomorrow, which one and why?

The PM is critical. PMs see the analytics and prototype tools differently than designers. Their answer about session replay and prototyping is usually the deciding vote.

Week 3 — cut

Anything with under 50% weekly usage among seat-holders, downgrade or kill. Anything with two tools doing the same job, consolidate to the cheaper or better one. Anything with no named owner, the audit is complete and the answer is cancel.

A realistic example from a 12-designer team I worked with:

Total before: roughly $69,000 a year. Total after: roughly $14,300 a year. The team got better tooling, not worse. The win is subtraction.

Week 4 — document

Write a one-page "Design Ops README" and put it in the design system repo or the team Notion. Include:

• The tool list with owners and renewal dates
• The kill/keep rationale (one line each)
• The "we already tried this" list (Framer, FullStory, Zeplin, whichever ones got cut)
• Onboarding instructions for the next hire so they don't re-buy what you just killed

The README is the artifact that protects the audit a year from now when somebody new joins and re-pitches Framer. You can point at the doc and say "we evaluated this in April 2026, here's why we passed, here's what would change the answer." That's the conversation you want to have once, not every quarter.

Stack-cost-per-designer is the number you want

By the end of the 30-day audit, you should be able to answer one question without a spreadsheet: what does our design stack cost per designer per month?

A reasonable 2026 number for a normal product team is $150 to $300 per designer per month, all in. If you're above $400, you have unused seats or duplicate tools. If you're below $100, you're either understaffed on tools or you're a 2-person startup using free tiers, which is also fine.

The number itself isn't the goal. The goal is having the number, knowing what's in it, and being the person who can defend the line items in a 15-minute meeting. That's the version of design ops finance leaves alone.

What "good" looks like in 2026

You don't need a fancy stack to ship great work. You need:

• Figma seats that match your editor count, no more
• One prototyping escalation tool, used by maybe two people
• One research tool, with a named owner
• One session replay tool, picked deliberately
• Stark on every Figma seat, axe in the browsers that need it
• Dev Mode for handoff, Storybook for component truth
• A README that tells the next hire what's already been tried

That's the whole stack. Six categories. Seven or eight tools. A defensible number. A finance partner who stops asking.

The win is subtraction.

Learn More

• Product Designer Job Description
• User Research That Moves the Roadmap
• Design System Contribution Without Breaking the Existing One
• Wireframing and Prototyping That Survives Engineering