The short answer: Optimal Workshop's constraint is not seats and it is not responses — both are unlimited. It is studies. The Starter plan is $199 per month billed annually and allows 5 studies launched per year. That works out to roughly $478 per study, and study number six requires a bundle purchase or an Enterprise quote. The best alternative depends on whether you need cheaper information-architecture testing specifically, or a research platform that does not meter the act of asking a question at all. Koji is the strongest choice for the second group.
Key takeaways
- Optimal Workshop publishes exactly two tiers: Starter at $199/month billed annually and Enterprise at custom pricing.
- Starter includes unlimited seats and unlimited participant responses per study — but 5 studies launched per year. That is $2,388 annually, about $478 per study.
- For the same $199 headline, Lyssna's Growth plan is $199/month monthly or $166/month billed annually and includes 5 in-depth studies per month — roughly 12x the annual study allowance.
- Research demand is rising into that cap. Maze's Future of User Research 2026 report (n~500, published 11 March 2026) found demand for research grew from 55% to 66% year over year, and the share of organisations where research is essential at all business levels nearly tripled from 8% to 22%.
- Koji does not meter studies at all. You pay per conversation: 1 credit for text, 3 for voice, starting at €29/month for 29 credits.
What Optimal Workshop actually costs in 2026
Read directly off Optimal Workshop's pricing page on 20 August 2026:
| Plan | Price | Seats | Responses | Studies | Buying |
|---|---|---|---|---|---|
| Starter | $199/month, billed annually | Unlimited | Unlimited per study | 5 launched per year | Self-serve |
| Enterprise | Custom pricing | Unlimited | Unlimited per study | Custom study bundles | Contact sales |
The page is unusually generous on the two dimensions most vendors squeeze — it really does say unlimited seats and unlimited participant responses per study — and unusually tight on the one that matters most to a working research team.
Why a study cap hurts more than a seat cap
A seat cap makes you ration readers. A study cap makes you ration questions, and that changes behaviour in a way that is genuinely bad for research quality.
At five studies a year you are budgeting roughly one per quarter with one spare. The practical consequences show up fast:
- You stop testing small things. A quick tree test on a single navigation label is exactly the sort of cheap, high-value study that a study cap makes irrational to spend an allowance on.
- Studies get bloated. When each launch is scarce, teams bundle four questions into one instrument. Combining a card sort with an unrelated survey produces worse data than running them separately.
- Iteration dies. Information architecture work is inherently iterative: card sort, draft a structure, tree test it, revise, tree test again. That is three studies for one navigation redesign — 60% of your annual allowance on a single project.
- The cap arrives at the worst time. Studies four and five typically get spent by August, which is exactly when a redesign hits its decision point.
There is nothing improper about metering studies; it is a defensible way to price a tool whose value is per-project. But you should price it honestly against alternatives, and the comparison is stark. Lyssna's Growth plan carries the same $199 monthly headline — $166 per month if billed annually — and includes 5 in-depth studies per month rather than per year.
What to look for in an Optimal Workshop alternative
- Check the metered unit before the price. Seats, responses, studies and conversations are four different bills wearing the same dollar sign.
- Separate the methods you actually use. Most teams buy Optimal Workshop for card sorting and tree testing and use the survey module twice.
- Ask what happens after the data lands. Dendrograms and pietrees are excellent at showing what people did and silent on why.
- Confirm participant sourcing. Bringing your own participants is cheap; buying them rarely is.
- Watch for the Enterprise cliff. When the only upgrade path is a quote, your sixth study has no published price.
The 8 best Optimal Workshop alternatives in 2026
1. Koji — best when you need the reasoning behind the structure
Card sorting tells you that 40% of users filed "Billing history" under Account rather than Payments. It cannot tell you that they did it because they were looking for a receipt to expense, not to change a card — which is the finding that actually changes your navigation.
Koji runs AI-moderated interviews at survey scale. The AI interviewer asks a real person to walk through where they would look for something and why, follows up when the answer is vague, and does it with hundreds of people simultaneously without a moderator on a calendar. You can combine that with Koji's six structured question types — open_ended, scale, single_choice, multiple_choice, ranking and yes_no — so a single study can ask people to rank candidate category names and explain the ranking.
Nothing is metered by study. Insights is €29/month for 29 credits, Interviews is €79/month for 79 credits (€290 and €790 annually, two months free). A text conversation costs 1 credit, voice 3, and a report refresh 5. Overage is a flat €1 per credit. New accounts get 10 credits free on signup, no card. Only conversations scoring 3 or higher on quality consume a credit, so incomplete or low-effort sessions do not bill.
Best for: teams whose IA problem is really a mental-model problem, and anyone tired of rationing studies.
Honest limitation: Koji does not produce a dendrogram. If you need a statistically clean closed card sort with similarity matrices, pair it with a dedicated IA tool rather than replacing one with the other.
2. Lyssna — the closest direct replacement, at 12x the study allowance
Formerly UsabilityHub. Card sorting, tree testing, first-click testing, five-second tests, preference tests and surveys, plus its own participant panel. Growth is $199 USD/month, or $166/month billed annually, and includes 5 in-depth studies per month alongside a quota of quick studies; there is a genuinely usable free plan and an Enterprise tier. For teams who like Optimal Workshop's methods and only object to the cap, this is the shortest move. See Lyssna alternatives for the fuller picture.
3. Maze — best for continuous testing wired into design tools
Maze is built around rapid, unmoderated testing on prototypes, with strong Figma integration and a continuous-discovery posture rather than a project-by-project one. Pricing is quote-led at the top end. Compare in Maze alternatives.
4. UXtweak — best budget IA toolkit
Card sorting, tree testing, session recording and a free tier that is more than a demo. Business pricing is custom. A reasonable landing spot when the objection to Optimal Workshop is purely cost.
5. UserTesting — best for watching real humans struggle
If what you want from an IA study is the video of someone failing to find the thing, UserTesting remains the category benchmark, with the category's largest price tag. See UserTesting alternatives.
6. Useberry — best for prototype-first teams
Strong on prototype flows and heatmaps with straightforward self-serve pricing. Less depth on classic IA research than Optimal Workshop or Lyssna.
7. Dovetail — best if the real gap is the repository
Some teams buy an IA tool and then discover their actual problem is that findings from previous studies are unfindable. Dovetail is an analysis and repository layer, not a testing tool — a complement rather than a replacement. See best UX research repository tools.
8. Your own recruiting plus a free card-sorting method
For small teams, an unmoderated card sort run with existing customers and a spreadsheet is a legitimate option. It is laborious and it does not scale, but at five studies a year the DIY route may already be what you are doing between allowances.
Choosing without a spreadsheet
- You like the methods, you hate the cap -> Lyssna.
- You test prototypes constantly -> Maze or Useberry.
- You need video of real struggle -> UserTesting.
- Budget is the whole objection -> UXtweak.
- Your navigation problem is really a mental-model problem -> Koji.
Before you rebuild anything, it is worth being precise about which method answers your question. Our guides to card sorting, tree testing and information architecture research cover when each one is the right instrument — and our usability testing guide covers the case where the structure is fine and the interface is the problem.
A practical migration plan
Export your existing study results first, including the raw participant-level data rather than just the summary charts — similarity matrices are painful to reconstruct. Then map your last twelve months of research questions against the five-study allowance you were working within; most teams find two or three questions they simply never asked because the allowance was gone. Those unasked questions are the real cost of the cap, and they are the right first studies to run on whatever you move to.
Finally, run one IA study in parallel. Same participants, same task, both tools. It takes a week and it settles the argument better than any feature comparison.
Frequently asked questions
How much does Optimal Workshop cost in 2026?
Optimal Workshop publishes two tiers. Starter is $199 per month billed annually — $2,388 per year — and includes unlimited seats, unlimited participant responses per study, and 5 studies launched per year. Enterprise is custom-priced with bespoke study bundles and requires contacting sales.
What counts as a "study" on Optimal Workshop's Starter plan?
Each launched study consumes one of the five annual allowances, regardless of how many participants respond to it. Because responses per study are unlimited, the plan rewards fewer, larger studies and penalises the small, iterative tests that information architecture work usually calls for.
Is there a cheaper alternative to Optimal Workshop with the same methods?
Lyssna is the closest like-for-like option. Its Growth plan is $199 USD per month, or $166 per month billed annually, and includes 5 in-depth studies per month rather than per year — roughly twelve times the annual study allowance at a comparable headline price. UXtweak is cheaper still and has a usable free tier.
Can AI interviews replace card sorting and tree testing?
They replace the explanation, not the measurement. A card sort produces a quantitative view of how a group clusters concepts; an AI-moderated interview explains the reasoning behind the clustering and catches the mental model your categories violate. The strongest IA research uses both — a card sort to see the pattern, interviews to understand it.
Does Optimal Workshop include participants?
You can invite your own participants at no additional cost, and the platform also offers paid participant sourcing. Recruiting costs sit outside the $199 subscription, so budget them separately when comparing platforms.
What is the fastest way to test a navigation change without a study allowance?
Run a short AI-moderated interview study asking people where they would look for a specific task, and why. With Koji this costs credits rather than a study slot — 1 credit per text conversation, 3 per voice conversation — so testing a single label is economically sensible instead of a waste of a quarterly allowance.
The bottom line
Optimal Workshop is a genuinely good information-architecture toolkit, and its pricing page is more honest than most: unlimited seats and unlimited responses are real commitments, not asterisks. But the unit it meters — studies — is the one unit that should never be scarce, because research quality depends on being able to ask small questions cheaply and often.
If your team spent last year deciding which questions were worth an allowance slot, the problem was never the price. It was the permission.
Try Koji free. Ten credits on signup, no card, no study cap — enough to run a real AI-moderated study on your navigation and find out not just where people looked, but what they thought they were looking for. Start free at koji.so.
