NEW Introducing Projects Introducing Projects: enter an iterative loop with the future

User simulation to help your AI make better decisions

Semilattice gives you and your agents accurate user feedback in minutes, with state-of-the-art models built from real human data.

Get started free See how it works

Projects › Upgrade campaign Context: Acme · Marketing PRSM

upgrade-campaign-brief.md v1

Upgrade campaign

Brief · Lifecycle marketing · owner: you

Context

4 in 10 premium subscribers hit their usage limit every month, but few upgrade on their own.

They hold back because the next plan doesn’t look worth the price.

Key messages

1.Offer the next plan up when usage hits 80%.

2.Email first, then an in-app banner a day later.

3.Show the price difference in every message.

4.Cap usage until they upgrade.

5.Campaign runs for two weeks from launch.

Success metrics

1.Primary: upgrade rate to the next plan.

2.We expect a 15% upgrade rate within two weeks.

“4 in 10 premium subscribers hit their…”

Context

Kept as context

“the next plan doesn’t look worth the price”

Supported 68%

68% say the price gap is why they haven’t upgraded

“Offer the next plan up when usage hits…”

Supported 72%

72% say they would upgrade if offered at the limit

“an in-app banner a day later”

Mixed 52%

52% say the banner makes no difference

“Cap usage until they upgrade.”

Not supported 1 in 3

1 in 3 say they would cancel rather than upgrade

“a 15% upgrade rate”

Needs depth 15%

Supported for annual plans only

campaigns

✳ Claude Code

> test upgrade-campaign-brief.md with premium subscribers before I send it

⏺ semilattice - create_simulation (MCP)(Premium subscribers)

⎿ 6 claims to test, 752 simulated users

⏺ semilattice - run_simulation (MCP)

⎿ Done in 3m 12s

⏺ Most of the brief is supported, but 1 in 3 say they would cancel if usage were capped.

✓ Supported 72% “Offer the next plan up when usage hits 80%.”
72% say they would upgrade if offered at the limit

~ Mixed 52% “an in-app banner a day later”
52% say the banner makes no difference

× Not supported 1 in 3 “Cap usage until they upgrade.”
1 in 3 say they would cancel rather than upgrade

Replace the usage cap with a reminder at 100%, and test that line again?

> yes▍

upgrade-campaign-brief.md Preview upgrade-campaign-brief.md

Upgrade campaign

Audience

Premium subscribers who used more than 80% of their monthly limit in two of the last three months.

Key messages

1. Offer the next plan up when usage hits 80%.

2. Email first, then an in-app banner a day later.

3. Show the price difference in every message.

− Cap usage until they upgrade.

+ Send a reminder when usage hits 100%.

5. Campaign runs for two weeks from launch.

Channels

Email on day one, sent from the account owner’s name.
In-app banner on day two, shown on the usage page only.

Success metrics

Upgrade rate to the next plan (primary).
Email click-through within the first week.

For

“Using Semilattice has been game changing when developing product features. We’ve been able to source early, frequent, and instant customer feedback that’s almost identical to real life customer survey responses. It’s been instrumental in helping us make the right product decisions at every stage of ideation and discovery.”

Jem Corden · Head of Product, MONEYME

Read the case study

Use cases

Rapid user feedback on any message, asset, or flow.

Check a plan before you commit budget, an ad before you pay for media, and a landing page before launch day.

01 · Plans and documents

See which parts of a plan users support.

Upload a campaign plan, a concept, a pitch or an insights report. Semilattice marks the lines users agree with, the ones they push back on, and what they expect that isn’t there.

Homepage refresh · Plan

Homepage refresh

Draft 2 Maya Okafor·Marketing

Key messages

1. Lead with the free trial: no card required to get started.

2.Put customer logos and a case study directly under the hero.

3. Security page stays as a link in the footer.

4.Newsletter sign-up is optional, shown after the first visit.

Success metrics

Trial sign-up rate (primary) · new vs. returning visitors.

72% of UK adults say they’d try it 4 in 5 UK adults miss this

02 · Ads and creative

Check ads, images and copy before you spend on them.

Upload ad variants, social posts or emails. See which headline people remember, which image they trust and which line puts them off, so you can change it before any budget is spent on it.

Upgrade campaign / Paid social 2 variants

Acme Sponsored Variant A

Never hit your limit again

Upgrade

64% remember this headline

Acme Sponsored Variant B

Upgrade to Pro today

Upgrade now

2 in 5 read “today” as pressure

03 · Landing pages and conversion

Find out where visitors will drop off.

Paste a landing page or an offer page. Simulated users go through it and show you where they hesitate, what reassures them and which version converts better, before launch day.

folk.app/autumn-offer

folk.

ProductPricingCustomers

The CRM your team will actually keep up to date

Headline B lifts clicks by 18%

Free for 30 days, no card needed Visitors find this reassuring

Monthly · Annual, save 20% $24 / seat 1 in 4 stall at the price

How it works

Upload your work. See how to improve it.

Upload a plan, an ad, a landing page or a flow, and Semilattice shows you how users will react and what to change. Use it for the decisions that would otherwise be made on instinct, because a full research study would take weeks.

1 / 3

01

Upload your work

A plan, an ad, a landing page or a flow. Drop it in, paste a link, or say what you’re thinking of doing. If anything is unclear, Semilattice asks before it starts.

Projects / Homepage refresh ＋ New project

homepage-brief.md

Homepage refresh

M

⎘ homepage-brief.md

The free-trial message is the risky change.

Read it. Are we testing the whole site, or just the homepage?

M

Just the homepage.

Testing how visitors will react to the homepage.

Testing the homepage brief

homepage-brief.md

02

Semilattice simulates user reactions

User models built from real human data predict how your users will react to every line, image and step. Results come back in minutes.

Projects / Homepage refresh ＋ New project

User reactions · homepage-brief.md

Running

Free-trial appeal

UK Adults 2026 · 12 questions

Discovery User research

Homepage walkthrough

staging.folk.app · homepage to sign-up

Conversion User journey

Headline variants

UK Adults 25–34 · 9 questions

Messaging User research

03

Get actionable feedback

Your work comes back annotated: what lands, what doesn’t, and what to change. Click any note to see the question that was asked and how each group answered.

Projects / Homepage refresh ＋ New project

homepage-brief.md

Homepage refresh

Key messages

1. Lead with the free trial: no card required.

2. Put logos and a case study under the hero.

3. Security page stays as a footer link.

4. Newsletter sign-up after the first visit.

Success metrics

1.Trial sign-up rate (primary).

2.New vs. returning visitors.

Supported 72%

72% of UK adults say they would try it

Mixed 48%

Logos help; most skip the case study

Not supported 4 in 5

4 in 5 don’t find it; link it from the hero

Agents

Give your agents the power to check with users.

Connect the Semilattice MCP to Claude, Claude Code, Cursor, or any agent you work with. Your agent can ask simulated users how they’d react to what it’s making, and revise the work before you see it.

Claude Opus 5 ▾

We're moving to usage-based pricing. How will free, pro, and enterprise customers react?

Run simulation·Free tier

Run simulation·Pro tier

Run simulation·Enterprise

Reactions split sharply by tier, so a single rollout will not land.

•Free users barely notice; lead with the lower entry price.
•Pro users fear runaway bills; ship a visible spend cap.
•Enterprise wants predictability; offer an annual commit.

Reply to Claude…

Claude

Connect →

✳ Claude Code - ~/app

> simulate the new onboarding

⏺

Run simulation

⎿ 200 simulated users

⏺

Fetch results

✓ completion 64% → 78%

The new flow lifts completion, but the phone step is where it leaks.

• Move phone verification to the end; most drop-off is there.

• Keep the single-field email start; it tested well.

• Add a skip on the workspace step for solo users.

> ▍

Claude Code

Connect →

Cursor

Plan · annual billing toggle

Thought 4sRead pricing.tsx

List simulations

Fetch results

Checked what we already know: two past runs cover this.

•71% prefer the monthly default; keep annual opt-in.
•Show the saving inline; it doubled toggle intent.
•No new run needed; wiring the toggle now.

Add follow-up…

Plan Composer 2.5

pricing.tsx

1export function Pricing() {

2 const [annual, setAnnual] = useState(false)

3 const price = annual

4 ? plan.yearly

5 : plan.monthly

6 return (

7 <BillingToggle

8 annual={annual}

9 onChange={setAnnual}

10 saving={plan.saving}

11 />

12 <PriceLabel value={price} />

13 {annual && <SaveBadge amount={plan.saving} />}

14 <PlanFeatures tier={plan.tier}

15 compact

16 />

17 )

18}

Cursor

Connect →

API

Build marketing and product tools with users in the loop.

Put user feedback inside the tools your team already uses. Automate testing of ad copy, emails, landing pages and onboarding flows from your own services, jobs and pipelines, and get back predicted reactions from simulated users.

Python SDK PyPI ↗ pip install semilattice TypeScript SDK npm ↗ npm install semilattice Any other language API reference ↗ GET /v1/user-models

Accuracy

Tested against real people, question by question.

Each user model is built from real human data. Accuracy is measured by comparing its predictions with how real people answered. Every model’s test results are shown in the product.

Models

Answers-1

Access: Public
Accuracy: Directional; What we found when we audited 1,000 predictions →

Ready-made models of audiences such as UK adults, developers and B2B software buyers, built from real survey data. Each result is the share of the audience that would give each answer.

US Adults UK Adults 2025 - All Adults Developers US B2B Software Buyers 2026 UK Adults 2026 - Consumer Finance +44 more

49

public models

88%*

average accuracy

*Mean across the 49 public models. A model’s accuracy is 100% minus the average gap between its predicted share and the real share for each answer, on questions held out from the model.

Anacreon

Access: Private
Accuracy: State of the art; Anacreon: state-of-the-art accuracy in simulating individuals →

Built from your own material: interviews, reviews and open-ended survey responses. It simulates hundreds of individual respondents, each modelled on one real person, so you can segment and re-cut the results.

On held-out questions it predicts a person’s exact answer 67.0% of the time, against 65.7% for the best published baseline.*

Exact-answer accuracy

Held-out individual-level questions

anacreon

67.0%

Park et al.

65.7%

Closeness on scale questions

1.0 is a perfect match

anacreon

0.77

Mega-Study

0.75

Socrates

0.74

Twin-2K-500

0.71

*Benchmarked on held-out questions against published academic baselines: Park et al. (2024), arXiv:2411.10109 · Toubia et al. (2025), arXiv:2505.17479 · Peng et al. (2025), arXiv:2509.19088 · Kolluri et al. (2025), EMNLP 2025

Accuracy research

answers-1

Mar 2026 Testing against the FCA Financial Lives Survey →
May 2026 What we found when we audited 1,000 predictions →

anacreon

Get started

Base

Get started free

Create an account and start getting rapid feedback now.

Pricing

Free

Uses answers-1 models

49 public user models

User research and user journey modes

The app, the MCP, and the API

Unlimited tests

Create account

Custom

Get a custom model of your users

Semilattice Custom is a partnership, not a licence. We build a model of your exact audience from your own research, so the feedback reflects your users, and we meet regularly to help you act on it.

Pricing from

$1,799 / month

Uses anacreon models

Everything in Base, plus:

A private model of your users

Organisation features with unlimited seats

Regular working sessions and support

Security and data handling

User simulation to help your AI make better decisions

Rapid user feedback on any message, asset, or flow.

See which parts of a plan users support.

Check ads, images and copy before you spend on them.

Find out where visitors will drop off.

Upload your work. See how to improve it.

Upload your work

Semilattice simulates user reactions

Get actionable feedback

Give your agents the power to check with users.

Build marketing and product tools with users in the loop.

Tested against real people, question by question.

Models

Answers-1

Anacreon

Accuracy research

Base

Custom

A guide to user-in-the-loop marketing and product work.