Sentiment
Scheduled Sep 18, 2026Engagement and performance data your HR team owns.
Replaces
- Culture Amp · Culture Amp Pty Ltd — no published list price
- Lattice · Lattice, Inc. — $13 per seat per month (Foundations), about $62,400/yr at 400 seats
Rented, per year
$62,400
Lattice list price at 400 seats.
Owned, per year
$4,680
AWS infrastructure only. Azure is $5,640.
Build
8 weeks
2 engineers. To parity on the core features above, including migration of historic survey cycles and review history from Culture Amp and Lattice exports.
Difficulty
4/5
Substantial. Compliance or realtime depth involved.
Reference organisation: 600 employees, 6 survey cycles a year, 3 years of review history, ~4M responses, single region, prod + non-prod. 3-year Compute Savings Plan on Fargate / 3-year reservation on Container Apps. Aurora minimum 0.5 ACU per cluster, non-prod stopped outside business hours. Excludes LLM gateway token cost, vendor support plans and any licensed survey instrument. Figures are infrastructure only and exclude the build.
What it is
Sentiment is an engagement and performance system: pulse and annual surveys with anonymity guarantees you can audit, review cycles with calibration, 1:1 agendas, and goals that roll up. It is aimed at organisations past roughly 300 employees, where the combined bill for engagement and performance has passed six figures and the vendor holds the most sensitive dataset in the company. The people data — who said what about their manager, who was rated what and why — sits in a schema you cannot inspect, under a contract you renegotiate from a weak position every year.
Why this one stops making sense
- Lattice lists Foundations at $13 per seat per month, with Performance at $10, Goals & OKRs at $8 and Engagement at $4 as separate seat lines. A 600-person company buying Foundations, Performance and Engagement is at $27 per seat per month — $194,400 a year.
- Culture Amp publishes no price at all. Every renewal is a negotiation in which the vendor knows your headcount, your engagement score and your renewal date, and you know none of their cost base.
- Anonymity is the whole product, and you are asked to take it on trust. You cannot read the vendor's schema, you cannot see their join keys, and you cannot show an employee the code that protects them.
- Review history is the evidence base for promotion, pay and, eventually, employment tribunals. Exporting it as CSV at the end of a contract is not the same as owning it.
- Per-seat pricing means the cheapest way to reduce spend is to survey fewer people, which is the opposite of what the tool is for.
What ships
- Pulse, annual and lifecycle surveys with branching logic and multi-language delivery
- Anonymity enforced by a minimum-respondent threshold applied at query time, not at report time
- Heatmaps by team, tenure, location and manager, with drill-down blocked below threshold
- Review cycles: self, manager, peer and upward, with configurable visibility per stage
- Calibration sessions with distribution views and a recorded rationale per adjustment
- 1:1 agendas with shared and private notes, carried-forward items and talking points
- Goals and OKRs with check-in cadence, roll-up to parent goals and drift detection
- Manager-level action plans generated from survey results, tracked to completion
- Full export of every response, rating and comment as a documented Postgres schema
- Retention rules per data class, including hard deletion of free-text comments on a schedule
Data model
Stack
- frontend
- Next.js 15 App Router + Tailwind + TanStack Query
- backend
- NestJS on Node 22 — REST, with a separate read-only analytics service
- database
- PostgreSQL 16, split into a directory database and an isolated response database
- cache
- Redis 7 for session state, aggregate caches and threshold-check memoisation
- queue
- BullMQ for cycle scheduling, reminder email and export generation
- auth
- OIDC against the customer IdP; SCIM 2.0 for joiner/mover/leaver
- search
- PostgreSQL tsvector over comments, scoped by the same threshold rules
- ai
- Optional theme extraction and comment clustering through a pluggable LLM gateway
Parity, honestly
What we match, and what we do not.
The rows marked No are the important ones. If one of them is the reason you bought Culture Amp, keep buying it — and we will tell you that on the first call rather than the fifth month.
| Capability | Culture Amp | Ours | Note |
|---|---|---|---|
| Pulse and annual engagement surveys | Yes | Yes | — |
| Anonymity threshold on reporting | Yes | Yes | Ours is enforced in the query layer and the code is readable by your works council. |
| Review cycles with peer and upward feedback | Yes | Yes | — |
| Calibration with distribution views | Yes | Yes | — |
| 1:1 agendas and shared notes | Yes | Yes | — |
| Goals and OKRs with roll-up | Yes | Yes | — |
| Cross-company engagement benchmarks | Yes | No | Culture Amp benchmarks against millions of responses from thousands of companies. We cannot manufacture that dataset and will not pretend to. If benchmarking against the market is why you bought it, keep buying it. |
| Validated survey instrument and IO-psychology question bank | Yes | No | Culture Amp's People Science team maintains and validates the question set. We ship your existing questions and can license a validated instrument separately, but we do not employ psychometricians. |
| Vendor-run SOC 2 and ISO 27001 attestation | Yes | No | Sentiment inherits your own cloud compliance posture. If your customers demand a third-party attestation for the HR system, that is a real reason to keep a vendor. |
| Bundled HRIS and payroll | Partial | No | Lattice sells an HRIS alongside. Sentiment reads your existing HRIS over SCIM or a nightly extract and does not try to be one. |
| Compensation review and merit cycles | Partial | Partial | Lattice charges $6 per seat per month for this as an add-on. We build the merit-matrix and budget-tracking parts; equity modelling is a separate engagement. |
| Native mobile apps | Yes | Partial | Installable PWA with push. A store-listed native app is a separate four-week engagement. |
| Sentiment analysis over free-text comments | Yes | Yes | Runs through your own LLM gateway. Comments never leave your tenancy and are not used for training. |
| Works council and GDPR data-subject workflows | Partial | Yes | Deletion and export run against a schema you can inspect, which is the only way to answer a works council properly. |
Reference architecture
Built twice, on purpose.
Every application in Techtons ships with a production architecture for AWS and one for Azure, using the same diagram grammar so you can read them side by side. Your cloud, your account, your bill.
Amazon Web Services
- Compute
- ECS Fargate services (api, analytics, worker) behind an ALB
- Data
- Two Aurora Serverless v2 PostgreSQL clusters — directory and responses — plus ElastiCache Redis
- Storage and edge
- S3 for exports and review attachments, CloudFront for the static front end
- Identity
- Cognito user pool federated to the customer IdP via OIDC/SAML
- Observability
- OpenTelemetry to CloudWatch and X-Ray; audit events shipped to a separate log group with a 7-year retention policy
- Infrastructure as code
- Terraform, one module per environment, remote state in S3 with a DynamoDB lock
Services
The decision worth arguing about
Two databases, not one. Survey responses live in their own Aurora cluster with no foreign key — and no security group path — to the directory cluster that holds names, and the only place the two can be correlated is a service that refuses to return any cell with fewer than five respondents. The obvious design is one database with Row Level Security, and we rejected it: RLS constrains the application, not a database administrator with a read replica and a quiet Sunday. The trade-off is not free. Longitudinal questions — did engagement move for the people who changed manager in Q2 — become an application-level join over salted cohort identifiers rather than a SQL join, which is slower to run and considerably harder to write correctly. We think that is the right price for being able to tell people that anonymous means anonymous and then show them the code.
Microsoft Azure
- Compute
- Azure Container Apps (api, analytics, worker) with KEDA scaling on queue depth
- Data
- Two PostgreSQL Flexible Servers — directory and responses — plus Azure Cache for Redis
- Storage and edge
- Blob Storage for exports and attachments, Front Door Standard for edge and WAF
- Identity
- Microsoft Entra ID directly, with Entra groups driving reviewer visibility
- Observability
- Azure Monitor and Application Insights, OpenTelemetry-native; audit events to a dedicated Log Analytics table
- Infrastructure as code
- Bicep, azd-compatible layout
Services
The decision worth arguing about
The same split, enforced by a different mechanism. The response server sits on its own private endpoint in its own subnet and is encrypted with a customer-managed key held in a Key Vault whose access policy deliberately excludes the subscription owners — so on Azure the separation is enforced by key custody rather than only by network topology, and revoking the key makes the data unreadable to everyone including us. The cost is unavoidable and worth stating plainly: Flexible Server has no scale-to-near-zero equivalent to Aurora Serverless v2, so the second database is a permanent monthly floor instead of something that idles down to fractions of a cent overnight. We run it on a B2s burstable to keep that floor near $60 a month rather than $200, and accept that a company-wide survey launch will burn through burst credits for the first hour of the window.
Sources
Every price on this page, with the page we read it from and the date we read it.
| Product | Plan | List price | Checked | Source |
|---|---|---|---|---|
| Culture Amp | Engage / Perform | Not published | 2026-09-05 | www.cultureamp.com/pricing Culture Amp publishes no list price. Its pricing page shows 'Request a quote' against every module and states that price depends on headcount, product and service tier, with annual billing only. |
| Lattice | Foundations | $13 / seat/month | 2026-09-05 | lattice.com/pricing Performance ($10), Goals & OKRs ($8) and Engagement ($4) are listed as separate per-seat lines on the same page. Minimum annual agreement is $4,000. |
Should you still be paying for Culture Amp?
Two weeks, fixed price. We audit what you actually use, map it against Sentiment, cost the replacement on your own AWS or Azure account, and give you a delivery plan. If the honest answer is to keep the licence, that is what the report will say.