# Ptally vs WTW Radar

> How Ptally compares with WTW Radar and Radar Live for insurance pricing and real-time rating: who each suits, how models are built and changed, testing and approvals, AI features and commercial model.

By Ptally Pricing Team, 4 November 2026. Category: Comparisons.
Source: https://ptally.ai/blog/ptally-vs-wtw-radar

WTW Radar suits insurers with an established pricing team that wants one suite for analytics, price optimisation and real-time rating, particularly in competitive personal lines markets such as UK motor and home. Ptally suits insurers and MGAs that want a rating engine for live quoting, with readable, tested models, AI built into the workflow and no upfront cost. If your priority is building and optimising prices from your own data inside one analytics suite, look closely at Radar. If your priority is getting a product live and keeping rate changes fast and controlled, look at Ptally.

## What WTW Radar is

Radar is the pricing and rating software of WTW (Willis Towers Watson). WTW describes it as "an end-to-end solution for rules, analytics, and model deployment" and says it is "trusted by more than 500 insurers globally". The suite includes modelling, optimisation and Radar Live for rate delivery, alongside WTW's Emblem modelling software.

**Radar Live** is the real-time rating engine. It has a long record in UK personal lines. Ageas UK uses it for motor pricing, where it handles millions of quotes a day and reacts to market changes on price comparison websites. WTW also offers Radar Live as SaaS: IMT Insurance in the US adopted Radar Live SaaS in 2022 as a single rating engine for its property and casualty lines, and Thimble, a US small business MGA, selected Radar Live for commercial lines pricing in March 2026.

Radar has supported Python since version 4.14 in 2022. In September 2024 WTW added real-time execution of Python models in Radar, with access to more than 8,000 Python libraries.

On AI, WTW launched **Radar Vision** in 2025, an AI-driven performance monitoring tool. **Radar 5**, released in October 2025, added generative AI so users can query Radar Vision in free-form text, plus machine-led modelling, browser-based cloud hosting and Radar Fusion for commercial underwriting.

WTW has published integrations with Guidewire PolicyCenter and Sapiens. It does not publish pricing.

## What Ptally is

Ptally is an AI enabled insurance pricing platform for insurers and MGAs. A model is a readable JSON file of ordered steps: lookups, calculations, transforms, exclusions and referrals. Every change runs against the model's regression test suite and goes through draft, approval and publishing, with full version history. One engine serves a real-time API, embedded quote forms, and chat and voice agents. Ptally has served over 50 million quotes and handles aggregator volume. It covers any line of business in the UK, Europe and the US.

## Side by side

| | Ptally | WTW Radar |
|---|---|---|
| **What it is** | AI enabled pricing platform and rating engine | Pricing analytics, optimisation and rating suite, with Radar Live for real-time rating |
| **Typical buyer** | Insurers and MGAs of any size | Typically insurers with in-house pricing teams; strong in personal lines |
| **Analytics and optimisation** | Not the focus. Ptally implements rating structures designed elsewhere | Core strength: GLMs, machine learning, optimisation and monitoring |
| **Where models come from** | Rating guides, Excel raters, US rate filings or any modelling tool | Built by your team in Radar, Emblem or Python |
| **Model format** | Readable JSON of ordered steps, tests included | Vendor specific, with Python models supported |
| **Testing** | Regression test suite runs on every change | Impact analysis and monitoring; detail varies by setup |
| **Approvals and audit** | Draft, approve, publish; every quote tied to a version | Governance for deployed models described; detail not published |
| **AI** | AI drafts and checks models; MCP server for Claude, ChatGPT and Cursor | Radar Vision monitoring; generative AI queries in Radar 5; machine learning in modelling |
| **Integration** | Real-time API, embedded forms, chat and voice | Radar Live real-time rating; Guidewire and Sapiens integrations |
| **Upfront cost** | None. No implementation fee | Not published; typically a licence and an implementation project |
| **Commercial model** | Up to 1% of GWP, a flat fee per model, or Enterprise terms | Not published |

## The key differences

### Scope

Radar covers much of the pricing process: analysing data, fitting models, optimising prices, deploying them and monitoring results. That breadth is why it is common in competitive personal lines. Ptally covers a narrower part in depth: turning a designed rating structure into a production model, testing it, approving it, serving quotes and managing rate changes.

### Who does the work

Radar is built for pricing analysts and actuaries who model and deploy rates themselves, typically with WTW or a partner supporting the first build. On Ptally your team can build models itself, through the dashboard or with an AI client. If you would rather not, Ptally's pricing experts can build the model from your documentation, with no implementation fee.

### How models are read and checked

A Ptally model is plain configuration that an underwriter, actuary or auditor can read step by step. Each test is a full set of inputs and the expected result, and the suite runs on every change, so you can see exactly which quotes a rate change moves. The [debug trace](https://ptally.ai/developer/debugging) shows every input, lookup and formula for any quote, and each quote records the version that priced it.

### How AI is used

WTW applies AI mainly to analytics and monitoring: machine-led modelling, Radar Vision's early insight on inflation, competitors and claims, and free-text queries in Radar 5. Ptally uses AI to build and maintain the production model itself: extracting rate tables from documents, drafting steps and checking them against worked examples. Through the [MCP server](https://ptally.ai/developer/mcp), Claude, ChatGPT or Cursor can read, validate, test and edit draft models. Agents cannot publish. People approve every release.

### Cost and time to first quote

WTW does not publish Radar pricing, so ask for the full cost before the first live quote. Ptally is free to build. On the Performance plan you pay up to 1% of the gross written premium your models write. The Per model plan gives a flat annual fee per live model, and Enterprise adds a self-host option, SSO, audit and approvals, and custom SLAs.

## When WTW Radar is the better choice

- You have an established pricing team that wants one suite for modelling, optimisation, deployment and monitoring.
- You compete on price comparison websites and want demand modelling and optimisation next to the rating engine.
- You want a long established vendor with a large base of personal lines users, especially in the UK.
- Your team already models in Emblem or Radar and wants to stay in one suite.

## When Ptally is the better choice

- You are an MGA or insurer launching or migrating a product and want it live without a licence or implementation project.
- Your rates come from a rating guide, spreadsheet or filing, or from a modelling tool you already use.
- You want models people can read and audit, with tests on every change.
- You want general AI clients to work on models under the same controls as your team.
- You want to pay as premium is written rather than before launch.

## Next steps

See how a Ptally model is built and tested on the [developer page](https://ptally.ai/developer), how quotes are priced through the [API](https://ptally.ai/developer/api), and the [pricing plans](https://ptally.ai/#pricing). Our guide to [pricing for aggregators](https://ptally.ai/blog/pricing-for-aggregators) covers comparison site volume, and the [buyer's checklist](https://ptally.ai/blog/insurance-pricing-platform-buyers-checklist) lists the questions to ask any vendor.

*This comparison reflects publicly available information as of November 2026. Vendor offerings change and vary by customer, so check current details with each vendor.*