Chesnara
Actuarial Modelling & Data Solution
Howden Insurance Actuarial & Longevity (IAL) — Proposal
27 July 2026
Private & Confidential
Contents
- Executive Summary
- Technology solution
2.1 Recommendation: Python for data and modelling
2.2 Python ecosystem
2.3 Target architecture
2.4 Benefits of our recommendation to Chesnara
- Data Solution
3.1 Scope
3.2 Our credentials
3.3 Our capabilities
3.4 Proposed solution
3.5 Ongoing support operating data solution
3.6 Codelivery
- Actuarial Model
4.1 Scope
4.2 Our credentials
4.3 Our capabilities
4.4 Proposed solution
4.5 Codelivery
- Implementation Plan
5.1 Delivery principles
5.2 Indicative timetable
5.3 Delivery path
5.4 Training
5.5 Post go-live support
- Resource Plan
6.1 Howden team
6.2 Co-delivery
6.3 Data operating secondments
6.4 Indicative effort
- Commercial model
7.1 Commercial structure
7.2 Commercial terms
7.3 Incentive for Chesnara resource
7.4 Combined data and modelling engagement
7.5 Fees
7.6 Post go-live support
7.7 Value included during build
- Appendix A: About Howden
8.1 Howden Group
8.2 Howden Insurance Actuarial & Longevity
- Appendix B: Case Study
9.1 Python model development
# 1. Executive Summary
The Scottish Widows Europe SA portfolio acquired as part of Project Sheen requires a documented, validated and audit-ready actuarial data and modelling solution that supports reporting requirements today and can be developed as requirements evolve.
Following discussions with Jason Treadwell and the wider Project Team, and having considered the clarification responses provided on 14 July 2026, our proposal has been designed to deliver those outcomes.
You will benefit from a documented, validated and audit-ready actuarial data and modelling framework built in Python, underpinned by actuarial modelling, data and reporting expertise and a strong focus on documentation, validation and governance.
The result will be a scalable, transparent and flexible solution that supports the project timetable, preserves the strengths of the current Lloyd’s process (including working-day-one model-point production) and gives Chesnara full ownership of the solution, with no ongoing licence costs an no vendor-specific lock in.
The table on the following page summarises the key features of our proposal and the benefits they provide to Chesnara.
We are incredibly delighted to build a long-term relationship with you and we look forward to discussing our proposal with you. Please get in touch with any clarification questions.
Amit Lad & Kim Durniat
| Area | Description | What this means for you |
|---|---|---|
| Python based solution | We will build your solution using Python, that has no ongoing licence costs, and no vendor lock-in period. The solution that we will build will be scalable and easy to extend. | A fully customised solution that meets your needs, with complete ownership and flexibility to adapt as future acquisitions are integrated. |
| Large expert team | Howden brings together over 80 insurance actuarial professionals by joining the insurance experts from Barnett Waddingham and Hymans Robertson. (See Appendix A) | Delivery by skilled experts with bench strength to protect the timetable when unforeseen issues arise. |
| Proven expertise | Extensive experience of model build, model transformation and audit support. We know what good model governance looks like. | Models and processes designed to meet internal governance standards, with documentation and testing aligned to what auditors look for. |
| Actuaries first | Led by actuaries, supported by a dedicated lead developer | More than a specification build: challenge on product features, simplifications, materiality and methodology, with models aligned to Technical Actuarial Standards and outputs fit for downstream reporting. |
2. Technology solution
Chesnara is not inheriting any models as part of this acquisition and does not have a modelling solution in place. Chesnara has a blank slate for choosing its technology platform.
2.1 Recommendation: Python for data and modelling
We recommend Python for both the data solution and the actuarial modelling solution.
In our opinion, it best meets Chesnara’s priorities:
- Governance and transparency
- Performance at scale
- Ease of implementation
- Full IP ownership
- Scalability for future M&A Python has become the industry standard across finance and related sectors because it combines powerful numerical capability with exceptional flexibility. Rather than relying on proprietary technologies that create ongoing licence cost and vendor lock-in, Python provides an open, well-supported platform that allows Chesnara to build, own and evolve solutions alongside the business.
This technology recommendation covers four linked decisions:
- Language / calculation platform — Python
- Data processing libraries — NumPy, Pandas and Polars (used where each is strongest)
- Hosting and runtime — Chesnara-hosted, Python + database
- Operating model — Chesnara retains full ownership after build ## 2.2 Python ecosystem
Python has a mature ecosystem of specialist libraries. We will use best-in-class tools for different aspects of data processing and calculation, rather than forcing one library to do everything. The following libraries will be core in our development.
| Library | Overview | Primary strength | Business benefit |
|---|---|---|---|
| NumPy | Foundational numerical computing library for highly optimised array and mathematical processing | High-performance numerical processing | Fast calculations; efficient analytics; support for advanced modelling workloads |
| Pandas | Industry-standard framework for structured business data from multiple sources | Data manipulation and business analysis | Rapid development; broad compatibility; powerful reporting; integration with existing extracts |
| Polars | Modern high-performance DataFrame library for large-scale processing | Large-scale data processing | Improved speed; lower memory use; headroom as volumes grow or further books are added |
2.3 Target architecture
Our solution will be hosted by Chesnara on its platform of choice. We will confirm the environment with Chesnara at kick-off. This will cover…
| Layer | Requirement |
|---|---|
| Runtime | Supported Python version (agreed at kick-off); package management (e.g. pip / poetry / conda as preferred by Chesnara) |
| Database | Access to a relational database for staging extracts, validated model points, run metadata, data reports and audit logs |
| Storage | Secure file / object storage for source extracts, assumption sets and output packs |
| Access control | Role-based access for developers, production users and read-only reviewers; segregation of non-production and production where required |
| Deployment | Repeatable deployment via continuous integration / continuous delivery (CI/CD) pipelines so changes are controlled, tested and reproducible |
| Connectivity | Controlled access to Lifeware / State Street extracts. |
2.4 Benefits of our recommendation to Chesnara
| Benefit | Detail |
|---|---|
| No ongoing licence costs | No per-user or per-policy model licence after build |
| Full ownership | Chesnara owns code and IP |
| Flexibility | Easy to extend for future M&A activity |
| Performance | Mature numerical and data stack suited to ~45,000–60,000 policies today, with headroom for future M&A activity |
| Auditability | Transparent calculation logic; reproducible runs; clear audit trail |
| Database integration | Native fit to Chesnara’s chosen database and file estate |
| AI-ready | Natural path for future analytics / controls innovation, subject to Chesnara approval and regulatory expectations |
3. Data Solution
3.1 Scope
The data solution is tool will:
- Take extracts from source systems (Lifeware policy and reinsurance data; State Street asset / investment data; economic assumptions from public sources; non-economic assumptions from Chesnara).
- Transform data into structured inputs for the valuation models. These inputs cover policy data, asset data, and assumption tables.
- Apply automated data integrity and quality checks, reconciliations, and produce an auditable log of transformations and overrides. You would also like ongoing resources to support with operating the data solution.
3.2 Our credentials
We have significant experience transforming, validating and analysing large, complex datasets across a wide range of projects and clients.
Our experience includes:
- Bulk Purchase Annuity (BPA) transactions: Supporting over 100 transactions, involving the validation, cleansing and transformation of large and complex policy datasets under demanding timelines.
- Continuous Mortality Investigation (CMI) work: Managing claims and death data from multiple large insurers, often received in different formats and structures, and transforming it into consistent datasets suitable for analysis and reporting.
- Tripartite data for SCR calculations: Processing, checking and cleaning Tripartite asset data from multiple asset managers, including State Street, into formats for use in Standard Formula SCR calculations.
- Data transformation for outsourced roles: Carrying out over [X] outsourced roles – each one required building a data solution at outset, and then operating it for each reporting cycle. ### 3.2.1 Benefits to you
Our proven track record means we can rapidly identify data quality issues, efficiently transform data into the required format, and provide confidence that the data underpinning actuarial calculations is complete, accurate and fit for purpose.
3.3 Our capabilities
As actuaries, we understand the data requirements that underpin reserving, capital, pricing and financial reporting. Our combination of actuarial expertise and software development capability enables us to translate complex actuarial and feature modelling requirements into robust, scalable data solutions.
Our capabilities include:
- Integrating with policy administration and source systems through APIs.
- Performing automated data quality testing, reconciliation and validation checks.
- Transforming data into bespoke formats tailored to actuarial, reporting and modelling requirements.
- Maintaining comprehensive audit trails of all transformations, adjustments and validation outcomes.
- Developing repeatable, controlled processes that reduce reliance on manual intervention while preserving actuarial oversight. ### 3.3.1 Benefits to you
By combining actuarial expertise with modern data processing tools, we can reduce manual effort, improve consistency, and accelerate data delivery. While robust controls, auditability and reproducibility ensure full transparency and auditability throughout the process and reduces the risk of downstream modelling and reporting errors.
3.4 Proposed solution
We will build you a customised python based solution that will run on Chesnara’s chosen platform. It will perform to Chesnara’s governance standards, and will be designed from the ground-up to be fast and automated to meet the one-working-day timeline.
We will run a series of structured workshop with Chesnara to scope and specify the requirements.
| Topic | Objective |
|---|---|
| Source extracts & schedules | Understand extract data structure, delivery timing, and ownership of extract production |
| Mapping rules & product taxonomy | How policy admin data, reinsurance data, assumptions data and asset data need to be mapped and combined to produce model ready inputs. |
| Data rules | Define the data quality rules, checks and tests that need to be carried out on data sources, including reconciliations to precious extracts. |
| Review process | Agreeing the data reports structure and contents produced, and how data issues are updated/ corrected and documented. |
That data solution build process will consist of:
- Establishing overall architecture: A version control repository, setting up the database objects, setting up the ability to carry out automated testing.
- Building pipelines: Code the engine of the data solution. This will read data extracts, carry out checks, transform to create model inputs, and saved outputs (including data reports) to database.
- Testing: We will construct test data to test the edge cases, and ensure data checks are working and identify the planted errors.
- Dry runs: We will carry out dry runs to test end-to-end process. Where needed, we will make improvements to the process.
- Documentation: We will produce full documentation of the data solution. The documentation will cover technical details of the data solution, operating the data solution for the data in scope here, and the principles on how to extend the data solution to work with future acquired business. ### 3.4.1 Benefits to you
You receive a solution designed around your business rather than adapting your processes to fit a pre-defined tool. This ensures the resulting data is immediately usable, while creating a scalable framework that can support future developments with minimal rework.
3.5 Ongoing support operating data solution
Our data solution is designed to be owned and run on Chesnara systems. Therefore, we recommend that ongoing resource to support operation is provided on a regular secondment basis.
We will agree ownership of the signoff process for the outputs of the data solution. Our expectation is that our team can run the data solution for Chesnara, however, the signoff of outputs remains within a valuation manager within Chesnara.
The timing of the secondments will coincide with the timing of when data can be provided. The length of the secondments are expected to be one working day for each data source, however, this may need to be extended if there are data errors that require more than a trivial remediation.
3.6 Codelivery
We will look to use available Chesnara team members throughout the process.
We will have significant discussions with Chesnara during the scoping and workshop phase of the project.
Our expectation is that as Chesnara recruits its anticipated BAU team, we can include them in the dry-run and documentation parts of the process. We can also involve them whilst we carry out ongoing operation of the data solution.
3.6.1 Benefits to you
Our approach will provide a suitable and hands-on way to transfer knowledge, and enable Chesnara personnel to own the operation in the longer run.
4. Actuarial Model
4.1 Scope
The modelling solution will:
- Allow all modelling to be performed within Chesnara
- model all business being transferred
- be capable of modelling financial options and guarantees
- Produce outputs for downstream reporting ## 4.2 Our credentials
We have significant building, migrating, and validating valuation models.
Our experience includes:
- Bulk Purchase Annuity (BPA) pricing model: Validated the pricing model, for a new entrant, including the complex underlying benefit structures.
- Simplified ESG and COG calculation model: Simplified market consistent ESG for five asset classes (cash, equity, property, gilts and credit), alongside an accompanying COG calculation model. Contains logic for investment and bonus management actions. Capable of running 10,000 simulations for a portfolio of 50,000 policies on a laptop in under 30 minutes.
- Migrated from vendor platform to python (See Appendix A): Carried out for a suite of valuation models covering 27 products for a global life insurer in c3 months and no audit queries. Since then, we have also migrated the models we have for our [X] outsourcing roles over to Python.
- Rebuilding business planning models: Methodology review and rebuild of business planning model achieving 1x order of magnitude faster run times.
- Standard formula calculator: For our outsourced roles, we have developed SIIMPLIFY, a Standard Formula calculator. Coverage of all Standard Formula risk modules, in a fully documented tool. ### 4.2.1 Benefits to you
You will benefit from our tried and tested model migration approach, adapted to fit your delivery plan.
Your final model will be accurate, fast, and meet governance standards.
4.3 Our capabilities
We are a team of actuaries supported by software development expertise. We understand valuation methodologies, and what is proportionate simplification. We have a strong knowledge and understanding of actuarial modelling standards, model governance, in insurers and insurance audit requirements. Our models are built on software engineering principles.
The business being transferred is currently modelled using Prophet. Our team have experience of developing, testing and using Prophet models before.
4.3.1 Benefits to you
Our delivery approach provides a robust, efficient, and scalable actuarial modelling solution that enhances reporting capability, reduces operational risk, strengthens governance, and develops in-house capability through targeted model and Python training. This ensures you not only a successful implementation but also the skills, knowledge, and documentation required to confidently own, operate, and evolve the solution in the future.
More specifically, you will benefit from:
- Strong actuarial expertise giving review of your feature specifications and translating into detailed model specs. Actuarial judgement and expertise in this process means that proportionality and materiality will be applied, and any errors will not be blindly built into the model.
- You will benefit from our Prophet expertise when we are reconciling the model we build to the Prophet models. Our expertise will enable our team to potentially narrow down and understand potential causes of differences more quickly.
- You will benefit from our models being faster, more flexible and more scalable by being build on software engineering principles. ## 4.4 Proposed solution
We will build you a customised python based model that will run on Chesnara’s chosen platform. It will perform to Chesnara’s governance standards, and will be designed from the ground-up to be fast, easy to use, and produce required outputs.
4.4.1 Key features of our model will be
| Area | Description | Benefit |
|---|---|---|
| Run controller | We provide a user-friendly Run controller that enables user to configure and execute model runs without accessing the underlying Python code. | The GUI provides an intuitive interface requiring minimal technical expertise, that allows for faster, more consistent execution of model runs and focussed results. |
| Model organisation | Our Python models are organised hierarchically to promote code reusability, consistency and maintainability. Shared functionality is centralised within common parent components, while product-specific requirements are implemented only where needed. | Improved consistency across products and modelling approaches with reduced development effort through reuse of functionality. Faster implantation of updates and regulatory changes with updates applied automatically across related products. |
| Output structure | The Python models developed can allow for a range of output formats to aid with the creation and analysis of the required regulatory submissions. Users can select via the Run Controller GUI output options to tailor the results to their needs. To ensure transparency and reproducibility, each run automatically produces a comprehensive audit trail. | Full transparency and traceability of model runs with reduced risk of inconsistencies when recreating historical outputs. Improved governance and audit readiness with reproducible results that support regulatory compliance. |
4.4.2 Scoping and specification
We understand that Chesnara is already carrying out significant work to review product features and to write feature specifications.
We will work closely with you to turn these feature specifications into model specifications. When doing this, we will discuss the features with you and provide expert actuarial input into areas such as anticipated materiality, and modelling.
4.4.3 Modelling of options and guarantees (including management actions)
We understand that you do not currently have access to an ESG, and it is out of scope for us to build you one. Furthermore, we understand that the with profits business, will still be managed as part of Scottish Widow’s With Profits Fund. This means that Chesnara will have to follow the management actions of that fund, without visibility of them.
We will discuss the modelling of this with you during the scoping phase of the project. We expect that the modelling solution will:
- Have full deterministic capability to calculate technical provisions without the cost of any guarantees or options.
- Have a switch to calculate the cost of guarantees and options using either
- closed form method
- a stochastic engine
- Will not model management actions, but the modelling framework will be such that it will enable adding bonus related managed actions in at a future point in time. #### 4.4.3.1 Benefits for you
Your model solution will be ready for use now, and can be easily extended in the future.
4.4.4 Access to Prophet and test plan
We understand that we will not have direct access to the existing underlying Prophet models. We understand that from October 2026, there may be some access to final model outputs and intermediate model outputs.
Our working assumption is that from October 2026, we will have access to the following outputs for two different reporting dates:
Full aggregate outputs that is currently pushed to downstream reporting processes.
Product level aggregated PV variables, in base and in standard formula stresses.
Policy level cashflows in base and in standard formula stresses, for a small sample of policies only. We will chose the policies to cover a wide range of product features. Once we understand what information will be available for us to test our model solution against, we will discuss and agree a test plan. This test plan will cover:
Unit testing of individual code parts with dummy data to confirm functionality with edge cases.
Replication of product level results in base and in standard formula stresses for two different balance sheet dates, to an agreed tolerance.
Replication of the policy level cashflows for the sample policies only in base and in standard formula stresses for two different balance sheet dates to an agreed tolerance
Replication of a smaller sample of policy level cashflows to a rudimentary Excel based baseline model. We understand that Chesnara is in the process of changing its auditors. When designing the test plan, we will engage with them early to confirm it meets their expectations.
4.4.4.1 Benefits to you
Your model build will meet appropriate governance standards, and the model build process will be designed to mitigate the risk of late audit challenge and late audit queries.
4.5 Codelivery
Throughout model build, testing and reconciliation, we will hold regular working sessions with Chesnara stakeholders to review methodology decisions, assumptions, modelling simplifications and reconciliation results. Team members will have visibility of the model architecture, controls framework, calculation engine and output structure, enabling them to develop a detailed understanding of the solution as it is built.
During validation and parallel-run activities, Chesnara personnel will work alongside our team in reviewing results, investigating differences, assessing materiality and approving outcomes. This will build the internal capability required at Chesnara to support future reporting cycles.
Knowledge transfer will be delivered through a combination of working sessions, formal training and detailed documentation. Training will cover model architecture, operation of the run controller, governance processes, testing approaches, model maintenance and Python fundamentals. The objective is not simply to hand over a model, but to equip Chesnara with the capability to understand, maintain and extend it confidently in the future.
4.5.1 Benefits to you
- Rapid development of in-house modelling capability.
- Greater transparency and understanding of model methodology and assumptions.
- Reduced operational risk.
- A smoother transition from implementation into business-as-usual reporting.
- Internal capability to make future model developments if required, without the need for external support.
5. Implementation Plan
We understand that the key driver is to have a fully validated actuarial data and modelling solution capable of supporting parallel reporting by June 2027, in advance of the end of the Lloyds TSA arrangements on 1 October 2027. Appointment is assumed from October 2026. Our implementation approach is designed around progressive delivery, frequent validation, close co-delivery with Chesnara’s developing actuarial team, and early engagement with incoming auditors.
The programme is structured around four integrated workstreams:
- Programme & Governance
- Data Solution
- Actuarial Model
- Testing, Validation & Knowledge Transfer
This allows data and modelling activities to progress in parallel while maintaining appropriate governance and control. Approved data inputs gate credible model parallel runs — we will not treat June 2027 as secured until data approval gates and the agreed model test plan have been met.
5.1 Delivery principles
Our implementation approach is based on the following principles:
- Gated specification before large-scale build — signed data and model baselines first.
- Incremental delivery and validation rather than a “big bang” approach.
- Continuous involvement of Chesnara personnel through a co-delivery model.
- Progressive documentation and knowledge transfer throughout the programme.
- Multiple production-style valuation runs prior to the June 2027 parallel run.
- Time-and-materials milestones with transparent actuals versus estimate.
5.2 Indicative timetable
The timetable below is indicative and will be baselined at mobilisation. Half-year and full-year 2026 exercises will be used as live-like dry runs where extracts allow. Prophet output reconciliations start from October 2026 where Lloyds Banking Group (LBG) can provide comparable results.
| Phase | Timing | Outcomes |
|---|---|---|
| Mobilisation | October 2026 | Governance, environment access, RAID log, milestone plan, workshop schedule |
| Scoping and specification (gated) | October – November 2026 | Signed data and model baselines; mapping / DQ catalogues; tolerances; Prophet comparison approach; management-action / SWL bonus boundary |
| Data solution build | November 2026 – January 2027 | Pipelines, automated DQ, mock then live-like dry runs, documentation |
| Core deterministic models | December 2026 – March 2027 | Non-profit unit-linked and with-profit deterministic builds |
| Regulatory frameworks | February – April 2027 | Solvency II (full Standard Formula SCR), Lux-GAAP / CAA, AoC / AoS model outputs |
| COG and hardening | March – May 2027 | COG switch (deterministic / Black–Scholes / stochastic hook-up); performance; controls |
| Data operating secondments | Through June 2027 | Howden-led working-day-one style operation and transfer to Chesnara |
| Parallel run | To June 2027 | Full test parallel; Prophet reconciliations where available; defect clearance; runbooks |
| Training and handover | May – July 2027 | Run controller training; Python knowledge transfer; final documentation |
| Post go-live support | 12 months from go-live | Discounted support and training queries (see Commercial model) |
5.3 Delivery path
We will use a tried and tested model migration and build approach, adapted to fit your delivery plan and applied consistently across the data solution and actuarial model. The same gated rhythm — set up, scope, build, test, train and complete — underpins both workstreams. Large-scale product build starts only after signed specification baselines, other than technical spikes and CI/CD scaffolding.
| Stage | Data solution | Actuarial model |
|---|---|---|
| Setup | Repository, database objects, CI/CD gates | Model hierarchy, run-controller concept, test harness |
| Scope | Mappings, quality rules, reconciliations, sign-off policy | Feature challenge; model specifications; COG / SWL boundary |
| Build | Ingest → check → transform → model-input pipelines | Core engine; NP UL; WP; SII / Lux-GAAP; COG switch; outputs |
| Test | Planted-error packs; mock and live-like dry runs | Unit, product, Prophet compare, UAT, parallel evidence |
| Train & complete | Runbooks; shadowing into BAU operation | Run controller training; documentation; handover |
When we have used this approach previously, we have delivered on-time, fully tested models with zero audit queries — demonstrating the robustness of our development approach, governance, and experience of working with auditors and understanding their requirements.
5.4 Training
Training runs through the programme, not only at the end: data and model architecture; data flows, assumptions and reporting outputs; operation of the data solution and the model run controller for business-as-usual; governance and testing approaches; model maintenance; and Python knowledge-transfer workshops. Materials include user guides, runbooks and reference documentation, with post-training support into the first live cycles.
5.5 Post go-live support
For 12 months from agreed go-live, Howden provides support at discounted time-and-materials rates for support and training queries (see Commercial model). Material change requests remain at standard rates unless otherwise agreed. The objective is a stable first year of Chesnara-operated reporting — not a multi-year managed service.
5.5.1 What this means for Chesnara
- A single critical path to June 2027 parallel reporting, with data gates protecting model credibility.
- Early specification that reduces rework and protects timetable certainty.
- Visible milestones you can track, challenge and steer.
- Hands-on secondments and training that transfer operating capability before TSA exit.
6. Resource Plan
This section sets out who does the work, how Chesnara and Howden co-deliver day to day, the operating secondments that protect working-day-one model points, and the indicative effort that feeds the Commercial model.
6.1 Howden team
Delivery is actuary-led with dedicated engineering support. There is no separate project manager — delivery management sits with the Project Lead. Surge capacity is available from the 80+ Howden IAL bench if unforeseen issues arise. Amit Lad and Kim Durniat remain your primary senior contacts.
| Role | Focus |
|---|---|
| Guardian Partner — Kim Durniat | Partner accountability, peer review, escalation |
| Project Lead — Amit Lad | Overall delivery, client management, plan / risks / milestones (no separate PM) |
| Data Lead — James Hadley | Pipelines, data quality, dry runs, operating secondments through parallel |
| Model Lead — Sam Underhill | Model specifications, product builds, Solvency II / Lux-GAAP frameworks |
| Lead Developer — Allan Engelhardt | Python architecture, CI/CD, run controller, deployment onto Chesnara platform |
| Testing Lead — Andrew Mason | Independent testing design, Prophet reconciliations, evidence packs |
| Consultants / Analysts — as required | Specification challenge, build support, investigation, documentation |
6.2 Co-delivery
We deliver with Chesnara’s team, not around them. Co-delivery is how Chesnara builds lasting ownership before TSA exit — and how Howden hours reduce when Chesnara takes on agreed tasks (see Commercial model).
- Specification workshops — Chesnara shapes and challenges mappings, data rules, feature specifications and modelling simplifications.
- Validation — joint review of data reports, exception packs, model results and materiality.
- Sign-off — Chesnara retains approval of data-solution outputs (typically a valuation manager) and user acceptance of the model.
- Progressive ownership — as your business-as-usual team is recruited, they shadow dry runs, documentation and data secondments and take over operation.
Illustrative Chesnara effort is around 0.5–1.0 full-time equivalent across the programme, peaking at user acceptance testing and the June 2027 parallel run. Greater Chesnara contribution reduces Howden time-and-materials burn (see Commercial model).
6.3 Data operating secondments
As set out under the Data Solution, the platform is designed to be owned and run on Chesnara systems. Through implementation and into the parallel window we recommend regular secondments timed to extract availability. Typical duration is around one working day per data source per cycle, extended only where remediation is non-trivial. Howden can operate the pipelines; Chesnara signs off outputs. The objective is business-as-usual ownership before TSA exit, preserving working-day-one model-point production.
6.4 Indicative effort
| Workstream | Howden hours | Approx. days (8-hour) |
|---|---|---|
| Data solution (build, dry runs, documentation and operating secondments through parallel) | 1,044 | ~131 |
| Actuarial modelling (specification, build, testing, training and handover) | 2,992 | ~374 |
| Combined programme | 4,036 | ~505 |
Estimates assume moderate Chesnara co-delivery and a manageable closed-book unit-linked + with-profit product set. They will be refined after scoping workshops confirm product count, extract complexity and the final test plan. Hours are managed on a time-and-materials basis against milestone estimates, with transparent actuals versus estimate.
Indicative Howden hours by grade (combined programme):
| Grade | Hours |
|---|---|
| Guardian Partner (Kim Durniat) | 112 |
| Project Lead (Amit Lad) | 328 |
| Data Lead (James Hadley) | 320 |
| Model Lead (Sam Underhill) | 680 |
| Actuarial Consultant | 1,000 |
| Actuarial Analyst | 596 |
| Software Developer (Allan Engelhardt + support) | 712 |
| Testing Lead (Andrew Mason) | 288 |
| Total | 4,036 |
Milestone-level hour schedules for the data solution and actuarial model are available on request and will be attached to the commercial schedule once day rates are agreed. Higher Chesnara contribution reduces Howden burn.
6.4.1 What this means for Chesnara
- Clear roles and a defined co-delivery model — you are not a spectator to your own build.
- Bench strength behind a named senior team if unforeseen issues arise.
- A commercial incentive to grow internal capability early and reduce external burn.
7. Commercial model
A time-and-materials build of a Chesnara-owned Python estate — not a multi-year platform licence. The commercial model matches how we deliver: transparent milestone estimates, an incentive for greater Chesnara involvement, and a clear boundary between implementation and ongoing software licensing.
7.1 Commercial structure
- Time and materials for the build and implementation phase, organised into milestones with estimated hours (see Resource Plan).
- Engagement covers data solution and actuarial model build, testing, training, knowledge transfer, and agreed operating secondments through the parallel window — not a multi-year platform licence.
- On payment for delivered work, Chesnara owns all code and intellectual property.
- No ongoing software licence fees. Infrastructure and hosting costs sit with Chesnara as host of the solution.
- Invoicing in pounds sterling (GBP), billed in the United Kingdom.
- Actual hours versus estimate reported transparently against each milestone.
7.2 Commercial terms
| Term | Proposal |
|---|---|
| Commercial form | Time and materials by milestone, with estimated hours (Resource Plan) |
| Scope of fees | Data + model build, testing, training, parallel-run support, and data operating secondments |
| Intellectual property | Chesnara owns all code and IP on payment for delivered work |
| Software licence | None ongoing; Chesnara hosts infrastructure on its platform of choice |
| Billing | GBP, invoiced in the United Kingdom |
| Cost control | Actuals versus estimate reported for each milestone |
| Combined programme | 5% reduction on Howden hours (or equivalent fee credit) versus standalone data + model estimates |
| Chesnara resource incentive | Higher Chesnara delivery of agreed tasks → lower Howden burn / effort credits on remaining milestones |
| Post go-live support (12 months) | Support and training queries at a proposed 20% discount to standard grade rates |
| Change after baseline sign-off | Standard time-and-materials rates unless otherwise agreed |
7.3 Incentive for Chesnara resource
Where Chesnara takes a larger share of delivery, Howden hours reduce. We will agree a baseline estimate per milestone, report actuals versus estimate, and reflect higher Chesnara contribution as lower Howden burn — with an effort credit on remaining milestones where agreed tasks are consistently delivered by Chesnara.
7.4 Combined data and modelling engagement
This proposal covers the data solution and actuarial modelling together. We propose a 5% reduction on Howden hours (or equivalent fee credit) relative to the sum of standalone workstream estimates, reflecting shared governance and overlapping testing.
7.5 Fees
Fee tables will be completed by applying agreed day rates to the hour schedules in the Resource Plan (programme total 4,036 Howden hours before any combined-programme credit). A standard rate card and a discounted post go-live support rate card will accompany the commercial schedule. Until rates are fixed, this is a structured time-and-materials offer — not a fixed-price bid.
7.6 Post go-live support
For 12 months from agreed go-live, support and training queries are charged at a proposed 20% discount to standard grade rates. Material change requests remain at standard rates unless bundled. This is support for a Chesnara-owned estate — not a managed-service licence term.
7.7 Value included during build
Within reasonable time-and-materials materiality during the build: actuarial challenge of product specifications and materiality; audit-oriented controls and documentation; knowledge transfer; and design hooks for future stochastic depth, asset modelling and further books.
Benefit for Chesnara: you buy delivery and ownership, keep commercial control through transparent time-and-materials reporting, and earn a discount for building your own capability early — without locking into ongoing software licence fees.
8. Appendix A: About Howden
8.1 Howden Group
Howden is a global insurance group with expertise across insurance, reinsurance, risk consulting, and employee benefits. With over 25,000 employees in more than 115 countries, we combine global reach with deep local insight.
Our scale and specialist capabilities allow us to support clients across all areas of insurance and risk management, from placement and structuring to capital optimisation and strategic advice.
Our collaborative model ensures access to a broad network of expertise, delivered through a responsive and client-focused approach.
8.2 Howden Insurance Actuarial & Longevity
Howden Insurance Actuarial & Longevity (IAL) was created through the merger of the Barnett Waddingham and Hymans Robertson insurance actuarial teams. Today, the team brings together 80+ specialists across general insurance, life insurance, and longevity.
We offer a full range of professional services to insurance companies, captives and mutuals, including:
- Provision of a Chief Actuary (UK SMF-20)
- Full outsource actuarial function, covering pricing and reserving
- Capital Modelling, Management & Validation for internal models
- Risk and assurance services,
- Regulatory and investment advice,
- Process transformation The Howden IAL team operates separately of any Howden broking or placement relationships with you. We maintain independence from our Howden colleagues which enables us to provide you with expert and impartial advice. Strong data governance and information security protections are in place, as demonstrated by our QAS and ISO certifications.
9. Appendix B: Case Study
9.1 Python model development
Our Insurance and Longevity Consulting team specialises in developing complex, bespoke financial models using Python, tailored to meet specific client requirements. A summary of a recent case study is set out below.
9.1.1 Requirements
A global insurance firm needed to migrate their life valuation systems to an open-source platform to increase the ability to maintain and further develop the model, whilst avoiding sizeable licence fees. The system needed to efficiently model cashflows for multiple reporting regimes and include 29 separate insurance products such as:
- Term Assurance
- Whole of Life
- Critical Illness
- Short-term credit Services Provided:
In three months, BW delivered a fully reconciled, tested and documented Python-based valuation system. This includes a user-friendly no-code interface, reducing potential errors and saving time by only exposing the options required at runtime.
The system takes inputs and produces outputs in identical formats to the previous system, allowing for maximum compatibility with the firm’s existing database and Excel driven out-of-model processes.
Additional output options were created to improve upon existing systems and provide new flexibility to produce both high-level summaries and detailed cashflow projects on request.
The models were approved for use for the year-end valuation process by the firm’s Audit Committee and underwent external audit without requiring any amendments.
9.1.2 Benefits to client
The client now has ownership of a modern, maintainable and flexible system with no ongoing licence fees.
Separation of the orchestration layer from the calculation layer means that actuaries, even with limited prior knowledge of python, are easily able to understand the calculation methodology and make precise, tested, and documented changes to this when required.
The time taken to set up and run the models has been reduced due to the graphical user interface and refinements to the processes.
One Creechurch Place, London, EC3A 5AFT +44 (0)20 7623 3806E info@howdengroup.comwww.howdengroup.comHowden is a trading name of Howden Insurance Brokers Limited, part of Howden Group Holdings. Howden Insurance Brokers Limited is authorised and regulated by the Financial Conduct Authority in respect of general insurance business. Registered in England and Wales under company registration number 725875. Registered Office: One Creechurch Place, London, EC3A 5AF. Calls may be monitored and recorded for quality assurance purposes. 03/24 Ref: 1375One Creechurch Place, London, EC3A 5AFT +44 (0)20 7623 3806E info@howdengroup.comwww.howdengroup.comHowden is a trading name of Howden Insurance Brokers Limited, part of Howden Group Holdings. Howden Insurance Brokers Limited is authorised and regulated by the Financial Conduct Authority in respect of general insurance business. Registered in England and Wales under company registration number 725875. Registered Office: One Creechurch Place, London, EC3A 5AF. Calls may be monitored and recorded for quality assurance purposes. 03/24 Ref: 1375