Prior Concept Cancellation Record — Harborline Buffer Series

Programme timeline · status 16 Oct 2026Read the full story →

Harborline

Aug 2025

Cancelled

Gate 0

Feb 2026

Stage 1

Business case

Gate 1

Apr 2026

Recycled

Gate 1

Jun 2026

Go w/ conditions

Stage 2

Development

You are here

Gate 2

Apr 2027

Gate 3

Oct 2027

Gate 4

Feb 2028

Launch

Mar 2028

Gate 5

Sep 2028

Cancelled 07 August 2025

Lighthouse Financial Services Company — The record of a product development programme this Gate Review Board cancelled. The Harborline Buffer Series, a registered index-linked annuity concept, passed Gate 0 in March 2025 and was stopped at Gate 1 on 07 August 2025. This document records the grounds, the vote, the execution of the cancellation protocol and the final accounting. It exists because a governance record containing only approvals is a record of nothing.

This is a separate, closed programme. Every figure in this document belongs to Harborline. None of it is part of, netted against, or rolled up into the Beacon Index Advantage programme. The two programmes share a board, a chair and a lesson — they share no funding, no schedule and no team. See §7.

Programme	Harborline Buffer Series — registered index-linked annuity (RILA) concept
Gate 0	06 March 2025 — GO
Gate 1	07 August 2025 — CANCEL, carried 1-4
Funding released	$1,140,000 (Stage 1 only)
Funding spent	$868,000
Returned to capital	$272,000
Final accounting filed	04 September 2025
Chair	C. Tyrrell, NPD Programme Manager (non-voting)

1. The concept

Harborline was a registered index-linked annuity: a contract crediting index-linked interest with a defined buffer absorbing a first tranche of index loss, and the contract owner exposed to loss beyond it. That downside participation is what makes a RILA a registered product. It brings the contract inside securities registration and brings its distribution inside broker-dealer licensing.

The strategic case was straightforward and, on the demand side, correct: RILA sales across the industry were growing faster than fixed indexed annuity sales, the carrier had no participant in that segment, and distribution partners were asking for one.

2. Why it was stopped

Three grounds were recorded. The first two are causes; the third is their consequence.

Ground 1 — The registration and distribution burden was not carried by the case

A registered product requires securities registration, prospectus delivery, and distribution through broker-dealer channels with licensing and supervision obligations the carrier's existing independent-agent distribution does not carry. Stage 1 costed that burden properly for the first time — which is what Stage 1 is for — and it was materially larger in both cost and elapsed time than the Gate 0 screening had assumed.

The Gate 0 assessment had scored feasibility on the carrier's ability to build the product. It had not adequately weighted the carrier's ability to distribute one through a channel it did not operate in.

Ground 2 — The buffered structure exceeded in-house hedging capability

Hedging a buffer requires an options strategy materially different from the capped-upside structures the Investments desk runs against the existing annuity block. The assessment concluded the capability could be acquired, but only by outsourcing execution — which contradicted the carrier's ALM operating model and added recurring cost to a product whose margin was already under pressure from Ground 1.

Recorded at the time: a guarantee the organisation cannot hedge with its own hands is a guarantee it does not fully control.

Ground 3 — Under Grounds 1 and 2 the product did not clear the hurdle

Loaded with the registration and distribution costs and the outsourced hedging cost, the projected return fell below the carrier's hurdle rate. No repriced structure tested during Stage 1 cleared it while remaining competitive on buffer level and cap.

The Board did not treat this as a pricing problem to be solved in Stage 2. It treated it as the arithmetic consequence of two structural facts that Stage 2 spending would not change.

3. The vote

Seat	Vote	Recorded position
Chief Product Officer — Executive Sponsor	Cancel	Segment remains strategically attractive; this structure is not the way in
Chief Actuary	Cancel	No tested structure clears the hurdle at a competitive buffer and cap
General Counsel & Chief Compliance Officer	Cancel	Registration and supervision obligations exceed the compliance operating model
Chief Financial Officer	Cancel	Capital better deployed against a product the carrier can distribute today
Head of Distribution	Continue	Channel demand is real and unmet; recommended re-scoping rather than cancelling
Carried 1-4		Cancel

The dissenting vote was right about the demand and wrong about the remedy — and it shaped what came next. Distribution's position was that the appetite was genuine and the carrier should find a way to serve it. The Board accepted the first half. What it declined to accept was that re-scoping this programme was the way, since neither the registration burden nor the hedging constraint was a matter of scope. The successor concept serves the same appetite from the other side of the registration line.

4. Execution of the cancellation protocol

Governance Model §13 sets out seven steps on a cancel. All seven were executed and are recorded here.

Funding stopped immediately. No Stage 2 tranche was ever released. No further commitments were entered after 07 August 2025.

Final accounting filed 04 September 2025, within twenty business days, to the CFO and the Executive Committee. See §5.

External obligations settled, not abandoned. Two engagements were live — outside securities counsel and an external hedging capability assessment. Both were terminated for convenience under their existing terms and paid to the point of termination.

Team released on a stated date with reassignment handled through functional leads. Fourteen people were engaged at least part-time; all were reassigned within the carrier.

Work product preserved and indexed — see §6. This is the step most often skipped, and skipping it is how an organisation pays twice for the same research.

This cancellation record filed, stating the decision, the vote, the rationale and the final accounting.

Re-entry permitted only at Gate 0, with a new business case and new funding. Applied — see §7.

5. Final accounting

Concept and product design - structure, buffer levels, rider	$284,000
Actuarial pricing and capital modelling	$246,000
Registration and distribution licensing assessment (external counsel)	$198,000
Hedging capability assessment	$88,000
Program management	$52,000
Total spent	$868,000
Stage 1 tranche released at Gate 0	$1,140,000
Returned to the capital pool	$272,000

The single largest line is the registration and distribution licensing assessment. That is the correct shape for a Stage 1 that did its job: the money went into establishing the fact that cancelled the programme, and it was spent before the far larger development tranche was committed.

$272,000 was returned rather than absorbed. Under a conventional single-baseline funding model the unspent balance would have sat inside an approved programme budget and been consumed by whatever the programme did next. Tranche-based release is what makes the return mechanical rather than a matter of anyone's discipline.

6. Work product preserved

Indexed and retained under step 5, available to any subsequent concept on the same terms as any other internal research:

Market sizing and segment analysis for index-linked products.
Distribution appetite research, including the partner conversations behind Distribution's dissenting vote.
The registration and licensing assessment — the most durable output, and the one that made the successor's Gate 0 case straightforward to frame.
The hedging capability assessment, including what the desk can and cannot execute in-house.
Actuarial pricing models for index-linked structures, with assumptions documented and dated.

What was not preserved: the programme. No team, no funding, no schedule and no approvals carried forward. Preserved research is an input available to anyone; it is not a running programme in suspension.

7. Why the successor is a new programme and not a restart

The zombie project. The characteristic pathology of stage-gate governance is the cancelled programme that returns under a new name, resumes where it stopped, and keeps its old funding and its old team. Where that happens the cancel was theatre — and every subsequent decision by that board is worth less for it.

The distinction is procedural and auditable, and it is set out here so that it can be checked rather than asserted.

Test	Harborline	Beacon Index Advantage
Programme status	Closed 07 Aug 2025; final accounting filed	New programme, screened at Gate 0 on 05 Feb 2026
Funding	Cancelled; $272,000 returned	New Stage 1 tranche earned at its own Gate 0
Entry point	—	Gate 0, not Gate 1. Nothing was resumed
Business case	Rejected at Gate 1	Written new; recycled once at its own Gate 1 before carrying
Team	Released and reassigned	Staffed independently
Product structure	Registered, buffered, downside participation	Non-registered, capped, principal protected — a different product
Elapsed between them	182 days between the cancel and the successor's Gate 0

The two programmes are connected by a lesson, not by a ledger. Harborline established that the registration burden — not the product idea, not the market — was the binding constraint. Beacon Index Advantage is a fixed indexed annuity precisely because of that finding, which is recorded as decision D-01. The successor serves the same demand from the non-registered side of the line, where the carrier's existing distribution, compliance model and hedging desk all already operate.

Stated plainly: this programme exists in the form it does because the Board cancelled the previous version of it.

8. What the Board changed as a result

Finding	Change
Gate 0 scored feasibility on the ability to build, not to distribute	Gate 0 must-meet criteria now include that the concept does not require a capability the organisation has already decided not to build; distribution appetite became a scored criterion in its own right
Hedging capability was assessed late in Stage 1	Hedgeability with in-house capability became a Gate 1 must-meet on the successor programme, certified by name
The strategic case outran the economics for two quarters	Cancellation thresholds are now published in advance for each gate rather than debated in the room (Governance Model §13.1)

These are the reason the successor's Gate 1 must-meet list reads as it does. A governance model that produces the same failure twice has not learned anything from producing it once.

9. Document control

Version	Date	Change
1.0	07 Aug 2025	Filed on the day of the decision. Grounds, vote and protocol steps 1, 3, 4 and 7.
1.1	04 Sep 2025	Final accounting added (§5); work product index completed (§6). Record closed.
1.2	05 Feb 2026	Reopened once, on the successor's Gate 0, to add §7 and §8 — the new-programme test and the governance changes. No figures altered.

Filed by C. Tyrrell, NPD Programme Manager and Chair of the Gate Review Board. Related: Stage-Gate Governance Model §13 and §18 · Gate Decision Framework §7.