Fund launch management that catches every slip the night it happens and holds first NAV
First NAV on the planned date: every slip caught early, re-planned and signed off.
See one case, screen by screen ↓Fund Launch Management is fund launch software that takes a mutual fund, ETF, share class or UCITS sub-fund from the product committee's decision to first NAV. Each launch is a row of dependency dominoes — board, registration, service providers, exchange listing, seed and operations — and seven agents plan the critical path, draft the board resolutions and filings, run the creation basket test files, chase the providers and catch a slip the night it happens. Launch PMs propose, the Head of Product approves, counsel signs filings and the treasurer co-signs seed moves.
AI in asset management: a buyer's guide to agentic AI for investment managers →
Updated
A fund launch is 30 to 60 dependencies, and most of them belong to someone outside the product team: the trust's board and its Section 15(c) approval, fund counsel, the SEC's 75-day review of a Rule 485(a) filing, the custodian, the fund administrator and transfer agent, the exchange, the CUSIP bureau, the authorized participants, the lead market maker and the seed desk. Today the fund launch process runs on a spreadsheet tracker, a weekly status call and hundreds of emails. The tracker knows a basket test is amber. It does not know that amber blocks the AP creation test, the seed wire and first NAV.
So a late provider step is found when a later date is missed. A custodian that was never asked to set up three new-issue municipal bonds rejects the creation basket test file, and the AP creation test, the seed wire and first NAV all wait on it. Every month a launch reaches first NAV late is a month without fees on its seed and early assets — about $21,000 a month on a $50 million launch at 50 basis points, by our estimate. And every gate is a control, with its evidence sitting in an email somewhere.
Estimated split for a launch project manager running nine launches, by hand and with the solution.
Seven agents plan, draft, test and chase every launch to first NAV; launch PMs propose, the Head of Product approves.
Quillfield Short Duration Muni ETF (QSDM) is an actively managed, transparent ETF in Quillfield ETF Trust, relying on Rule 6c-11. First NAV is set for Tuesday, March 16, 2027, with $25.0 million of seed. On the evening of February 22 the custodian rejects its basket test file. Here is what happens next, screen by screen, in the working solution.
Nadia's Today screen shows 24 launches in flight — 19 on track, 3 at risk, 2 with a regulator — 7 decisions waiting for her, 11 days recovered by the agents this month and 146 provider tasks open, 11 past SLA. First under “Needs you”: the Short Duration Muni ETF, approve a re-plan or first NAV moves. Beside it, each agent's overnight work in plain words, with its version, its confidence and a link to its source.
Launch Conductor: “Overnight: 1 domino tipped, the QSDM basket test file. The agents recovered 4 of 6 days by 8:36 a.m.; 2 need your decision before the seed desk's 10:00 a.m. cut-off.”
The launch is one row of dominoes in six bands — Board, Registration, Service providers, Exchange listing, Seed, Operations — running to a fixed First NAV. Each names its owner, an agent or a person, and its date. When the basket test file failed, the Conductor tipped it and leaned every domino it knocks over with its new date: the AP creation test from Feb 24 to Mar 4, the seed wire and seed portfolio from Mar 2 to Mar 10, both NAV dry runs, and first NAV from Mar 16 to Mar 24.
“Basket test tipped at 6:51 p.m.: 6 business days at risk · first NAV would move to Mar 24.”
The custodian rejected 3 of 58 lines in test file 2, “security not found”: Lake Avalon Public Utility District 2027A, Calder Bay Port Authority 2027 and Westbrook County School District No. 4 2027B. The Basket agent found all three in Quillfield's security master, set up on Feb 16 from the official statements. They are new issues that settled Feb 18–19, and nobody had asked the custodian to set them up.
The Basket agent never changes the basket's composition; cash in lieu needs the portfolio manager.
At 7:02 p.m. the Provider agent emailed the custodian's ETF services team a pre-filled setup file with the three official statements, asking for the expedited channel. At 7:48 a.m. the custodian had set up two and asked for a call schedule; the agent answered from official statement page 3, and the third was set up by 8:31. At 8:36 a.m. the Basket agent ran test file 3: 58 of 58 accepted.
Every step of the agents' run is on the slip, with its time and its source — and can be replayed.
At 7:20 p.m. the night before, the Readiness agent had checked each bond on its own against the official statements and the evaluated pricing vendor: all three agree with the security master. With test file 3 passed, 2 business days are still at risk. Option A, recommended: move the seed wire from Mar 2 to Mar 9 and run NAV dry run 1 on the test-file portfolio on Mar 4, the live run on Mar 11 — first NAV holds Mar 16. Option B, cash in lieu for the three bonds, gains nothing. Option C accepts the slip to Mar 18.
“The Readiness agent never reads the Basket agent's working; it checks the same facts from the official statements and the pricing vendor.”
Raj Malhotra, fund operations lead, agreed in the thread to run dry run 1 on the test portfolio, and the administrator confirmed its plan v3. Tom submitted option A as the maker. When Nadia opens “Review and decide” she sees the dominoes as they stand this morning — 4 recovered overnight, 2 need her decision — and the decision card from Tom beside them: move the $25.0M seed wire to Mar 9; first NAV holds Mar 16; the treasurer co-signs the wire after her.
Tom cannot approve his own re-plan: “You submitted this re-plan (maker). The Head of Product approves it.”
Nadia asks the agents in plain words: “Why March 9 and not March 4?” The Launch Conductor answers from the treasury seed capital policy and the authorized participant's email: the seed desk funds on Tuesdays only and needs 3 business days' notice after a passed AP creation test; Kestrel Markets runs the creation test on Friday, Feb 26; three business days later is Wednesday, Mar 3, so the first Tuesday is Mar 9. Mar 4 is a Thursday and would need a policy exception from the treasurer.
“Every answer carries its source; I change nothing without your decision.” — Launch Conductor
The approval shows what she is signing — move the $25.0M seed wire from Tue Mar 2 to Tue Mar 9, dry runs Mar 4 and Mar 11, first NAV holds Tue Mar 16 — who proposed it and the evidence: test file 3, treasury policy §4.2 and Kestrel's email. She picks reason code RP · Re-plan within float and approves and signs. The dominoes stand up again.
“Approved as Nadia Okafor · Head of Product Development · 9:07 a.m. ET · recorded on the audit trail.”
Seed is money, so it takes two signatures. Claire's screen shows the $25.0M seed — 20 creation units of 50,000 shares at $25.00 — the wire date moved from Mar 2 to Mar 9, 2027, and the treasury rule: Tuesdays only, 3 business days' notice. Beside it, the Readiness agent's first-NAV checklist, each check with its source: pricing coverage 212 of 212, basket test file 58 of 58, fee waiver in force. Only the treasurer can press “Confirm the wire for Mar 9”.
Treasury policy §4.2(c): “Any change to a scheduled seed wire needs the fund treasurer's confirmation.”
The banner over the dominoes reads “Re-planned by Nadia Okafor at 9:07 a.m.: First NAV holds March 16. Seed wire for Mar 9 confirmed by Claire Donovan.” The AP creation test is Feb 26, the seed Mar 9, dry runs Mar 4 and Mar 11. Then the Conductor proposes a change to the series-trust ETF playbook — custodian setup for new-issue municipal bonds from 3 to 6 business days — so the next muni launch asks a week earlier. It goes to Nadia as a change request.
The Launch Conductor never moves a date, a seed wire or a gate on its own; it proposes.
The same ETF launch, seen by five of the people who carry it to first NAV — what a slip looked like before, and what it looks like now.
Seven agents plan, draft, test and chase every launch to first NAV; launch PMs propose, the Head of Product approves.
Plans the critical path from the fund type and the playbook, watches every domino, computes what a slip knocks over in business days, recovers what the agents can and proposes re-plans with options and first NAV dates.
Drafts board resolutions and the board book from the precedent library and the term sheet, every term cited, and tracks each gate with its minutes.
Drafts Form N-1A and N-2 sections and comment-response letters from the filing as effective, cites every changed sentence and maps each SEC comment to its response and page.
Prepares the ticker reservation, CUSIP request, listing application and Form 8-A, and tracks NSCC and DTC eligibility.
Builds the creation basket — the portfolio composition file — and its test files, runs them with the custodian, reads every reject file and reconciles it against the security master.
Chases service providers by email from the launch mailbox, reads their replies in any language with the original beside the English, and updates the domino with the reply as evidence.
Checks seed and first-NAV readiness — pricing coverage, NAV dry runs, Rule 6c-11 website disclosures — and re-checks the other agents' fixes from independent sources.
Approves every plan and gate. The agents propose; a named person decides.
Every screen has a conversation with the agent that owns the work. Ask about a domino, a date, a provider or a filing — every answer cites its source, and nothing changes without a person's decision.
Why March 9 and not March 4?
The seed desk funds on Tuesdays only and needs 3 business days' notice after a passed AP creation test (treasury policy §4.2). Kestrel runs the creation test on Fri Feb 26; three business days later is Wed Mar 3, so the first Tuesday is Mar 9. Mar 4 is a Thursday and would need a policy exception from the treasurer. With the administrator's plan v3, first NAV holds Mar 16.
What did the custodian say?
At 7:48 a.m. the custodian set up 54218P KQ7 and 13921C HX2 and asked for the call schedule of 95783R DM5. I answered from official statement p. 3: callable on or after Dec 1, 2029 at par. At 8:31 a.m. they confirmed all three.
What about option B?
Option B keeps Mar 2 by putting the three bonds into cash in the first creation. It changes the seed basket, which Hannah Cole decides, and Kestrel needs 2 business days to load a custom basket, so it gains nothing. Option C accepts the slip: first NAV Mar 18, a new notice to the exchange, and the launch marketing date moves.
What did the São Paulo sub-custodian say?
In Portuguese, on Feb 22: two documents are missing for the CVM registration, the proof of the fund's CNPJ registration and a signed power of attorney, and it needs 6 business days after receipt. The original is shown beside the English.
13 screens from the working solution, on its sample data. Pick one to see it large.
The rules every launch runs on are settings. The administrator proposes a change; the Head of Product approves it; every change is versioned and logged.
| Setting | Default | Choose from |
|---|---|---|
| Playbook per fund type | Series-trust ETF: 26 dominoes, about 112 days | Series-trust ETF · ETF share class · New share class · Standalone mutual fund · UCITS sub-fund · Interval or tender-offer fund |
| Tip a domino when the forecast slips by | 1 business day | Changed by a change request the Head of Product approves |
| Ask a person when an agent's confidence is below | 90% | Set per agent, from 90% (Launch Conductor, Listing and Provider agents) to 95% (Basket and Readiness agents) |
| Escalate a provider task past SLA after | 2 business days | Provider SLAs: custodian 3 business days per request, sub-custodians 10 |
| Escalate an open decision to the COO after | 1 business day | Changed by a change request the Head of Product approves |
| Freeze dates inside | 5 business days of first NAV | Inside the freeze, only the Head of Product moves a date |
| Seed movements | Two signatures: Head of Product, then treasurer | Every seed movement needs the fund treasurer's confirmation |
| Board book sent | At least 7 days before the meeting | Changed by a change request the Head of Product approves |
| Residency and retention | United States; launch records 6 years (Rule 31a-2) | UCITS sub-fund records in the EU (Ireland, Luxembourg); provider email 5 years (Rule 204-2); entitlements per trust |
Every figure is labelled: a target the solution is built to, an estimate, a typical published result, or a proven one.
“demo” = seen in the working solution, on its sample data · “target” = the design goal, measured in the live solution · “estimated” = our estimate (fees a month: $50 million × 0.50% ÷ 12) · “typical” = published figures (Cerulli Associates via 401k Specialist — 953 active ETFs launched in the US in 2025, 84% of all new US ETFs; Securities Act Rule 485(a)(2) — a new series effective 75 days after filing; a series-trust administrator's estimate, published by Nasdaq and Advisorpedia in “The opportunities and mechanics of launching an ETF” — 100–130 days in a series trust, 6–9 months for a new standalone trust). People, firms, funds, tickers, CUSIPs, issuers and service providers named on this page are fictional — characters and sample data in the working solution.
Fund Launch Management is fund launch software that takes a mutual fund, ETF, share class or UCITS sub-fund from the product committee's decision to first NAV. Each launch is a row of dependency dominoes — board, registration, service providers, exchange listing, seed and operations — and seven agents plan the critical path, draft the board resolutions and filings, run the creation basket test files, chase the providers and catch a slip the night it happens. Launch PMs propose, the Head of Product approves, counsel signs filings and the treasurer co-signs seed moves.
The ETF launch timeline depends on the structure. One series-trust administrator's published estimate is about 100 to 130 days in an existing series trust and 6 to 9 months in a new standalone trust, and a new series' Rule 485(a) filing goes effective 75 days after it is filed. In the working solution each fund type has a playbook with its baseline: about 112 days for a series-trust ETF, 84 for an ETF share class, 60 for a new share class, 95 for a UCITS sub-fund and 210 for a standalone mutual fund.
How to launch an ETF in an existing series trust, in order: product committee approval; the board approves the new series and the advisory agreement under Section 15(c); the 485(a) is filed and SEC comments are answered; the ticker, CUSIP, listing application and Form 8-A; the custodian, administrator and transfer agent, lead market maker and authorized participant agreements; the portfolio composition file, basket test files and the AP creation test; the seed wire; NAV dry runs and Rule 6c-11 website disclosures; the Rule 497 prospectus; first NAV. In the solution each step is a domino with an owner, a date and its evidence — an ETF launch checklist that shows what a late step knocks over.
The Basket agent reads the reject file and reconciles each line against the security master, and the Launch Conductor tips the domino and shows the days at risk and every later date it moves. The Provider agent asks the custodian to fix it, a second agent re-checks the fix from independent sources, and the retest runs. Whatever days the agents cannot recover go to the Head of Product as options with their first NAV dates.
Yes. Besides the series-trust ETF, the solution has playbooks for an ETF share class of an existing mutual fund under the firm's exemptive order, a new share class, a standalone mutual fund launch in a new trust, an Irish or Luxembourg UCITS sub-fund and an interval or tender-offer fund. The Launch Conductor builds the critical path from the fund type and the target first NAV date, and a launch PM submits it for the Head of Product to approve.
The Provider agent sends chasers from the launch mailbox and reads every reply, in the provider's language, with the original shown beside the English and the reply attached to the domino as evidence. Tasks carry SLAs — 3 business days per request for the custodian, 10 for sub-custodians — and escalate 2 business days past due. It never agrees to terms or fees.
Yes. The agents propose; they never move a date, sign, file, record a vote, wire money or change a basket. A launch PM proposes each plan and re-plan and the Head of Product approves it with a reason code, fund counsel alone signs filings, the treasurer confirms every seed movement, and every step is on a hash-chained audit trail.
The Agentic Solution Engine builds and deploys it from your requirements and documents — your launch playbooks, past launch trackers, board precedents and filings — and it goes live once every quality gate has passed. We will walk you through it on a launch you have in flight first.
Every 15(c) answer cited and checked against last year, ready 10 days before the deadline.
1-min demoFee BenchmarkingEvery peer fee your board sees, cited to the peer’s own fee table and signed.
1-min demoFund RationalizationEvery subscale fund, each quarter: a cited case argued both ways, signed before the board.
1-min demoWe'll run Fund Launch Management on your own playbooks and a launch you have in flight.