A smaller, sharper Grainfather — and a clear account of what is resolved and what is not
We are asking the board to support and approve a revised strategy for the brand — the four points below, taken together. This is a mandate to execute, not a set of individual product approvals. The decisions that sit underneath it are delivered through defined processes with named return dates.
Along with cost reductions across labour, marketing, IT and logistics. Reduce the range, and put the G40 into Bevie Handcraft’s Proposed Deletions Process — the process that systematically works through a fully considered decision, and will have resolved it by the first board meeting of 2027.
FY25 86% of revenue, plus D2C. Reduced investment in AU/NZ. Canada to stop on regional compliance grounds and cost to implement. Any remaining resource is focussed on consumer and customer activities targeted to these regions specifically, beyond the limited AU/NZ spend that earns its keep.
A defined review, not an open one: the G40 and other products outcomes, the AU/NZ delivery model decided and implemented, and a fully costed SKU & spare parts run-out and write-off plan. By this time all three revenue engines are running (Brouwland, D2C in EU, US to MoreFlavor).
Not a deferral. By 2028 the decisions taken today will largely have played out (G30 core + range, NPD brand requirement, D2C returns, EU & US distributor deliveries) — and the lease renewals force the infrastructure call. That is when we can settle the brand’s ongoing shape.
"We built for the COVID market. We now right-size for the real one — around a position only Grainfather owns."
→How we got here, in five steps
The G30
All-grain, all-in-one, premium, app-connected, quality, well supported — a category Grainfather defined. Copycats followed, but the original G30 and its associated gear (fermenters, glycol chiller and so on) remain rated and wanted.
Surge to $15M
Lockdown demand took us to a ~$14.7M peak. We scaled range, formats and footprint to match it. Indications were that greater volume capacities were wanted.
The peak was the anomaly
Demand normalised, low-priced rivals crowded in, and the EU warehouse closure removed roughly half our brand revenue. We had also bet on bigger formats, and the market is going the other way: small batch.
The enthusiast is still here
And still chooses us for the ecosystem — app, consumables, community, support — not on price. Demand continues to sit with the G30 + associated products and below.
Retrench and defend
Shrink to the G30 ecosystem, run the large formats out — with G40 into the Proposed Deletions Process to establish where, if anywhere, it earns — and defend the position only we hold, and for which people are prepared to pay.
→Where every decision actually sits
This recommendation is not one decision. Some of it is in front of you now for support, and some we are deliberately leaving open because we do not yet have the evidence to close it.
The revised brand strategy as a whole, and the mandate to execute it: the reduction to core range, the cost reductions, the concentration on EU + UK + US, reduced investment in AU/NZ, and a signature product (G40) into the Proposed Deletions Process.
The G40 outcome, the AU/NZ delivery model, and a costed portfolio run-out and write-off plan. All three are SMT decisions — reported at that meeting, not brought back for approval.
Read more
- The G40 outcome from the Proposed Deletions Process, resolved by the SMT.
- The portfolio run-out — not just G40. How long it actually runs, what it costs, and the write-off estimates that go with it, covering SKUs and spare parts across the range.
- The AU/NZ delivery model — having approved reduced investment, we settle and implement the mechanics, including possible D2C-only.
Why that meeting. The Proposed Deletions Process needs clean data. G40 and G70 will have been back in market long enough for the post-rework re-buy period to wash through, so we read genuine sell-through rather than a defect-suppressed year. Brouwland and Grainfather.com/EU will be several months established and MoreFlavor operational — so the review runs on trading evidence rather than assumption.
Board calendar note. Meetings have run January, April, August and November; the schedule is being reset alongside the move to a January–December financial year (tab 02, item 4), so this is anchored to the first meeting of 2027 rather than to a month.
The shape of the brand, settled once the evidence is in. Five things resolve between now and then, most of them consequences of what we are asking you to approve today. We are not asking for more time.
Read more
- Whether the reshaped core focus is working.
- The G40 decision closed out, and with it the G70 run-out largely complete.
- Whether the modular system belongs under this brand or not, with a roadmap for the retrenched category alongside it.
- How Brouwland, MoreFlavor and D2C in Europe have actually performed.
- The forcing function — the NZ and AU lease renewals in 2029, most materially AU, require the brand call to precede them rather than follow.
We are waiting on these answers, and they arrive on this timetable.
→Three things that strongly informed our recommendation
A format bet failed. The brand did not.
We invested in bigger systems on the view that brewers wanted to scale up. They did not. G40 is down 71% from its FY22 peak and G70 down 79% from its FY21 peak, and both of those declines happened while the products were fully available — before the FY26 defect issue. Meanwhile demand continues with far less decline at G30 scale and below. Exiting the large formats corrects a bet that did not pay off; it is not a retreat from a good position.
Make the brand what it is at its core, then reassess on clean sales with the opportunities realised.
FY26 is unfair to judge — a major product defect on G40 and G70, and new initiatives that have not had time to be realised: Brouwland as distributor in the EU, D2C for Grainfather switched back on in the EU, and the forthcoming change of distributor in the US. The brand deserves more time to see how these play out. We are not avoiding the decision, we are balancing the considerations.
Exiting the brand costs more than concentrating.
When looking at brand profitability estimates, Grainfather's EBITDA loss includes shared overhead that does not disappear on exit — it redistributes onto the other brands. Beyond that: ~$450k of other-brand revenue is at risk in the EU (retailers like to buy large Grainfather products then pack out their orders with our other brands), the Brouwland agreement itself exposed, a warranty tail that survives the decision, and leases that run to 2029 regardless. There is no cash release available from a more drastic decision today.
→Not a wind-down
Cheap on a spreadsheet, expensive in practice — and slow. Unless we write off saleable stock in our warehouses and at distributors, it still has to run out, and the brand has to be supported the whole way down, then to a legal minimum beyond that. A deletion decision taken today would not end the costs today.
→Not a request for investment
We are not asking to fund a turnaround. New product development has not grown this brand — NPD has contributed close to nothing since FY24 — so the growth this plan counts on is recovering the EU and US through better partners, not new equipment. That is a statement about this plan, not about the category’s future. We will build a roadmap for success in the retrenched category; it informs the Key Brand Milestone in 2028 rather than this meeting.
→A major step towards profitability
$328k of cost action is real, it is largely within our control, and this plan delivers it. It does not by itself make the brand profitable — fully allocated, Grainfather remains loss-making over the next 12 months. But on current estimates the following twelve months reach +$353k before indirect costs, with revenue back to FY25’s level on a better mix and the cost base lower again (tab 08). This is a first step, not a one-off.
The risk sits on the revenue side, not the cost side. The improvement depends on revenue growing 13%, and the three engines carrying that growth — Brouwland, GF.com D2C in Europe, and the US move to MoreFlavor — are weeks old or not yet live. The +$49k also sits before Customer Service headcount, early-stage NPD engineering and the cost of the run-out itself. We name those here rather than footnote them, because they are the difference between this line and a profit.
Separately, and outside every figure in this pack: the RahrBSG to MoreFlavor transition carries a one-off −$416k effect on the company forecast. It says nothing about demand — see tab 02, item 2.
Read this before the figures
Nine things that shape how the numbers in this pack should be read. Each is referenced again where it applies, but they are collected here so none of them is a footnote you find after forming a view.
FY26 is not a fair read of the brand
A thermal defect took G40 and G70 out of the UK, EU and US from January 2026 — eight of FY26's twelve months — while stock was returned to China for rework and shipped back to regions. The scale of it: large formats did $395k in the four months before the stop-sell, then −$41k in the four months after, as returns exceeded sales. Where a clean read matters, we use FY25.
The RahrBSG buy-back sits outside these numbers
Moving from RahrBSG to MoreFlavor means buying back existing inventory and on-selling less of it into the new distributor — a one-off credit of about −$416k. It is an anomaly and says nothing about underlying demand, so it is excluded from every FY26 figure in this pack. It is included in the recent business reforecast to Rahr for financial expectation purposes. If the two documents differ, this is why.
For scale: FY26 US revenue is $608k, so the effect is close to two-thirds of a full year of US sales — which is why the US case is framed as a next-twelve-months recovery rather than an FY26 one. The size of the credit is driven by two things: excess stock of Grainfather products generally, from container purchase decisions, and in-progress rework stock of G40s and G70s due back in the US over the coming weeks.
“Profit before indirect costs” is not profit
The line this plan improves is struck before shared overhead. Indirect costs — management, property, shared functions — are allocated to Grainfather by revenue share and sit below it. So the $328k improvement — from −$279k if we do nothing to +$49k with the changes — is a milestone, not profitability; fully allocated, the brand is still loss-making. It also means Grainfather's reported result moves with how other brands perform. This treatment lines up with the brand profitability data shared at each board meeting, at the “before indirect costs” line.
The financial year changes at this meeting
Everything here is on the current September–August year. Forward figures are labelled “Next 12 Months” — September 2026 to August 2027 — deliberately, not FY27, because the new FY27 means January–December 2027. The period spans the FY26S stub year plus the first eight months of the new FY27, so it does not read across to either budget in front of you today in other formats outside of this topic — that is, FY26S or FY27.
Margin is only measurable from FY25
Ahead of FY25 we ran a programme of work that enabled us, from September 2024 onwards, to see and measure margin at customer, brand, product and region level — which we had not had until then, because of intercompany complexities. Margin is therefore only accurate from that period, so you will not see it referenced before then. It is also why the eleven-year chart shows revenue but no margin line.
There is no clean year to measure against
Every year that could serve as a baseline is distorted by something structural, so comparing any two of them compares different businesses.
| Year | Brand | Why it is not a clean baseline |
|---|---|---|
| FY19 | $9.54M | Neither G40 nor G70 had launched. The G30 was three years old and still in its heyday. |
| FY20 | $11.81M | COVID began. G70 launched. |
| FY21 | $14.69M | COVID in full flight. The brand peak — but an artificial one. |
| FY22 | $13.10M | G40 launched. G30 fell $3.6M against a $2.6M G40 gain — direct cannibalisation. Still partly COVID. |
| FY23 | $8.39M | Normalisation began — but also the EU warehouse closure and the shutdown of Grainfather.com/EU. |
| FY26 | $3.60M | G40 and G70 out of market for eight of twelve months on the defect. |
So we do not lead with percentage declines. Where a comparison appears we say what sits inside it. The case rests on unit economics, margin and demand direction — not on headline decline rates.
Grain Mill data distorts what it touches
We hold significant excess Grain Mill stock and are using it as a discount and giveaway to move other SKUs. That affects where its costs and values land in our ERP, and therefore in our financials. Treat any Grain Mill figure as indicative. It is also blurring other products where we promote them together, such as G30 and G40 — but extrapolating that out is too difficult to be worth the time.
Revenue in the forward plan includes products we are deleting
The Next 12 Months holds revenue partly because deleted lines keep selling through the run-out. Costs fall before revenue does, which buys time — but it reverses: part of next year's revenue comes from products that will not be there the year after.
All figures are NZD
Unless stated otherwise. Where a table shows values in thousands, the header says so.
→Key dates for reference
The events behind several of the points above, in sequence.
EU warehouse closed
Bevie Handcraft’s EU warehouse, based in the Netherlands, was closed. Grainfather.com/EU was shut at the same time, removing both the physical and the direct channel in the region.
G40 and G70 blocked for sale
The defect issue stopped sales of both formats. Stock was returned to China for rework.
Brouwland distributorship begins
First orders placed under the new distribution arrangement — the EU’s route back to market after nearly three years.
Grainfather.com/EU reopened
Direct sales recommenced, with pick and pack provided by Brouwland. The EU had been without a direct channel for 35 months.
Corrected stock still awaited
G40 and G70 reworked stock is currently on water to most regions. Sales have been blocked for roughly seven months and counting.
Act now on what's ready. Optimise what's unproven. Settle the brand's shape at the Key Brand Milestone in 2028, ahead of the 2029 infrastructure renewals.
01Act now
- Cut labour, marketing, IT and warehousing costs.
- Confirm deletion of GC4, the Whirlpool Arm and the 25L Sparge Water Heater; commence run-out. G40 enters the Proposed Deletions Process.
- Reduce investment in AU/NZ. Stop selling into Canada.
02Optimise · to the first board meeting of 2027
- Test the three new revenue engines: Brouwland, GF.com D2C in Europe, MoreFlavor in the US.
- Resolve the G40 through the Proposed Deletions Process. An SMT decision, reported at that meeting rather than brought back for approval.
- Decide and implement the AU/NZ delivery model, including possible D2C-only. Also an SMT call, reported not approved.
- Return with a dated, costed run-out plan for the portfolio — not just G40 — including write-off estimates.
- Why the evidence will be there: the Proposed Deletions Process needs clean data to work with. G40 and G70 will have been back in market long enough for the post-rework re-buy period to wash through, so we read genuine sell-through rather than a defect-suppressed year — a measurement precondition, not a reason to keep them. Brouwland and Grainfather.com/EU several months established; MoreFlavor operational.
- So that meeting can settle the G40 outcome and the AU/NZ model on evidence rather than assumption.
03Key Brand Milestone · 2028
- Not a deferral. By 2028 most of what we are asking you to approve today will have played out.
- We will know whether the reshaped core focus is working.
- The G40 decision will be closed out — and if deleted, gone, with the G70 run-out largely complete alongside it.
- We will know whether the smaller-format modular brew system belongs under this brand, or not — and we will have a roadmap for success in the retrenched category to put alongside it.
- We will have real evidence on Brouwland, MoreFlavor in the US, and D2C in Europe.
- And the forcing function: NZ and AU leases renew in 2029 — most materially AU. Whether we carry large equipment drives how much space we commit to, so the call has to precede the renewals rather than follow them.
→Portfolio
Fifteen SKUs stay and account for 58% of sales. Two are already running out. G40 goes into the Proposed Deletions Process, and three more are recommended for removal — GC4, the Whirlpool Arm and the 25L Sparge Water Heater, together 8.0% of sales. None leaves a gap: GC2 has grown as GC4 declined, and the 18L Sparge Water Heater stays.
Full detail
FY25, on the current NZD extract at weekly rates — the same basis as the regional table below and the rest of the pack, so the G40 figure agrees in both places. Covers the master-SKU products named: about $148k of FY25 consumables and accessories cannot be attributed to a product family and sits outside the table, which is why the total reads $3,807k rather than the brand’s $4.28M. FY26 is distorted for G40, G70 and the Grain Mill, but reads straight for everything else (tab 02, items 1 and 7).
| Action | SKUs | Sales (NZD) | % of sales |
|---|---|---|---|
| Keep | 15 — G30 core, fermentation, accessories (incl. Stainless Steel Paddle) | 2,200 | 57.8% |
| Consider | 5 — Hop Plate, pressure transfer, Whirlpool & Aeration Paddle, pump kit, growler | 91 | 2.4% |
| Run out | 2 — G70, Grain Mill (already decided) | 454 | 11.9% |
| G40 — in Proposed Deletions Process | 1 — G40 (resolved by the process, not today) | 757 | 19.9% |
| Remove — to be finalised | 3 — GC4, Whirlpool Arm (G40/G70), SWH 25L | 305 | 8.0% |
| Total | — | 3,807 | 100% |
→Why G40
The G40 is the better system and the defect is fixable. Neither is the reason. A bigger, more complex machine costs materially more to build beyond the premium it attracts, and too few brewers will pay for it. It goes into the Proposed Deletions Process, which resolves the decision by the first board meeting of 2027: on FY25 it earns its best margins in the US and Europe — 76% of its sales. The UK sits between those and the weaker markets, at a materially lower price per unit — a pricing question rather than a settled verdict. NZ margin needs looking into, which the deletions process will deliver.
Full detail
The G40 is the better system, and the heating defect is fixable. Neither is our reason. The reason is that a bigger, more capable machine costs materially more to build, so it must carry a premium — and too few brewers will pay it. That gap has not closed and we see no reason it will.
| Evidence | What it means |
|---|---|
| 3 G40s per 10 G30s historically | Demand sits with the smaller format even when both are sold. |
| Large-format revenue −40% in a year, fully available | A demand problem, before the stop-sell. |
| ~$90k warranty cost vs $59k across the entire core | Disproportionate support load. |
Into the Proposed Deletions Process
- Current sales and margin delivery — definitively, on reconciled transaction data.
- Cannibalisation — what G30 recovers if G40 goes. G40's launch cost G30 $3.6M against a $2.6M gain, so some of that should return.
- The inventory tail — run-out economics, timing and discounting needed.
- Spare parts that must be carried for the warranty tail, and those written off. This is the largest single unknown, and it swings entirely on the G40 decision: G40 shares a significant parts inventory with G70. Delete it and much of that inventory becomes a write-off; retain it in any market and we consume those parts over the following years instead. The same stock, valued two completely different ways — which is why a write-off estimate cannot precede the decision.
- The five regional versions and their compliance regimes, assessed market by market.
- Cost of a permanent fix to the heating defect, where retention is on the table.
- G40-specific costs that fall away — IT, connectivity, support and service load.
- Retailer and market response — what a deletion signals to distributors and retailers, whether it changes their willingness to keep buying the rest of range, how they perceive our brand direction and how we manage that. This is commercial as well as reputational: the run-out only recovers value if retailers keep ordering lines they know are being discontinued as well as the other lines in the range we are continuing to sell.
- The opportunities as well as the disadvantages — the process is not built to confirm a deletion.
G40 goes into Bevie Handcraft’s Proposed Deletions Process. We are not asking the board to approve a deletion today, and we are not asking for a decision at the 2027 meeting either — the process resolves it, and we will report the outcome. It works the question market by market, and it may not end in a clean exit.
What the data shows, market by market, on FY25 — the only year whose margin data we can rely on. This is not a pass/fail test: the Proposed Deletions Process weighs these figures alongside inventory, spare parts, the five compliance variants and the other inputs listed above. But the spread is wide enough to be informative:
| Region | Revenue (NZD) | Gross profit (NZD) | GM% | What it shows |
|---|---|---|---|---|
| United States | $329k | $103k | 31.4% | Strongest the best G40 market, and by a wide margin. |
| Europe | $243k | $67k | 27.3% | Strong largest by volume. |
| UK & Ireland | $110k | $28k | 25.0% | Weaker on price sells at $1,180 a unit against $1,349 in Europe, so pricing is worth a look. |
| Australia | $55k | $4k | 7.0% | Marginal not a loss, but not a margin worth serving. |
| NZ & Pacific | $19k | −$0.3k | −1.8% | Loss-making the only region actually losing money. |
| Total | $757k | $201k | 26.5% | — |
The version complexity cuts both ways. G40 carries five regional versions — five electrical compliance regimes, with some unique parts behind each:
→Run out, not write off
There are two ways to stop selling something: write the stock off, or run it out and recover the value. We recommend run-out every time. There is $2.10M of stock across the eleven affected SKUs, clearing over roughly three years. The brand has to be supported throughout — service, warranty, parts, platform. None of that stops on the day a deletion is approved.
Full detail
Two ways to stop selling something: write the stock off, destroying inventory we have paid for in our warehouses and at distributors — or run it out and recover the value. We recommend run-out every time. It takes years, and the brand must be supported throughout: service, warranty, parts, platform, enough marketing to shift the stock. None of that stops on the day a deletion is approved. Legal minimum obligations then continue beyond that (tab 04).
| Status | SKU | Units at run-out | Stock value (NZD) | Est. months to clear | Last region | Approach |
|---|---|---|---|---|---|---|
| In Proposed Deletions Process | G40 (if deleted) | 562 | $695,546 | 10 | EU | Sell through with minimal discounting needed. |
| Run-out | Grain Mill | 2,266 | $687,209 | 36 | NZ/AU | Being used to move other SKUs — heavily discounted or as a freebie. $106k of the value is the NZ version. |
| Run-out | G70 | 232 | $323,557 | 20 | NZ/AU | Promote where needed to shorten the run-out. |
| Remove | GC4 | 216 | $232,871 | 18 | NZ | Sell through with minimal discounting needed. |
| Remove | Sparge Water Heater 25L | 664 | $84,855 | 36 | NZ/AU | Promote where needed to shorten the run-out. |
| Remove | Whirlpool Arm (G40/G70) | 562 | $9,951 | 25 | NZ | Promote to align with the last of G40 or G70, then write off. |
| Consider | G30 Hop Plate | 678 | $23,315 | 11 | AU | Move stock where needed to level the run-out. |
| Consider | GF30 Fermenter Cooling Pump Kit | 366 | $19,069 | 22 | US | Move stock where needed to level the run-out. |
| Consider | GF30 Fermenter Pressure Transfer | 315 | $14,333 | 11 | UK | Move stock where needed to level the run-out. |
| Consider | Whirlpool & Aeration Paddle | 497 | $9,557 | 8 | NZ | Move stock where needed to level the run-out. |
| Consider | Swing Top Growler 2L | 38 | $1,417 | 6 | UK | Move stock where needed to level the run-out. |
| Total | 11 SKUs | 6,396 | $2,101,680 | up to 36 | — | — |
Already in run-out — $1.01M. Grain Mill and G70, decisions already taken.
G40, in the Proposed Deletions Process — $696k. The single largest holding, but the shortest tail at 10 months. Contingent on the process outcome.
Remove — $328k. GC4, the 25L SWH and the whirlpool arm.
Consider — $68k. Immaterial by value across five SKUs.
G40 clears fastest of the large holdings. At 10 months it is the shortest tail on the list despite being the biggest single value — so a deletion outcome would not commit us to a long wait.
The Grain Mill is already doing a job. It is being used as a discount and giveaway to move other SKUs, which is why its data distorts everything it touches — and why 36 months is a working estimate rather than a forecast.
→Markets
Europe, the US, the UK and D2C are the focus — 86% of FY25 revenue. Europe and the US get investment, the UK is defended, and D2C is the only growing channel at $864k, up 58%. AU/NZ moves to hold on reduced investment, and Canada stops on compliance cost.
Full detail
| Region | Stance | Basis |
|---|---|---|
| Europe | Invest | Two new strands, both weeks old: Brouwland, and GF.com D2C switched back on. Was ~half of peak revenue. FY26 $1,565k — 43% of the brand. |
| United States | Invest | MoreFlavor replacing RahrBSG. FY26 $608k, down 36% on FY25 ($946k, recomputed basis). The transition carries a one-off −$416k effect held outside these figures — see tab 02, item 2. |
| UK & Ireland | Defend | FY26 $818k at ~26% — the second-largest market. Healthy — don't disturb it. |
| D2C / Online | Invest | FY26 $864k, up 58% on FY25 — the only growing channel in the brand. ~$1.1M assumed for the next twelve months; return has fallen but is still 5×+ on spend, and it improves margin. |
| Australia & NZ | Hold | FY26 $561k revenue (AU $385k, NZ $177k); NZ runs at roughly break-even on gross margin. Decided: reduced investment — digital spend pulled back, no dedicated marketing resource or activities targeting the region, and continued digital spend only where it makes financial sense. With the SMT: the delivery model, including possible D2C-only — to be settled and implemented ahead of the first board meeting of 2027, and reported there. SKUs for this region add complexity to manage. |
| Canada | Exit | FY26 $50k. Full country compliance costs more than the market returns. Unless the distributor carries the risk. The distributor has stock they will sell through, but will not buy more. |
→People
Brand manager. Upcoming parental leave lets us test a reduction in marketing time and focus. Minimal responsibilities will be shared into other existing roles. A same-level role is required on return post parental leave, so this needs to be kept in mind for future decisions.
Engineering. App developer stays full-time — reducing hours risked losing them, and an agency would cost more. Note the engineering line assumes no change, but actual Grainfather time runs at 30–40%; the rest is other brands.
→Footprint and the 2029 deadline
Three warehouses: NZ, AU and UK. Grainfather holds 13% of the space in use on $3.6M of revenue — a larger tenant than its revenue suggests, because the products are bulky. It is the least space-efficient brand at every site. The NZ and AU leases renew in 2029 — most materially AU — so the brand decision has to land in 2028.
Full detail
Three warehouses globally: NZ, AU, UK. The US and Canada run through distributors. Europe is a hybrid — direct to retailers from the UK warehouse, plus an appointed EU distributor. The table below shows warehouse utilisation as at end July 2026 and Grainfather's share of that space, on a locations basis.
| Site | Lease to | Locations (capacity) | Locations used | Utilisation | Grainfather locations | GF % of space used | GF % of capacity |
|---|---|---|---|---|---|---|---|
| AKL · New Zealand | 2029 | 2,962 | 2,267 | 77% | 285 | 13% | 10% |
| QLD · Australia | 2029 | 2,117 | 1,099 | 52% | 100 | 9% | 5% |
| WAL · United Kingdom | 2033 | 2,387 | 1,824 | 76% | 267 | 15% | 11% |
| Total | — | 7,466 | 5,190 | 70% | 652 | 13% | 9% |
Three things worth drawing out.
1. Grainfather is a larger tenant than its revenue suggests. It holds 13% of the space in use across the three sites while producing $3.6M of revenue. At the UK site it is the single largest brand by locations (15% of space used). At Auckland it is fourth, behind Still Spirits (29%), Mangrove Jacks (19%) and unassigned stock (16%). At the AU site it is fourth (9%).
2. It is the least space-efficient brand at every site — and that, not stock depth, is what drives the footprint. Auckland holds 58 units per location of Grainfather stock against Still Spirits' 762 and Mangrove Jacks' 115; the UK site holds 106 against Still Spirits' 473. Large equipment in large cartons. This is the substance of the 2029 question: whether we carry large formats determines how much space we must commit to, and no amount of stock discipline changes that.
3. The AU site's spare capacity is not a Grainfather question. QLD runs at 52% utilised against 76–77% at the other two, but Grainfather is only 5% of its capacity. Shrinking or exiting Grainfather there would release very little space, so that decision has to rest on other brands' needs.
→Innovation
As a business we still believe in equipment innovation and need the engineering resource for it. It may or may not be for Grainfather. That is why the roles are protected while brand costs fall. The engineering team also provides some support service to the brand, which continues until we have no product and have met our legal requirements. We currently have the role of Engineering Manager empty, and are starting a short piece of external consulting work to review, assess and make recommendations for this function before progressing with any replacement plan.
We are not asking you to fund innovation as this plan’s growth story, and we are not presenting NPD as one — the track record does not support it, and the established core is what earns today. That is a statement about this plan, not about the category’s future.
What we will do is build a roadmap for success in the retrenched category. We are not offering the solution today, because we do not yet have it — and presenting one would be a request for investment we are not making. One candidate is the modular brew system, currently in early development, which could in time take the position the G30 holds now. It may become relevant to Grainfather, or not.
These take up to three years to reach market, so nothing here is imminent. The roadmap is an input to the Key Brand Milestone in 2028, not a request in front of you now. All we ask today is that the decision doesn’t foreclose it.
Why we can't simply switch this off — and what we still don't know
The legal reality behind scaling back, a full list of what remains unresolved, and a note on how far each figure in this pack can be trusted.
→Warranty and legal obligations do not shrink with the brand
| Obligation area | What it requires | Risk if under-resourced |
|---|---|---|
| Statutory / implied guarantees | Acceptable quality, durability and fitness-for-purpose guarantees (CGA, ACL, EU Directive, UK CRA, US state law) apply regardless of brand strategy. | Claims can be brought for years after purchase, driven by expected product life rather than our sales activity. |
| Express warranty terms | The advertised three-year warranty is a contractual promise that must be honoured as stated. Our terms contain no carve-out for discontinuation. | Under-resourcing support creates breach-of-contract exposure, not merely reputational risk. |
| Spare parts and repair access | Several regimes imply parts and repair capability must remain reasonably available for a reasonable period after supply. Norway requires five years post-sale, independent of the stated warranty term. | The most common real-world trigger for a statutory breach claim — more so than the warranty clock itself. |
| Product safety and recall readiness | Must be able to identify affected units, contact purchasers and action a recall for the product's reasonably expected life. | Regulators expect recall readiness independent of current commercial focus. |
| Data retention | Warranty registration data (serials, purchaser details) must be retained for the liability tail under applicable privacy law. | Retention should match, not fall short of, the warranty and liability tail. |
What doesn't change
Warranty terms already promised stand until they expire naturally. Parts and repair capability must remain reasonably available — this is where under-investment is most likely to create legal exposure, rather than marketing spend. Recall and safety monitoring continue at the same standard as any active product line.
Where scale-back is genuinely safe
Marketing, new product development and market expansion are commercial choices, not legal obligations. Warehouse footprint can be right-sized, provided a minimum part and repair reserve is maintained against the outstanding warranty tail.
→Still open — flagged, not resolved
These were raised in SMT sessions and have not yet been closed out — they require more time, analysis and decision-making to close, which is in plan for the next few months.
| Open item | What's needed | Bearing on the recommendation |
|---|---|---|
| G30 vs G40 US sell-through | Like-for-like sell-out data from the US distributor, not just container and order data. | Would sharpen, but is unlikely to reverse, the G40 case for the US region. The distributor change may bring a positive sales trajectory too. |
| A US and Europe only G40 | Whether keeping one regional version rather than exiting everywhere could preserve profit with less complexity. Parts cross-over with G30 needs checking with product and engineering. | A genuine alternative to full exit; not yet modelled. This is an explicit input to the Proposed Deletions Process, which resolves it by the first board meeting of 2027. |
| AU/NZ delivery model | Reduced investment is decided. The model is not: whether to move to D2C-only, stay as-is, or take a harder line still. Needs scenario planning, and is complicated by Australia still being strong on physical store sales, unlike the other regions. | An SMT decision, not a board one. We will settle and implement it ahead of the first board meeting of 2027 and report the outcome there. Anything that changes the warehouse footprint still has to resolve by the 2028 Key Brand Milestone, ahead of the 2029 renewals. |
| Service load by product | Whether support tickets can be sliced by product or SKU, not just fault type. | Would quantify the labour intensity of the Remove portfolio suggestions. We are also active on a project on AI in Customer Service, which we expect to bring time benefits — especially to this brand. |
| Digital agency retainer | Narrower than it first looked. D2C-allocated digital spend continues in AU/NZ where it earns, so the question is only whether pulling back the non-D2C activity reduces the retainer or simply reallocates it. | Affects the size, not the existence, of the AU/NZ saving. Question is out with the agency at the moment. |
→Resolved through brand review
- App resourcing. The SMT considered reducing the app developer role — based in France, on contract — to 0.5 FTE. On further consideration it was decided that reducing the role carried significant risk: the team member would need to find full-time work elsewhere, putting even the 0.5 at risk. We also have other brand developments likely to use this role, so we may achieve the reduction in focus and cost on the Grainfather side while investing in other brand development.
- Engineering resourcing. These roles are critical for business equipment innovation, and the majority of their time is now focussed elsewhere. The Modular Brewing System sits in "explore" and will resolve both whether the development progresses and which brand to apply it to. Consensus is that the project is worth investment through the next stages irrespective of brand.
- Brand manager maternity cover. No backfill during leave, which lets us test operating with less dedicated marketing resource on the brand. Critical point: we need to provide a same-level role on return, and we are confident we can do that across other brands if needed.
- Canada. The SMT Session 2 deck recorded this as Review. Since then, compliance issues affecting particular Canadian regions and a costing of full country-wide compliance across all key equipment have led to a decision to stop selling into Canada — unless the distributor elects to carry that compliance risk itself. See tab 03.
The questions we expect — and our answers
1. Does the brand make money — or can it?
No. FY26 closes at −$357k; the unchanged plan for the Next 12 Months is −$279k.
Cost actions are worth $328k — taking the Next 12 Months from −$279k to +$49k. But that is profit before indirect costs (tab 02, item 3), so fully allocated the brand is still loss-making. It also assumes 13% revenue growth: 8.5% is needed to reach zero, flat revenue gives −$89k.
Two things frame it. Infrastructure costs are allocated by revenue share and fixed until the 2029 leases — exiting wouldn't remove them, it would move them onto the other brands. While they are fixed anyway, we may as well use the space and earn against it. And the EU D2C site has only just been switched on, so none of that revenue is proven yet.
The direct cost side is ours to manage and this plan gets it largely done. The revenue opportunity is only weeks old in the EU (Brouwland and D2C) and not yet realised in the US (the RahrBSG to MoreFlavor change).
2. What are we stopping?
Products. GC4, the whirlpool arm and the 25L Sparge Water Heater are confirmed for removal. G40 — 19.9% of FY25 sales — enters the Proposed Deletions Process rather than being deleted today; together the four are 28% of FY25 sales. G70 and the Grain Mill were already deleted and are running out.
Markets. Reduced investment in AU/NZ — digital pulled back, no dedicated marketing resource or activities targeting the region, continued spend only where it earns. Stop selling into Canada.
Cost. Brand manager to 0.3 FTE during parental leave, with the residual responsibilities absorbed into existing roles. Marketing spend not tied to the D2C platform or NPD is cut. $63k of Particle IoT hosting comes out when the contract ends in March 2027 — though CloudAMQP, already in the stack at about $750/month, is the replacement, so the net saving is real but smaller than $63k. And a partial 3PL reduction — $126k against a $147k budget — from moving US and EU logistics from Mainfreight to MoreFlavor.
Not stopping: warranty, parts and service. Legal obligations, not choices.
3. What is the cost of doing nothing — or of exiting?
Doing nothing costs $279k over the Next 12 Months, and keeps us building a format the market won't pay for.
Exiting is not the saving it looks like. The reported loss includes allocated overhead that doesn't disappear — it redistributes onto the other brands. On top of that: ~$450k of other-brand revenue at risk in the EU, the Brouwland agreement itself at risk, a warranty tail that survives the decision, stock that must run out or be written off, and sunk leasehold costs across the warehouses.
4. What happens to our other brands if we exit Grainfather?
This is the part that makes exit expensive, and it sits outside Grainfather's own P&L.
Grainfather appears in 47% of EU orders and anchors the basket. Share of each brand's EU sales that arrive inside a Grainfather order: On The Rocks 97%, Keg King 96%, Still Spirits 87%, Mangrove Jacks 54%. On our attrition assumptions that is ~$450k of other-brand revenue at risk.
Beyond the revenue, removing Grainfather would be a material change to the intent of the Brouwland agreement — so the agreement itself should be assumed at risk. UK, Australia and NZ are assessed as low or no cross-brand risk.
5. Why believe 13% growth from a brand that just missed its budget?
Fair challenge — FY26 is expected to land about 9% under.
What's different: two structural changes took effect part-way through FY26 — Brouwland and GF.com D2C in Europe — with the US still on RahrBSG in downturn and not yet moved to MoreFlavor. That move lands before September 2026, so the Next 12 Months is the first full year of all three. On top of it, two products were effectively out of market for eight months of FY26 on the defect issue.
What isn't: none has a track record with us, and we'll have two to three months of EU data by the meeting.
Approve on the cost actions. Hold us to the revenue at the first board meeting of 2027. Get G40 and G70 back in market to continue sell-through.
6. If G40 and G70 were unsellable for seven months, how do you know demand isn't there?
We don't rely on FY26 for that. Both formats declined hard while fully available: G40 is down 71% on revenue and 76% on units from its FY22 peak; G70 down 79% and 81% from its FY21 peak.
In the four months before the stop-sell G40 was running only 5% below its FY25 rate — so demand had stabilised, at 26.5% margin. G70's revenue was up over the same window on a product already scheduled for deletion, which reads as run-out clearance rather than renewed demand.
Same conclusion for both. They sell. They just don't earn.
7. G40 is the better product and you say the defect is fixable. So why is it in the deletions process?
Because better isn't the same as viable.
A bigger, more complex machine costs materially more to build than the premium it attracts, and too few brewers will pay for it. The tell is in the ratio: we still sell roughly three G40s per ten G30s, and large-format revenue fell 40% in a year while both formats were fully available.
Fixing the defect removes a cost. It doesn't close that gap. That said, G40 goes into the Proposed Deletions Process rather than being deleted today — it earns solid margins in the US and Europe, which are 76% of its sales, so a blanket exit would give up margin we have not yet proven we should lose. The process resolves it market by market and reports at the first board meeting of 2027. The UK earns less per unit — $1,180 against $1,349 in Europe — which is a pricing question for that process to weigh.
8. You have almost 300 SKUs. Isn't the range over-extended?
No — and it's the easiest wrong conclusion to draw from the count.
190 of 294 SKUs are spare parts we're legally required to keep available. They generate 5.3% of revenue but exist primarily to service warranty. They have been reviewed and tightened as far as we can go, and will reduce further once the warranty period on deleted SKUs expires. Another 30 are point-of-sale, reviewed on an ongoing basis.
The commercial range is tight: 34 equipment SKUs drive 74% of revenue, and the top 25 of any type drive 80%. The issue isn't breadth — it's revenue per SKU, which the retrench addresses.
9. What does run-out cost, and how long does it take?
Roughly three years, and we now have it by SKU. There is $2.10M of stock across the eleven affected products. The longest tails are the Grain Mill and the 25L Sparge Water Heater at 36 months; G40, despite being the largest single holding at $696k, clears fastest at about 10 months. Detail on tab 03.
What we do know: G70 was deleted and still did $117k in the first eight months of this year. Three completed run-outs each took roughly three years from peak. Expect a multi-year tail.
That tail is also why the Next 12 Months holds revenue while we delete products — we keep selling the deleted lines through the period, so costs come down before the revenue does. It buys time. It is a short-term benefit that reverses: part of next year's revenue comes from products that won't be there the year after.
The alternative is writing off saleable stock, which we don't recommend. A dated, costed plan for the whole portfolio — including write-off estimates — comes to the first board meeting of 2027.
10. Isn't holding AU/NZ with reduced investment just a slow exit?
Partly, and we won't pretend otherwise — it may well end there. If the delivery model doesn't work and the numbers keep declining, reduced investment is the road to an exit. We would rather say that now than discover it later.
Decided: reduced investment. We pull back on digital spend, and there is no dedicated marketing resource or activities targeting the region. We do continue some digital spend on the brand there, but only where it makes financial sense. With the SMT: the delivery model, including whether to go D2C-only. The scenario work is still to do, but this is ours to settle and implement — we will report the outcome at the first board meeting of 2027 rather than bring it back for a decision.
FY26 revenue is $561k (AU $385k, NZ $177k), with NZ at roughly break-even on gross margin. Against that, it still consumes warehouse space and complexity — and Australia is the one market that hasn't shifted online, which weakens the D2C-only case there specifically. Anything that changes the warehouse footprint still has to be settled by the 2028 Key Brand Milestone, ahead of the 2029 leases — for the AU site particularly, given its size.
11. Why do we need the app?
Because it is why customers pay a premium — and our research says so specifically, unprompted. The app controls the equipment; competitors’ apps do not. Connected control is now an expected purchase driver, and it is the clearest thing separating a Grainfather from a cheaper copy. It is the USP the retrench is built to defend.
And its own cost is far smaller than it looks. The ~$165k shown at tab 08 is the whole digital stack, not the app. It breaks down as $16k app, $38k website (Shopify $36k plus $2k translation), and ~$50k of shared services — Klaviyo, Zendesk, Google Services and the Alumio integration — which is Grainfather’s apportioned share and so belongs in this brand’s costs.
The remaining $63k is Particle IoT hosting, a legacy contract ending March 2027. CloudAMQP is already in the stack as the replacement at about $750 a month, so the app’s steady-state infrastructure cost settles at roughly $25k a year.
But $16k is infrastructure only, and that is not the whole cost. Maintaining the app currently takes 1.0 FTE Dev Ops team member, engaged as a contractor. That sits in the labour line of the P&L, not in this stack. So the honest figure is a modest infrastructure cost plus a full-time resource — still defensible for the brand’s central differentiator, but it should be read that way rather than as a $16k commitment.
One caveat. The stack figure sits on draft FX rates, and the component split is a reasonable estimate rather than a reconciled allocation. Treat the totals as indicative.
12. What is the long-term growth story?
Near term, not new product development. The FY22 wave lifted the group 1–3% and mostly cannibalised G30; later updates coincided with group declines of 10–45%; NPD has contributed roughly nothing since FY24. So for the next two years the answer is recovering EU and US through better partners, and growing consumables around the G30 without major investment.
Longer term, we will build a roadmap for success in the retrenched category. We are not offering that solution today — we do not have it yet, and putting one forward would be a request for investment we are not making. One candidate is the Modular Brewing System, in early development, which could in time take the position the G30 holds now. It may or may not end up under the Grainfather brand.
Three years to market, so the roadmap informs the 2028 Key Brand Milestone rather than this meeting. All we ask today is that the decision doesn’t foreclose it.
13. Your own analysis said this was “not a deletion case.” Which is it?
Both, at different levels.
Not a deletion case at brand level — we are not recommending you delete Grainfather. The deletion is at product and region level, and it is what makes the concentration real. Concentrating on what earns means stopping what doesn't.
14. What happens if the EU recovery fails?
Europe is two independent bets: Brouwland, and the GF.com D2C site we've switched back on. They can fail separately.
If both fail, the revenue assumption goes with them — Europe carries most of the assumed growth. That takes the Next 12 Months from +$49k to roughly −$89k.
It doesn't make exit cheap. The ~$450k cross-brand exposure and the warranty tail are unaffected.
15. How does this line up with the stub year budget?
It doesn't align, and both papers are in front of you today. The periods are set out at tab 02, item 4.
This pack is on the current September–August year, so the forward figures cover the 12 months from September 2026 — spanning FY26S plus the first eight months of the new FY27. The $4,070k and the +$49k don't read across to either budget.
We haven't restated it. Re-cutting into a four-month stub and an eight-month remainder adds apportionment judgements and changes nothing about the decision. Most of the identified cost changes also don't commence until into the new FY27 — parental leave timing, the Particle close-out — so the shapes differ as well as the periods.
We brought this forward, and we ran it with dedicated focus
How the SMT reached this recommendation, and why the timing is earlier than you were originally promised.
→Why this is in front of you in August, not January 2027
A recommendation on Grainfather's future was originally scheduled for January 2027. At the April 2026 board meeting the Managing Director advised that the SMT would bring it forward to August 2026. The reasoning was straightforward: we did not believe the answer would change in the intervening six months, so there was no value in waiting and real value in acting sooner. Work started the following month and ran continuously from there — brainstorming, two rounds of pre-work, two working sessions and the actions between them, all against that self-imposed deadline.
April board meeting
MD informs the board that Grainfather will be brought to the table at the next meeting, rather than waiting until January 2027.
Initial SMT brainstorming
First pass on approach, and assembly of the data sets the work would need.
Pre-work for SMT Session 1
Analysis prepared and circulated so the session could start from evidence rather than opinion.
SMT Session 1 — shortlist the options
Reviewed pre-work and aligned on the strategic landscape, with all directional options considered and either kept in or removed. Restructure / Retrench was shortlisted for deeper investigation.
Working through SMT Session 1 actions
Closing out the questions the first session generated.
Preparation and pre-read for SMT Session 2
Second round of analysis built and circulated ahead of the session.
SMT Session 2 — stress-test the preferred path
The last working session before the board. Tested the forward P&L line by line, the G30 retrench and G40 exit case, cross-brand exposure, warehouse economics and the innovation pipeline. Left the room with a recommendation the group was willing to defend.
Board meeting — this recommendation
A specific, costed recommendation and a two-phase plan for approval — not a request for more time.
→Resolved in SMT workshop / session 1
| Option | What it means | Outcome and reasoning |
|---|---|---|
| Restructure / Retrench | An intentional, strategic refocus — deprioritise some markets, double down on others, and keep the channels that earn. | Recommendation Intentional focus: deprioritise ANZ, back EU and US, keep D2C given its ROI. Be selective, reduce costs, go hard on a limited portfolio. |
| Invest & Fix | Fund a turnaround — pivot the brand toward consumables and back new product development to rebuild it. | Held in view Not dismissed, but aware any request for investment in the brand would be a contradiction, and the SMT were not confident brand investment could deliver returns. |
| Harvest | Strip back all investment with no real strategy — do nothing, and hope the brand sustains itself. | Set aside Stripping investment with no strategy — judged a wind-down in disguise within 2–3 years, and likely not credible with the board. |
| Wind Down | Exit the brand entirely and close it out. | Struck off Too expensive: warranty tail, the inventory at risk across the whole range, and collateral damage to other brands. For scale, the eleven SKUs already heading for run-out hold $2.10M of stock on their own (tab 03); a full exit would put the rest of the range at risk alongside them. |
| Sell & Licence | Sell the brand outright, or licence it to a third party to operate. | Parked No realistic buyer today. Worth revisiting only after a turnaround — at which point the rationale weakens. |
→Resolved in SMT workshop / session 2
- The forward Forecast against an Adjusted scenario, and the assumption behind every cost action.
- The G30 retrench and G40 exit — unit economics, the price premium a larger system has to carry, defect and warranty exposure, and the historical ratio of roughly three G40s sold per ten G30s.
- EU cross-brand exposure, quantified at about $450k of other-brand revenue.
- Whether the ~$120k AU warehouse saving survives a like-for-like rent comparison. It does — the higher per-square-metre rate on a smaller footprint is already applied. Note this is not a near-term saving: it is only available at the 2029 lease renewal, and as an allocated indirect cost only Grainfather's revenue share of it would land in this brand's result. Detail and the underlying figures are at tab 08.
- Whether our early-stage NPD project should be framed as a Grainfather commitment. It should not — not yet, and possibly not ever.
The decline is real, global and volume-led — and the core still earns
→Eleven years of revenue and margin
→Gross margin, clean window only
Margin is only reliable from September 2024 onward. Before that, intercompany accounting through FY21–FY24 distorts it far enough that the brand shows a negative margin in FY23 — an accounting artefact, not trading. So we show two points rather than an eleven-year line.
| Basis | FY25 (NZD) | FY26 (NZD) | Read |
|---|---|---|---|
| Reported gross margin | 29% | 27% | The two-point step reads as deterioration. It is not. |
| Trading margin, excluding run-out write-offs | 30.5% | 30.4% | Flat. The core earns what it earned last year. |
The entire 2-point difference is inventory write-offs on stock being run out — G40 and the GC2/GC4 chillers — booked as cost against zero revenue (−$81k in FY26). That is the cost of the decision in front of this board, not erosion in the business that remains.
→Reconciling the three FY26 revenue figures
Three different FY26 numbers appear across our own working papers. Rather than quietly pick one, here is what each is. Our fiscal year runs 1 September to 31 August.
| Figure | Value (NZD) | What it actually is | Status |
|---|---|---|---|
| FY26 ACT+FCST | $3,603k | Actuals for the ten months to 30 June 2026 plus forecast for July and August, from the FY26 reforecast. This is the figure the board decision rests on and the one used throughout this pack. | |
| — actuals, Sep 2025 – Jun 2026 | $3,026k | Ten of twelve months, closed. Averages $303k a month. | |
| — forecast, Jul – Aug 2026 | $577k | Two months at $289k a month — slightly below the run rate of the ten closed months, not above it. The EU distributor ramp is the largest component. | |
| Legacy figure: $2,957k | $2,957k | Not a forecast vintage. The same ten months of actuals as the row above, taken from the transaction export before June had closed — June was $68.5k short at the moment of extract. Ten months of a twelve-month year, nothing more. Superseded. |
→A volume problem, not a pricing problem
Price and mix held up through the decline — the customers who remain still pay full price. Units fell from roughly 72k in FY19 to about 26k, while average price per unit has held near $140 throughout. The fall is in volume, not in what people will pay.
The pipeline that might have offset this has contributed roughly 0% of revenue since FY24, down from a 21% peak — and even that peak largely substituted for G30 sales rather than adding to them.
→What NPD actually did
The uncomfortable pattern, told straight:
- • G70 (FY20) — the only launch that clearly grew its family, at +72% volume. That uplift is distorted by the COVID effect.
- • The FY22 wave (G40, S40) — lifted the group just 1–3%. Mostly substitution, cannibalising G30.
- • Post-FY22 updates — coincided with group declines of 10–45%, not recoveries.
- • The exception: GCAST launched at 61% first-year margin — accessories, not equipment, are where the margin has been.
→The SKU count is not the problem — and this is the easiest wrong conclusion to draw
Nearly two-thirds of the SKU count is spare parts that we are legally obliged to keep available, generating 5.3% of revenue. The commercial range is far tighter than the headline count suggests.
| Category | SKUs | % of range | Revenue (NZD) | % of revenue |
|---|---|---|---|---|
| Equipment | 34 | 11.6% | $4,001k | 74.0% |
| Accessory | 31 | 10.5% | $978k | 18.1% |
| Spare part | 190 | 64.6% | $284k | 5.3% |
| Consumable | 9 | 3.1% | $143k | 2.6% |
| Point of sale | 30 | 10.2% | $2k | 0.0% |
| Total | 294 | 100% | $5,408k | 100% |
Concentration on the same data: the top 8 SKUs are 50% of revenue, the top 25 are 80%, and the top 56 are 95% — so 19% of the range produces 95% of the money.
→Large formats: a real two-year decline, then a self-inflicted stop
This distinction matters, because the two halves of the story prove different things.
| Calendar year | G40 + G70 revenue (NZD) | Change | What it shows |
|---|---|---|---|
| 2023 | $1,632k | — | Product fully available |
| 2024 | $1,221k | −25% | Genuine demand decline, product available |
| 2025 | $736k | −40% | Genuine demand decline, product available |
| 2026 to 30 Apr | −$16k | stop-sell | Net negative — credits and returns exceeded sales during the stop-sell. Not a demand signal. |
→EU cross-brand exposure, quantified
Grainfather appears in 47% of EU orders (322 of 679 over twelve months) and is more than 80% of order value in 117 of those, with a further 42 in the 50–80% band. It is a genuine anchor product.
| Anchor bucket | Non-GF revenue (NZD) | Assumed attrition | Est. revenue lost (NZD) |
|---|---|---|---|
| >80% Grainfather orders | $47,898 | 90% | $43,109 |
| 50–80% Grainfather orders | $278,816 | 50% | $139,408 |
| <50% Grainfather orders | $1,782,826 | 15% | $267,424 |
| Total at risk | $2,109,540 | — | $449,941 |
How exposed the other brands are
Share of each brand's EU sales that arrive inside a Grainfather order: On The Rocks 97%, Keg King 96%, Still Spirits 87%, Mangrove Jacks 54%. These are the brands that would take the sharpest volume hit.
And the acquisition effect
52% of EU customers' first order included Grainfather. Those customers have since spent $1.27M on other brands. Nine of 71 active EU customers are at least 50% dependent on Grainfather; four are over 80% dependent.
→The large-format thesis didn't hold — and the market moved the other way
This is the clearest read across all three of our evidence sources. When we invested in bigger formats we were betting that brewers wanted to scale up, and would pay for the privilege. Our own NPD sales, the market data and customer feedback all now say otherwise: that thinking has either turned out not to be true, or the market has moved on from it. Demand has gone toward smaller formats — G30-sized and below.
The mechanism matters, because it determines whether the problem is fixable. It is not that the larger systems are inferior — the G40 is the better machine. It is that a bigger, more capable system costs materially more to build, and the price it must therefore carry exceeds what enough brewers are willing to pay. The margin confirms it rather than driving it: large formats earn 26.7% gross margin against the core's 27.1% — essentially level, for a machine that costs materially more to make. We are already pricing below what the cost base warrants to chase volume, and still not winning it.
So stepping back from the large formats is not a retreat from a good position — it is correcting a bet that did not pay off. G70 is already out; where the G40 lands is for the Proposed Deletions Process to resolve. The G30 ecosystem sits on the right side of that shift and remains the part of the brand that earns, and the app ecosystem is a moat aligned with where appliances are heading: rivals compete on price, not platform.
These are directional, research-sourced proof points (Brewers Association/AHA, Grand View Research, Future Market Insights, IWSR, Star Tribune, Allied Market Research). Market-size estimates vary between firms and should be verified against primary sources before any external use. They corroborate the recommendation; they do not drive it.
A path to breakeven before indirect costs — and what that does and doesn't mean
The forward numbers, what drives the improvement, how sensitive it is, and what sits below the line.
It also goes a step further than that report. The depth of work behind this review let us identify further costs that attach directly to the brand — shown here as Additional direct costs — which the routine report does not split by brand. The result is a cleaner view of what the brand carries before any overhead is allocated to it.
Indirect costs — management and administration, property and warehouse rent, and other shared functions — sit below this line and are allocated to Grainfather separately. So the +$49k in the Next 12 Months column is not a profit forecast. It says the brand covers its own direct costs with a small margin. Once its share of company overhead is applied, Grainfather remains loss-making over that period. The mechanics are at † below.
→Next 12 Months P&L — path to breakeven before indirect costs
| NZD 000s | FY26 Est 12 mths |
Next 12M Forecast |
Next 12M Adjusted* |
Further 12M Estimated |
|---|---|---|---|---|
| Revenue | 3,603 | 4,070 | 4,070 | 4,300 |
| Cost of sales | 2,631 | 2,884 | 2,884 | 2,924 |
| Gross margin | 972 | 1,186 | 1,186 | 1,376 |
| GP % | 27% | 29% | 29% | 32% |
| Labour (marketing, IT, engineering) | 539 | 561 | 362 | 300 |
| Marketing | 208 | 241 | 196 | 150 |
| Direct costs | 747 | 802 | 558 | 450 |
| GP post direct costs | 225 | 384 | 628 | 926 |
| GP % post direct costs | 6% | 9% | 15% | 22% |
| Distribution | 132 | 149 | 149 | 155 |
| Inventory write-offs / warranty | 184 | 203 | 203 | 194 |
| IT | 166 | 166 | 103 | 75 |
| 3PL — US/EU | 100 | 147 | 126 | 150 |
| Additional direct costs | 582 | 665 | 581 | 574 |
| Profit before indirect costs | (357) | (279) | 49 | 353 |
Direct costs is labour and marketing only — the two the business identifies by brand as a matter of routine, and the same definition that report uses. On that line the two line up: $747k here against $751k in the report’s January–December view. Different windows, but labour and marketing move little between them, so the definitions clearly agree.
The other lines do not tie, and are not meant to. This column is FY26, September to August, on the current forecast. The report carries both a January–December view and a September–August one — and that September–August column was prepared for an earlier board, on the forecast as it stood then, so it reads lower than this. Period and vintage both move revenue and gross margin, which is why the comparison holds at the direct costs line and not above it.
Additional direct costs — distribution, inventory write-offs and warranty, IT and 3PL — are not costs we split by brand in the ordinary course. The depth of work behind this review let us extrapolate them, which is why this pack can carry a profit before indirect costs line that the routine report cannot. It is additional detail, not a different basis.
Line items are rounded to the nearest $1k and carry a $1–2k rounding difference from the profit line, inherited from the source workings.
- Revenue — $4.3M. Back to FY25’s level on a materially better mix: G40 fully sold through, G70 likely the same, and the Grain Mill gone or held only for genuine sales rather than discounting. The US and Europe performing, with extra volume through D2C.
- Gross margin — 32%. Fewer higher-cost SKUs and more lower-cost, better-margin ones, with D2C a larger share of brand sales at a better margin.
- Distribution. Held at the same percentage of revenue.
- Inventory write-offs and warranty. Broadly unchanged — we are still clearing spare parts on discontinued lines.
- Labour. The brand manager role fully out, and Dev Ops weighted towards other brands.
- Marketing. Lower spend overall, and more work brought in-house using AI.
- IT. Shrinks with fewer regional connections for the brand.
- 3PL. Similar rate of cost.
†Indirect costs, how they're treated, and why brand profit can only ever be an indication
Indirect costs — management and administration, property and warehouse rent, other shared functions — are allocated across brands by share of revenue, and sit below the line in the table above. So +$49k is not profit: it means the brand covers its own direct costs with a small margin, and remains loss-making once its share of overhead applies. The principle is set out at tab 02, item 3. Three consequences matter when reading the figures on this page:
- Grainfather's fully-allocated result moves with other brands' performance, not only its own. If other brands grow while Grainfather doesn't, Grainfather's share of group revenue falls, it carries less indirect cost, and its bottom line improves — with no change whatsoever in how Grainfather actually traded. The reverse is equally true. Movement in fully-allocated brand profit, in either direction, is partly allocation arithmetic rather than performance.
- The revenue growth in this plan draws more indirect cost with it. If Grainfather grows 13% while other brands hold flat, its revenue share rises and it carries a larger slice of the indirect pool — partly offsetting the improvement the growth is meant to deliver. The line in the table above is unaffected, because indirect costs sit below it; the fully-allocated result is not.
- The loss is not a cash sum we would recover by exiting. Grainfather's indirect cost allocation exists whether or not Grainfather exists. Closing the brand would not remove those costs — it would redistribute them across Still Spirits, Mangrove Jacks, Keg King, On The Rocks and the rest, making each of them look worse by the same arithmetic. The genuine saving from exit is only the cost that is specific to Grainfather. This is a further reason Wind Down is not the saving it appears to be, over and above the cross-brand revenue risk on tab 07.
For these reasons, brand-level profitability here should be read as a directional indication and a forecast rather than a precise result. Two of the three variables that determine it — other brands' revenue, and the size of the shared cost pool — sit outside this brand's control.
→How much revenue growth does this line actually need?
This is not in the SMT Session 2 deck and we think you should have it. Holding the Adjusted cost base and the 29% gross margin constant, and flexing only revenue — note this is the profit before indirect costs line, so zero here is not brand profitability:
| Revenue scenario | Revenue (NZD) | Growth on FY26 | Gross profit (NZD) | Profit before indirect costs (NZD) |
|---|---|---|---|---|
| Budget as planned | 4,070 | +13.0% | 1,186 | +47 |
| Breakeven on this line | 3,908 | +8.5% | 1,139 | 0 |
| Half the planned growth | 3,837 | +6.5% | 1,118 | −21 |
| Revenue holds flat at FY26 | 3,603 | 0.0% | 1,050 | −89 |
The read: the cost actions do most of the work — they close roughly $330k of the gap — but they do not by themselves get the brand to zero even on this line. The plan still needs about 8.5% revenue growth, against 13% assumed, leaving roughly four and a half points of headroom. Given FY26 came in about 9% below its own budget, that headroom is thin, and the board should treat the revenue assumption as the principal risk in this plan.
And the harder read: every figure in the right-hand column is before Grainfather's share of indirect costs. Full brand profitability requires clearing this line and then covering that allocation — which needs materially more revenue than +8.5%, or a smaller group overhead pool, or both. That is beyond this plan's 12-month horizon. What this plan does is get the cost base right; profitability then depends on the revenue following.
→The cost actions behind "Adjusted"
- Labour — UK brand manager reduced to 0.3 FTE during maternity leave (0.3 to cover mat costs); app developer modelled at 0.5 FTE but since resolved to remain full-time, funded partly from other project work. Excludes any customer service time.
- Marketing — spend not allocated to the D2C platform or NPD is cut.
- IT — $63k of Particle IoT hosting removed from March 2027 when the contract ends.
- Warehousing — partial 3PL reduction ($126k against $147k budget) from moving US/EU logistics from Mainfreight to MoreFlavor.
→What the revenue assumption rests on
Growth from FY26 to the next 12 months is attributed principally to Europe (about $1.9M total, 22% growth) and D2C (about $1.1M total, 22% growth). These depend on three things performing as expected, none of which has a track record with us yet: the Brouwland partnership, the GF.com D2C site now switched back on in Europe, and the US distributor transition.
Gross margin is assumed at 29%, based on FY25. FY26 came in at 27%, held down by grain-mill promotions across both D2C and distribution channels. If FY26's 27% persists rather than recovering to 29%, that is a further ~$87k of gross profit against the plan.
→What is not in the +$49k
Named explicitly so none of this is a surprise later:
- Customer Service employee costs — unresourced in the forward numbers. Support and warranty obligations do not shrink with the brand (tab 04). The team's preferred lever is AI to reduce headcount time rather than reducing service quality.
- Early-stage NPD engineering costs — excluded, and in any case not a committed Grainfather investment. See the innovation section on tab 03.
- The SKU rationalisation's own impact — the revenue and margin effect of whatever the Proposed Deletions Process decides for the G40, the operational cost of running it out, and any inventory write-down risk.
- Engineering capacity reality — the engineering line assumes no change to current allocation to Grainfather, which has been running at 30–40% business-as-usual, the rest on other brands. Shared capacity, not dedicated.
- Warehouse rent reduction — an upside, but not before 2029. Both sites could be downsized at renewal — Brisbane on a seven-year term, NZ on eleven. Both are gated on the 2029 lease renewals — the leases run until then regardless of what is decided about the brand, so none of this is available in the near term. Note too that rent is an allocated indirect cost, so only Grainfather's revenue share of any saving would land in this brand's result (see † above).
→Warehousing — the footprint behind the 2029 decision
| Site | Current | Utilisation | Smaller premises |
|---|---|---|---|
| Brisbane Airport (AU) | 2,600 sqm, ~$420k/yr avg ($161/sqm), 7-year term | 52% | ~1,500 sqm at $200/sqm ≈ $300k/yr |
| Rosedale (NZ) | 4,320 sqm, ~$637k/yr avg ($147/sqm), 11-year term | 75–80% | ~3,000 sqm at ~$200/sqm ≈ $600k/yr |
The AU case was challenged in SMT Session 2 on exactly the right point — whether the higher per-square-metre rate on a smaller footprint had been applied. It had, and the case holds provided a comparable $/sqm base is maintained. Note also the sunk leasehold investment: $560k in Brisbane (2022) and roughly $500k in NZ, which is part of why exit is not free.
A note on who benefits. Any downsizing gain is a business-level property saving, and rent is one of the shared costs allocated across brands by revenue share (see † above). The full benefit accrues to Bevie Handcraft; the portion that shows up in Grainfather's own brand P&L is only Grainfather's allocated share of it. Both framings are legitimate — worth being precise about which one is being quoted, because the two are materially different numbers.
→D2C marketing — the return, on both bases
Return on spend has fallen sharply but remains positive. We show margin return as well as revenue return, because margin is the decision-relevant number and it fell considerably further.
| Period | Spend + retainer (NZD) | Revenue (NZD) | Revenue return | Margin (NZD) | Margin return |
|---|---|---|---|---|---|
| FY25 | $36,737 | $545,128 | 14.8× | $275,818 | 7.5× |
| FY26 — 9 months actual | $123,410 | $662,518 | 5.4× | $285,777 | 2.3× |
| FY26 — full-year forecast | $163,905 | $867,940 | 5.3× | $373,214 | 2.3× |
Spend rose more than four-fold while margin dollars stayed almost flat — $275.8k to $285.8k across the comparable nine-month periods. The team attributes the fall largely to grain-mill giveaways skewing the numbers, plus a genuine step-up in the cost of buying market control and data, versus what was previously described as "stabbing in the dark." At 2.3× margin return it remains clearly ROI-positive, and the team's judgement is that this is a value trade-off rather than a loss — but the trend is the thing to watch. Worth noting what this data does not show: the other benefits the brand gets from the increased work — sharing of brand, building of communication databases, and similar.
→The digital stack — app, D2C and shared services
The app is a genuine differentiator — competitors’ apps do not control the equipment. It is also the smallest part of the stack it sits on. The stack costs roughly $165.5k/year ($13.8k/month); excluding Particle the run rate is about $102.6k/year.
| Component | Per year | What it is |
|---|---|---|
| The app | $16k | Infrastructure only. Excludes the resourcing to maintain it — currently a 1.0 FTE Dev Ops team member on contract, carried in the labour line rather than here. |
| Particle IoT hosting | $63k | Legacy contract, ends March 2027. CloudAMQP is already in the stack as the replacement at about $750/month, so the net saving is real but smaller than $63k. |
| Website / D2C | $38k | Shopify $36k plus $2k translation. Needed with or without the app. |
| Shared services | ~$50k | Klaviyo, Zendesk, Google Services and the Alumio integration. This is Grainfather’s apportioned share, so it belongs in this brand’s costs. |
Particle's contract ends 31 March 2027, after which it drops to roughly $500/month or a free plan — the source of the $63k saving in the Adjusted plan. Note that CloudAMQP, already in the stack at about $750/month, is described as the Particle replacement, so the net saving is real but smaller than the gross $63k implies.
Slice it yourself
An interactive deep-dive built on real figures — filter by region and group, walk the launch timeline, compare before and after each product introduction, and check productivity year by year. FY16–FY25 are actuals recomputed from source. Updated 27 July 2026: FY26 in this tool now carries the $3.60M ACT+FCST reforecast — actuals to 30 June plus forecast for July and August — applied to each region on its existing category mix, so it agrees with every other figure in this pack. Margin has been removed from this tool entirely, including the before→after table, because the FY21–FY24 series is intercompany-distorted. In Brewing Accessories, Grain Mill and GCAST are excluded from the before→after rows — Grain Mill data is distorted by promotional stock clearance, GCAST is too small to read — so accessory launch evidence rests on the SWH family. Reconciliation on tab 07.