The Elevate
Playbook

Commercial outcomes

Operating outcomes

What the table changed

1. Family Wellness was under-served by a factor of four. It was 16% of customers and 27% of contribution, with the highest retention in the book — and had no dedicated bundle, no dedicated creative and no email journey. The Family Pack SKU exists because of this row.
2. First-Timers were being over-acquired. Cheapest CAC, worst everything else. Paid social had optimised into them because they convert fastest. Bid strategy was moved off first-purchase conversion and onto predicted 180-day value, and blended LTV rose 19% in two quarters with no change in spend.
3. Active Performance justified a price increase. Highest AOV, second-highest retention, lowest price sensitivity in testing. The Performance bundle was repriced up 12% with no measurable change in conversion.
4. Travellers convert on format, not formula. They are the only segment where the stick pack outsells the tub on first order — 3.4:1. Serving them tub creative was costing an estimated $340K a year in wasted impressions.

1. The subscription price is the real price. 68% of orders are subscription orders, so the subscription price is quoted everywhere as the headline. The one-time price exists, is visible, and is never the number on the ad.
2. No promotional discounting, ever. Smart Greens has run zero sitewide sales since March 2024. The welcome kit — shaker, scoop, five travel sticks, $34 of physical value at $11 of cost — replaced the discount entirely. Contribution per first order rose $9.20 and the customers acquired retain 8 points better.
3. Volume value is structural, not promotional. Per-serve price falls from $3.07 on sticks to $2.23 on the Family Pack. The discount lives in the range architecture, where it builds order value permanently, rather than in a campaign, where it trains the market to wait.

Three models, one business

• The legacy channel model pays commission on introductions and settles on a monthly cycle governed by a legacy commission structure that can change.
• The direct model pays nobody, needs subscription billing, dunning, proration and self-serve amendment, and settles instantly.
• The creator model pays a flat or tiered commission to a partner outside the legacy channel, on verified sale rather than click, in whatever market they happen to live in.

Run these on separate systems and the failure mode is immediate and specific: a customer introduced by the legacy channel, who later buys through a creator's storefront, whose subscription then renews direct. Who gets paid? Under Smart Greens' 2023 stack the answer was nobody, twice, and a ticket was filed. That exact scenario occurred 1,840 times in 2023 and was resolved manually every time.

What Elevate One resolved it with

1. A single commercial ledger. One order record carries every attribution claim against it — introducing legacy channel partner, referring creator, acquisition channel, subscription lineage — and the payout engine resolves them by rule rather than by ticket.
2. A configurable attribution hierarchy. Smart Greens' rule: the introducing legacy channel partner retains a 4% commission for eighteen months regardless of subsequent channel; the creator earns full commission on the order they source. Both are paid. The margin cost is 2.1 points and it ended the conflict permanently.
3. Payout as a shared service. Legacy channel partners, creators and affiliates are paid by one engine across six markets and four currencies, with the tax and licensing logic held once rather than per-channel.
4. Legacy commission structure as configuration. Plan rules moved out of code. The change that took seven months in 2022 was made in six days in 2025, and tested against a twelve-month replay of live orders before it shipped.

The two findings that decided it

1. Standing still was not free — it was $26.6M. The default option carried almost the full cost of the change with none of the benefit, because the integration overhead and agency dependency continue whether or not anything improves. This single line moved the decision more than any projection.
2. Building in-house cost more than standing still. Not because engineers are expensive, but because 23 specialist hires at 4.5 months average time-to-hire means the team is not complete until month fourteen — and the first nine months produce cost with no revenue. Two of the three most senior roles took over seven months to fill in the modelling, which is optimistic.

Programme metrics, rolling twelve months

The four rules that make it work

1. One workflow, not campaign-by-campaign. A creator moves from discovery to first payment without a spreadsheet or a manual approval. Median time from outreach to activated storefront is nine days; under the 2023 process it was never completed.
2. The honest check-in is a required deliverable. "30 Days of Better Routines" asks creators to document a real month, and the day-20 check-in must address what they would change. Content that admits friction converts at 5.8% against 3.1% for uniformly positive content — a finding that survived three quarters of testing.
3. Commercial infrastructure, not a discount link. Every creator gets a storefront, a code, curated bundles, and live commission and payout visibility. Payout is seven days. This is why 62% grant paid rights: they can see what the content earns.
4. Rights are captured at brief, not negotiated after. Usage terms are agreed before production. The 62% clearance rate against an industry norm nearer 25% is almost entirely a process artefact, and it is what lets the creator programme feed the ad account.

Podcast sponsorship

1. The category's most efficient channel, and the one direct sellers systematically overlook. 34 shows, host-read, mid-tier — health, endurance and business — chosen for host trust rather than audience size.
2. Show-specific landing pages and codes mean attribution is exact rather than modelled. Podcast CAC is $52 and the channel produces the second-highest subscription take-up in the book at 76%.
3. The offer is a format, not a discount. A free travel pack with first subscription converts listeners without training the market to wait for a sale.

Giveaways and collaborations

1. Adjacent, never competing. Equipment, apparel and whole-food brands — audiences that overlap on values without overlapping on spend.
2. Entry grows the list, not the discount habit. Follow both brands and join the newsletter; no purchase, no code. 14 partnerships added 118,000 email subscribers at an effective cost of $0.41 each.
3. Prizes are product-led — the Starter Kit plus the partner's equivalent — so winners experience the routine rather than redeem a voucher.

Community and IRL events

1. Sampling converts where argument fails. Taste is the single most common objection in cancellation-reason capture at 23%, and it cannot be answered with copy. 96 activations across six markets last year, sampling 41,000 people.
2. The brand hosts, with creator partners. Activations are run by the brand and its creator partners, which keeps the customer-facing surface consistent and gives the creator channel a live moment to build content around.
3. Every activation is tracked. Location-specific codes and landing pages mean in-person activity is attributed like any other channel. Event CAC is $31 with a 7.9% same-week conversion rate.
4. Events feed the content engine. Activation footage becomes organic content and paid creative, amortising the cost across three channels rather than one.

The journal

1. The compounding asset. 180 expert-reviewed articles, 2,400 ranking long-tail queries, 640,000 organic sessions a quarter at zero marginal cost per click.
2. Articles feed email, not the reverse. Long-form is written first and syndicated into lifecycle journeys, which keeps the education layer consistent and halves the content workload.

The welcome flow — the highest-leverage sequence in the business

1. Opens with the quiz result, not a discount. The first email reflects the customer's own answers back as a recommendation — personalisation they can verify rather than a claim they have to accept. 58% open rate.
2. Second email is the ingredient breakdown. "What's actually in the scoop." Education positioned before any commercial ask.
3. Third addresses objections directly. The three questions everyone asks before a first order, answered without deflection — including the one about taste.
4. The offer arrives last and stays modest. By design, the discount is not the reason the sequence gets opened. Removing the discount from email one and moving it to email four raised sequence conversion from 7.9% to 11.2%.

Retention — where the LTV gain actually came from

1. Replenishment timing is per-customer, not per-SKU. Predicted from actual consumption signal rather than a fixed 30 days. Failed-to-reorder rate fell 6.1 points.
2. Failed-payment recovery is the cheapest retention there is. Dunning with card-update prompts recovers 71% of failed renewals — worth $2.9M a year, at essentially no marginal cost.
3. Cancellation-reason capture drives the roadmap. Taste 23%, cost 21%, forgot to use it 19%, travel 11%, other 26%. Each reason routes to a different save offer, and the Hydration+ flavour work came directly from that first line.
4. Win-back waits. The series does not fire for 90 days. Testing at 30 and 60 days produced higher unsubscribes and lower conversion in both cases.

1. Launching paid acquisition before the creator and referral infrastructure existed. ($1.4M and nine weeks.) The original plan ran acquisition first because it showed revenue fastest. With no organic proof for paid to draw on, CAC ran 40% above plan for nine weeks and the account was paused while the creator and referral infrastructure was completed. The rule that came out of it — creator and referral proof ships before paid — is the single most valuable line in this playbook and it was learned the expensive way.
2. A twelve-question finder. (Six weeks of build, discarded.) The first version was thorough and completed by 24% of starters. Cutting it to six questions raised completion to 62% and recommendation accuracy did not measurably fall. Questions seven through twelve were being answered by people who were going to convert anyway.
3. Urgency mechanics. ($0 net, real damage.) A countdown timer and a live-activity popup were tested in 2024. Both lifted immediate conversion — 6% and 4% — and both produced measurably worse month-three retention. Net twelve-month contribution was negative in both cases. They were removed and the practice was banned in the brand guidelines.

4. Paying creators on clicks for one quarter. ($210K of leakage.) The first commission model paid on tracked clicks because it was simpler to implement. It rewarded volume over relevance within three weeks and attracted precisely the wrong partners. Moving to verified-sale commission, refund-adjusted, cost a quarter's rework and removed an estimated $340K a year of ongoing leakage.
5. Translating creative rather than localising it. (Two markets, four months.) Markets three and four launched with subtitled versions of home-market assets. CAC ran 41% above forecast until casting and copy were rebuilt locally. The saving on production was roughly $80K; the cost in wasted media was closer to $600K.
6. Under-resourcing creator onboarding in year one. (Ongoing.) The creator programme scaled to 340 partners because the workflow was rebuilt in Scout, but activation coaching for new creators was under-funded from the start — the honest check-in, the storefront setup, the first brief. Closing that gap is modelled at a 15–20% lift in per-creator productivity across the current cohort.

The platform

• Unified commercial ledger — legacy channel, direct and creator settling against one customer, one inventory position and one tax model across six markets.
• Configurable legacy commission structure — rules held as configuration, testable against a replay of live orders. Seven months to six days.
• Payout as a shared service — legacy channel partners, creators and affiliates paid by one engine, four currencies, seven-day cycle.
• Subscription engine — self-serve pause, skip, swap, two-tap cancel, per-customer replenishment prediction and dunning.
• Scout — AI-assisted creator discovery, outreach, CRM, gifting, UGC tracking, campaign operations and payment in one workflow, with rights capture at brief.
• Storefronts at scale — provisioned as shared infrastructure for creators and affiliates, not built per-partner.
• Guided selling — the finder, the segment model and the proof-matching engine behind the result page.
• Lifecycle automation — journeys that can reference subscription state, segment and legacy channel relationship in one message.
• One data layer — eleven systems to three; 31-hour attribution latency to real time.

The delivery model

1. A team that had done this before. Six specialist roles were hired internally rather than twenty-three, because the platform came with the operating knowledge attached. The recruitment timeline — 4.5 months average per specialist role — is the hidden cost that sinks in-house programmes, and it was the deciding factor in the board comparison.
2. Sequencing as the deliverable. The order of the build was the most consequential decision in the programme, and it is the thing a platform alone cannot supply. Storefronts before paid social is not a feature; it is judgement about how these businesses actually fail.
3. Nineteen weeks to first incremental revenue, against a modelled 9–12 months for the in-house path.
4. $15.3M over three years against $30.9M to build it and $26.6M to do nothing.

Which inputs actually dominate

1. Subscription share is the highest-leverage input. It moves recurring revenue, retention and the enterprise value multiple simultaneously. It is the only one of the six that changes what the business is worth per dollar of revenue rather than just how many dollars there are.
2. Cost to serve is the one most businesses cannot state. Smart Greens' figure was distributed across eleven vendors, four departments and two capitalisation policies. Establishing it took four weeks and it is what funded the programme — the argument was a reallocation, not a capital request.
3. Conversion rate is the cheapest to move and the most ignored. It required no additional media spend and applies to every future visitor. Replacing the category grid with a guided finder did most of the work.
4. CAC is a constraint, not a goal. Driving it lower by acquiring cheaper customers is how the paid account originally optimised itself into the worst-retaining segment in the book. It is read alongside twelve-month value or not at all.

What the six produce

1. Revenue composition, which is the actual objective — 39% recurring at Q2 2026 against 8% at the start.
2. The enterprise value case. Brands with concentrated legacy revenue trade at 0.8–1.4× revenue; hybrid businesses with meaningful recurring revenue at 2.5–4×. The same $500M business is worth roughly $600M under one model and $1.6B under the other.
3. The build-versus-buy comparison, since all three paths in that table are the same six inputs run forward under different cost and timeline assumptions.
4. The pace of the transition. Nineteen weeks to first incremental revenue was a function of sequencing, not of spend.