# Flo & Frankie: how every number in the strategy was built

This file is the full derivation of the money model in the Search & AI Strategy, written so
you, or your own AI assistant, can verify every figure against your own accounts. Two markets
run through the same chain independently: New Zealand (floandfrankie.com, NZD) and Australia
(floandfrankie.com.au, AUD). The two currencies are never added together anywhere.

Data windows: Google Search Console 2 September 2025 to 2 September 2026, per property.
GA4 twelve months to 4 September 2026 for channel totals, and the last 90 days for
landing-page economics. Keyword volumes pulled September 2026. Ahrefs pulled 4 September 2026.

## Step 1: the keyword basket, triangulated across three sources

We built a basket of head terms for the categories you sell (dresses, denim, knitwear,
outerwear, footwear, bags, jewellery, homeware, gifts, store-level terms) and pulled each
term's monthly search volume from three independent sources per market: Google Keyword
Planner, SEMrush and Ahrefs. The working volume per term is the **median of the three**,
because each tool is wrong in a different direction and the median discards the outlier.

- **102 generic head terms** per market, plus **3 terms proven in your own Search Console**
  (cocktail glasses, wine glasses, stocking fillers, added after a gap-check sweep found you
  already earning clicks on them: cocktail glasses alone earns 513 NZ clicks a year at
  average position 7.4).
- **20 stocked-brand terms** (frank green, assembly label, nude lucy, status anxiety, saben
  and others) are triangulated the same way but **held out of the money model entirely**,
  because someone searching a brand name may want the brand's own site. They are reported as
  a separate territory: 63,210 searches a month in NZ, 399,370 in AU.

The generic core, the only thing the money model uses:

| | NZ | AU |
|---|---|---|
| 102 triangulated generic terms | 217,660 /mo | 1,064,560 /mo |
| 3 GSC-proven additions | 4,690 /mo | 20,700 /mo |
| **Measured head-term core** | **222,350 /mo** | **1,085,260 /mo** |
| Core searches per year (x12) | 2,668,200 | 13,023,120 |

The long tail beyond these terms is real (your NZ account alone wins 225,579 named generic
clicks a year, far beyond this basket) but it is deliberately excluded from the pool, so the
share figures below stay honest and every volume traces to three sources.

## Step 2: the winnable click pool

Not every search ends in an organic click: ads, shopping carousels, AI answers and
zero-click results take their share first. We model **35% of searches surviving as winnable
organic clicks**:

- NZ: 2,668,200 x 0.35 = **933,870 winnable clicks a year**
- AU: 13,023,120 x 0.35 = **4,558,092 winnable clicks a year**

## Step 3: the share you hold today, measured, not assumed

We matched every core term, plus its geo-suffixed and prefixed variants ("linen pants",
"linen pants nz", "womens linen pants", "buy linen pants"), against the top 25,000 queries
in your own Search Console per market (each query classified as own-brand, stocked-brand or
generic; only generic-classified queries count, and each query counts once). Clicks you
actually won on those queries over twelve months:

- NZ: **47,370 clicks** / 933,870 pool = **5.07% share** (reported as 5.1%)
- AU: **318 clicks** / 4,558,092 pool = **0.007% share** (reported as ~0%)

## Step 4: the value of one non-branded click

GA4 does not label a session branded or non-branded, so we split organic traffic by landing
page: sessions landing on the **homepage are treated as branded** (people who searched your
name and clicked the top result) and sessions landing anywhere else as non-branded. This is
a stated simplification in both directions. Last 90 days, per market:

| Step | NZ | AU |
|---|---|---|
| Organic sessions, 90 days | 241,274 | 26,129 |
| Homepage (branded proxy) sessions | 71,442 | 5,457 |
| Non-homepage sessions | 169,832 | 20,672 |
| Homepage transactions | 1,811 | 184 |
| Non-homepage transactions | 2,791 | 450 |
| **Non-branded conversion rate** | 2,791 / 169,832 = **1.6434%** | 450 / 20,672 = **2.1769%** |
| Organic revenue, 12 months | $2,480,235.51 | A$377,863.31 |
| Organic transactions, 12 months | 16,963 | 2,634 |
| **Average order value** | **$146.21** | **A$143.46** |
| **Value per non-branded click** | 1.6434% x $146.21 = **$2.4029** | 2.1769% x A$143.46 = **A$3.1228** |

The same homepage proxy splits the 12-month organic revenue: the homepage carries 40.97% of
NZ organic revenue (90-day ratio applied), so the non-branded slice is
$2,480,235.51 x 0.5903 = **$1,464,060.58 a year**. In AU the homepage share is 30.12%, so
the non-branded slice is **A$264,061.40** of the A$377,863.31 total.

## Step 5: the scenarios

Revenue per scenario is simply: **pool x share x value per click**. New revenue is the
clicks gained over today, times the click value. NZ scenarios grow the measured 5.07% share
by +4, +8 and +13 points, justified by measured striking distance (page one to two positions
across most of the head terms). AU scenarios are absolute 1%, 2.5% and 5% shares, deliberately capped low
because AU starts from near-zero authority (DR 30, 787 AU keywords).

**New Zealand (NZD):**

| Case | Share | Clicks / yr | Revenue / yr | New revenue / yr |
|---|---|---|---|---|
| Today | 5.07% | 47,370 | $113,824 | - |
| Conservative | 9.07% | 84,725 | $203,583 | +$89,759 |
| Mid | 13.07% | 122,080 | $293,342 | +$179,518 |
| High | 18.07% | 168,773 | $405,540 | +$291,716 |

Each 1% of the NZ pool = 9,338.7 clicks x $2.4029 = **$22,440 a year**.

**Australia (AUD):**

| Case | Share | Clicks / yr | Revenue / yr | New revenue / yr |
|---|---|---|---|---|
| Today | 0.007% | 318 | A$993 | - |
| Conservative | 1.0% | 45,581 | A$142,342 | +A$141,349 |
| Mid | 2.5% | 113,952 | A$355,854 | +A$354,861 |
| High | 5.0% | 227,905 | A$711,708 | +A$710,715 |

Each 1% of the AU pool = 45,580.9 clicks x A$3.1228 = **A$142,342 a year**.

## Step 6: the twelve-month curve

The compounding chart plots the mid case arriving per market: today's non-branded revenue
plus 10% of the mid-case new revenue by month 3, 35% by month 6, 70% by month 9 and all of
it by month 12. The month-12 value equals the mid-case table row exactly, in each market's
own currency. The two lines are never summed.

## What is deliberately NOT in this model

- **The customer-value ladder (first-year and lifetime spend).** GA4's purchaser counts here
  are cookie-based and report 98% of buyers as first-time (93,278 of 95,011 NZ purchasers),
  which is not a credible read of repeat behaviour for a fashion retailer with 22 boutiques.
  Rather than price repeat purchases on bad data, every figure in the strategy is
  **first-order only**. The ladder gets added from your Shopify customer cohort reports when
  that access lands, and because repeat spend sits on top of everything here, first-order is
  a floor.
- **Stocked-brand demand** (63,210 /mo NZ, 399,370 /mo AU). Partially addressable at best,
  so it is a strategy territory, never revenue in the model.
- **AI prompt demand** (est. 39,400 to 48,500 /mo NZ, 192,100 to 236,400 /mo AU across the
  132-prompt library). Counted separately from search so nothing is double-counted; the
  bands are directional floors anchored to the same keyword volumes (12% chat share,
  1.3x to 1.6x fan-out uplift), replaced by observed citation tracking once measurement runs.
- **Branded revenue.** Your branded organic revenue sits outside every scenario, unchanged.

## Verify it yourself

1. Search Console, floandfrankie.com, 2 Sep 2025 to 2 Sep 2026: total clicks about 967,157.
   The AU property over the same window: about 56,708.
2. GA4 NZ property, twelve months: Organic Search about $2,480,236 on 16,963 transactions
   (branded and non-branded together). AU property: about A$377,863 on 2,634.
3. Search Console NZ, query "maxi dress": average position about 18.7; "maxi dress nz":
   about 4.2. That pairing is the striking-distance pattern the strategy is built on.
4. Ahrefs (or any DR tool): floandfrankie.com.au DR 30 with 787 AU keywords, against
   dissh.com.au at DR 44 with 7,588 and sportsgirl.com.au at DR 55 with 12,872.
5. GA4 NZ, source chatgpt.com, twelve months: about 6,280 sessions, 92 transactions,
   $13,945 revenue. AU: 2,633 sessions, 33 transactions, A$3,884.