Be the face paint brand
AI recommends.

1. Executive summary

• You already lead the niche, on authority the field cannot match. Body FX carries the highest domain rating of any NZ specialist in this category (26, against 2.9 for the nearest face paint rival) and holds 92 keywords in the NZ top 3, a 35% top-3 hit rate. Yet Costume World, at a domain rating of 10, ranks for 8 times as many keywords, purely on catalogue breadth. The authority is earned. The pages are unbuilt. Ahrefs
• The answer layer has no NZ incumbent, and AI is filling it from offshore. Across 11 real buying and how-to questions we ran through live answers, Body FX was cited in 2. Every technique and safety question came back with US, UK and AU publishers. The sharpest one: you hold 146 Ben Nye SKUs, the deepest range in the country, and "where can I buy Ben Nye in NZ" is answered with two smaller rivals. Where a real page exists (wig rental, bio glitter) you already get cited first, so the mechanism works. The pages do not exist. AI answer scan
• October is a cliff, not a ramp. October alone is 27% of the year's searches, 3.3 times an average month, and 22% of your online revenue. September runs only 30% above baseline, then October triples, so rankings and AI answers have to be in place before September to count. That makes this a build for October 2027, the first season you can meet with a full answer layer instead of a buy layer alone. GKP monthly Shopify

So the whole opportunity is one sentence: you hold 28.3% of the non-branded organic clicks in your category, and every extra 1% is worth about $1,208 a year in first orders and roughly $2,661 a year once a customer's repeat buying is counted. Your branded traffic is not part of this and does not change. Because you start from the lead, the growth this plan asks for is consolidation of a market you already head, not conquest of somebody else's.

Each non-branded organic visit is worth $1.48 in first-order revenue, measured from your own analytics: non-branded traffic converts at 1.95% on a $64.73 order, scaled 1.17x for the online revenue GA4 does not capture against Shopify. Your Shopify cohort grid then gives the ladder: a customer's spend grows about 32% by the end of their first year and reaches $177 over six-plus years (your 2020 buyers, still spending), so the same visit is worth $1.95 in first-year value and $3.26 over a customer's life, and the lifetime rung is a floor that keeps rising. The three value columns below price each scenario's clicks at each rung, all of it non-branded. Your $24,000 of branded organic revenue sits on top of all of it, unchanged. GA4 + Shopify cohorts Modelled, directional

Read every column as value created per year at that pace, all of it non-branded. The mid case adds +$26,180 a year of new first-order revenue on top of today: about 346 extra customers a year, worth +$34,578 by the end of their first year and +$57,668 over their lives. The mid case asks you to grow today's clicks 1.77x in a niche where nothing you target scores above 4 for difficulty. And because the market is so October-loaded, roughly $16,400 of the mid case's first-order value lands in the October window alone (about $36,100 at the lifetime rung). The lifetime column is a floor from your own six-year cohorts, still climbing. Shopify cohorts Modelled, directional

Triangulated NZ search volume, September 2026. The full keyword, prompt and cluster research (129 keywords, 130 prompts, 11 territories) behind every number here lives in The data tab. GKP + SEMrush + DataForSEO

2. How AI-ready you are today: the AEO scorecard

We score every brand out of 100 against our AEO Maturity Model: four pillars (Technical, Content, Authority, Measurement), each out of 25. Certified is 80 out of 100, Professional on all four, the standard we build clients to. You are at 48 today, near the top of the Emerging band, and the shape of your 48 is unusual: three pillars sit at the top of their level and one sits on the floor.

In plain English, what holds each pillar under its gate: Technical reads well to engines but the homepage fails Google's real-user mobile speed test and the 696 collections run on defaulted titles with mostly missing descriptions. Content has a real FAQ and a 54-post archive, but no researched plan and none of the how-to layer engines draw answers from. Authority is the strongest specialist profile in the category, accumulated rather than managed, and AI citations show it: 2 of 11. Measurement is the floor: GA4 is installed but unread, and no Search Console property existed for bodyfx.co.nz until this month. The detailed findings, and the four moves that sum to exactly +32, are in the appendix.

3. The market you are missing: search and AI

Face and body paint looks like one search, and it is closer to a thousand. People search by product, by brand (Kryolan, Ben Nye), by design (butterfly, skeleton, tiger), by buyer (kids, sensitive skin, professionals) and by moment (Halloween, festivals, school galas), and they increasingly put the same questions to an AI chat. The demand is real, the difficulty is close to zero everywhere, and the AI answers for the category are being formed right now, almost entirely from offshore sources.

Search volume triangulated across Google Keyword Planner, SEMrush and DataForSEO (median of three), NZ, September 2026. Rank is your current NZ position from Ahrefs; there is no Search Console history for bodyfx.co.nz to read yet (the property was only created this month), so positions are tool-estimated until your own data accrues. "kids face paint" reports from one source only, so treat the 590 as directional; the intent is core either way. The full triangulation is in The data tab. GKP + SEMrush + DataForSEO Ahrefs

Zoom out and the funnel reads like this. The whole category is about 408,557 NZ searches a year. The 129-keyword basket we measured directly is 166,080 of those (41%); the rest is the long tail of character designs, colour variants and brand-line searches, sized from the full keyword universe on a consistent scale and stated as a floor. About one search in five survives as an organic click somebody wins: 81,711 a year, and you take 23,150 of those. On top of the searches sits the AI layer: live AI Overviews already sit on 9 of your basket's SERPs (face paint nz, fake blood, liquid latex, temporary tattoos, biodegradable glitter among them), and the buying and how-to prompt library we built runs to a directional 157,000 to 193,000 AI prompts a year, counted separately so nothing is double-counted. The full prompt library and territory maths live in the appendix and The data tab.

4. The opportunity, in dollars

The table in the executive summary is built from four numbers, all of them yours, and all of them non-branded. Here is the chain in full.

Roughly one search in five survives ads, image packs, AI answers and zero-click as an organic click somebody wins, a deliberately conservative rate in a category this visual. You already take more than one in four of the survivors. Modelled

Each of those clicks is worth about $1.48 to you in first-order revenue (the measured figure from the executive summary, and the one the table multiplies) and about $3.26 over a customer's life with you, on a six-plus-year floor that is still climbing.

That is the whole model: win more of the 81,711 clicks. The conversion rate is held flat at today's measured figure in every scenario, so any conversion improvement (and the URL-mapping work suggests there is one to be had) sits on top.

Three lines, one for each column of the value ladder, each pricing the acquisition pace reached at that month; hover any month for the maths. The phase-in runs roughly 10% of the mid uplift live by month 3, 35% by month 6, 70% by month 9 and all of it by month 12, then climbs toward the high case by month 24. One timing truth sits under the curve: because October is 27% of the year's searches, the season this build is aimed at is October 2027, met for the first time with the full answer layer live and aged. What can move for October 2026 is the refresh-and-fix list on pages that already rank, worth doing in the first weeks, not worth promising a season on. Modelled, directional Shopify cohorts

Two levers, both visible in your own data. Coverage: the pages that do not exist yet (the answer layer, the brand pages, the kids collection, the store page) start ranking in a market where maximum difficulty is 4 and 35% of your existing keywords already sit top 3. And conversion of what you already hold: striking-distance terms one or two places off the top 3 (bald cap, temporary tattoos nz, body glitter, white eyeliner) move up on refreshed titles, descriptions and internal links, and the junk rankings (an all-products URL ranking for a 1,500-a-month term) get real pages. The conservative case assumes mostly the second lever lands; the high case assumes both do.

Every figure here is non-branded organic. The AI prompt layer, your everyday-makeup crossover rankings, the physical store's foot traffic and the trade accounts are all deliberately outside the model, so each is upside on top of the table, not inside it. Your $24,000 of branded organic revenue is likewise stated separately and unchanged.

5. The six-month transformation

A done-for-you AI and Search transformation over six months, built and run on our side so there is nothing for your team to learn or maintain. It moves through three phases: foundations (the measurement layer stood up properly for the first time, the homepage speed fix, unique titles and descriptions across the collections, the bloat pruned, your baseline captured), owning your territories (the answer layer across kids safety, technique and design ideas, the brand and store pages, the buying guides, calibrated to your voice and drawing on Myrthe's existing tutorial catalogue), and authority and compounding (turning an accumulated backlink profile into a managed one, with the earned-media story only you can tell). We move into each phase as your site is ready for it rather than on a stopwatch, and everything we build is yours to keep. The full breakdown, deliverables and your time commitment live in The offer tab.

Traffic, orders, revenue and cohorts are from your own Google Analytics and Shopify (12 months to September 2026, NZD). Rankings are Ahrefs NZ estimates, September 2026, because no Search Console history exists for bodyfx.co.nz yet. Search volumes triangulated across Google Keyword Planner, SEMrush and DataForSEO (NZ), September 2026. Competitor authority from Ahrefs, one source across the whole table. AEO score from our AEO Maturity Model against a live crawl and Google's real-user speed data. Revenue figures are directional, modelled from your own measured conversion and order values, and the model gets re-run on your own Search Console data as it accrues.

Appendix A: your channels and current position

You have built a real business: $652,343 in total sales across 8,971 orders in the last year, up 1.6% on the year before, with $417,200 through the online store, $182,200 through the Morningside shop and $50,900 of trade draft orders. A 44.5% returning-customer rate on orders, driven by a pro and trade core that reorders steadily while the Halloween crowd buys once. This is growth on a proven base, and the base is unusually seasonal, which is exactly why the timing discipline in this plan matters.

GA4, NZ-filtered, 12 months to 7 September 2026, NZD. New Zealand is 97.7% of tracked revenue, so the model runs NZ-only. Paid channels together carry roughly $132,800, about 38% of NZ revenue. Organic search runs flat at 2,000 to 2,500 sessions a month with a 2.6x October spike and no growth trend across the year. Two small rows are left off the table for space: organic shopping ($2,707) and organic social ($853), each a blend of branded and non-branded traffic like every channel row here. GA4

Two and a half thousand indexable URLs producing 263 ranked keywords is the surface problem in one picture: 696 collections (against 1,825 products) full of expired campaign pages that rank for nothing, and no informational layer at all. When a page is built to answer a real query, it goes top 3 more than a third of the time. Ahrefs

Appendix B: the scorecard in detail

Here is what sits behind the 48, pillar by pillar, and how quick each piece is.

Appendix C: the demand detail

Working medians, 129 keywords, 13,840 searches a month, 166,080 a year. The whole-category figure of 408,557 a year is this basket times a 2.46 long-tail factor, derived on a single consistent scale from the full keyword universe (character designs, colour variants, brand lines the core basket never counts) and stated as a floor. Terms Body FX cannot credibly win (other cities' costume shops, car-paint noise, offshore intent) were removed from the universe rather than counted and hand-waved. GKP + SEMrush + DataForSEO Ahrefs universe

The basket's monthly search curve, from Keyword Planner's own monthly data: a baseline of 8,000 to 13,000 searches a month, then 41,560 in October. October is 27% of the year, 3.3 times an average month, and the spike has no ramp: September runs 30% above baseline, then October triples, then November is back to normal. The sharpest skews are the ideas terms (halloween face paint ideas 88% October, halloween makeup ideas 85%, witch face paint 83%), which is exactly the content layer that does not exist yet. A counter-seasonal layer runs underneath it: glitter and festival demand peaks in summer, the pro and trade base is flat year-round, and Easter, Matariki and Christmas give small secondary hooks.

We built a library of 130 real questions buyers put to AI engines about your category, anchored to the measured keyword volumes, then ran a sample through live answers in September 2026. Here is who gets named today.

Live answer checks, 9 September 2026, recorded per prompt in The data tab. The pattern across the library: cited on 2 of 11 sampled, and the two wins are the two purpose-built pages. 46 of the 130 prompts are October-loaded, and 106 of 130 are ones Body FX can uniquely own. Prompt volumes are directional floors anchored to the keyword medians with a conservative fan-out uplift; the library totals a directional 157,000 to 193,000 prompts a year. Directional

Appendix D: where to play, your content territories

The same demand, grouped into the eleven territories a plan executes against. Priority weighs demand, the October clock, whether AI answers are already live, and whether Body FX can win it uniquely.

Counts and volumes computed from the keyword and prompt libraries (129 keywords and 130 prompts, each assigned exactly once); October share is the volume-weighted share of each territory's searches that lands in October. Content requirement: roughly 76 pieces to own the priority cores, about 152 for full category coverage. The full territory maths is in The data tab. Cluster model

Appendix E: who you are up against, and the authority play

Ahrefs, September 2026, one source across the whole table so the comparison is like-for-like; NZ columns are country-scoped, domain rating and referring domains are global. One context line worth having: Look Sharp, the national party chain, runs at DR 30 with roughly 36,300 NZ organic visits a month, the ceiling a retail chain sets in this category; it is a chain with many stores, so it does not belong in a specialist table. And one note on your own profile: the raw backlink count (349,000) is inflated by a handful of sitewide links, so referring domains (471) is the number we quote. Ahrefs

The sequencing: the foundations and content phases get you cited on your own answers and convert the authority you already hold. The press and citation work runs alongside and then compounds for years after. It is the longest lever, and you start it from the front of the field, not the back.

Appendix F: where the numbers come from

Measurement runs in our own platform, built specifically for Shopify brands, so the dashboard, the data and the method are yours to keep, not rented from a third-party tool. The full research behind every table is in The data tab, and the complete money-model workings, every assumption and every equation, are published alongside this report as a plain-text file your own AI can verify: bodyfx-model-workings.md. Method