↓ Download PDF
Team Empathy  ·  Search & AI Strategy

Own the answer,
not just your name.

A Search and AI strategy for Neat

Getting Neat found on ChatGPT

New Zealand   |   Prepared for Abby  ·  July 2026

1. Executive summary

The short version. You have built a brand 100,000+ Kiwis love: roughly $3.9M a year online, grown almost entirely on paid ads. The quiet problem is that when someone opens Google or ChatGPT and asks "best natural perfume", "which magnesium for sleep" or "best collagen powder NZ", they find Chemist Warehouse, the health publishers and your competitors, not Neat. Almost all of your search traffic today is people already typing "neat". That non-branded demand is large and high-intent, it converts about five times better than your paid social, and once you own it, it keeps sending buyers after the ad spend stops. This is not about out-spending Meta: organic and AI search compound instead of resetting to zero, and you own them rather than rent them.
Three things to know
~$530K
new revenue a year on the table at a 20% share of the searches you are missing
~48,400
non-branded category searches a month in NZ; you capture ~2%
~$389K
already earned from organic in 12 months, converting at 10.6–11.5%
42 / 100
AEO readiness today, Certified is 80
Where you focus, at a glance
Fragrance
~29,000 searches / mo
The prize
"perfume nz" alone is 27,100 a month at difficulty 5, and your 63-scent dupe range is built to answer the designer long tail behind it.
Wellness
~9,900 searches / mo
Fastest growing
Collagen and magnesium demand, most of it with live AI answers. Magnesium glycinate is up 86% year on year.
The answer layer
38 of 56 prompts live
The moat
Skincare, collagen and perfume questions already trigger an AI answer, and no NZ brand owns them. First mover wins.

Triangulated NZ search volume, July 2026. The full keyword, prompt and cluster research behind every number here lives in The data tab. GKP + Ahrefs + SEMrush

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 42 today, with a very fixable shape: your plumbing and authority are ahead of most NZ brands, and the gap is the answer-layer content and measurement.

Neat42/100
Developing · well-plumbed, answer-poor
Technical14/25
Content9/25
Authority14/25
Measurement5/25
+38
points to Certified. Your agentic plumbing (your site already welcomes AI crawlers and runs Shopify's agentic-discovery feed) and your backlink authority are ahead of the pack. The climb is content shape, schema and measurement, the fast-moving pillars.
Where you are  ·  where you need to be
Neat
42 / 100
Today · Developing, 42 / 100 Where you need to be · Certified, 80 / 100

The detailed findings, how the scoring works, and where every number comes from are in the appendix.

3. Your search & AI market, laid out

This is the full picture of the demand sitting around your brand, built the way we build it for every client: keyword volumes triangulated across Google Keyword Planner, Ahrefs and SEMrush (the median of three, so no single tool's blind spot skews the model), a 56-prompt AI-search library mined from real questions people ask, and a cluster map that organises all of it into the territories a content plan can actually execute. The full workbooks are yours in The data tab; here are the headlines.

The keywords: what New Zealand searches every month
KeywordKeyword PlannerAhrefsSEMrushWorking volumeDifficulty
perfume nz27,1009,60027,10027,1005
collagen powder nz3,6002,3003,6003,6001
magnesium glycinate nz3,6009002,9002,9000
dry skin moisturiser1,9001001,9001,9000
natural deodorant nz1,0006001,0001,00023
vitamin c serum nz1,0004501,0001,0000
essential oils nz1,3009001,0001,00026
mens cologne nz7205007207200
collagen nz7203507207200
marine collagen nz4804004804800

Top 10 of 47 terms researched; 43 of 47 reached Tier 1 confidence (all three tools agree). The basket totals ~48,400 non-branded searches a month, and it is a deliberate floor: the designer-name long tail behind your 63-scent dupe range (Versace, YSL, Billie Eilish and hundreds more) adds tens of thousands of addressable searches on top. Most terms sit under difficulty 15, which is unusually winnable. GKP + Ahrefs + SEMrush

The AI prompts: what people ask the engines

AI engines do not invent demand, they intercept it. We built a 56-prompt library from real sources (question searches, review language, the roundups AI cites), each anchored to the triangulated volumes. 38 of the 56 already trigger a live AI answer in New Zealand, and Neat is cited in none of them.

Real question buyers askEst. demand / moAI answer live?
What is the best collagen powder?~7,700Yes
What is the best vitamin C serum?~5,600Yes
Is hyaluronic acid or niacinamide better?~4,300Yes
What is a cheaper alternative to designer perfume?~2,800Yes
What does face serum actually do?~1,800Yes
What is magnesium glycinate good for?~1,700Yes
Is apple cider vinegar good for you?~810Yes
Are perfume dupes as good as the original?~420Yes

A sample of the 56-prompt library, all of it in The data tab with per-prompt demand and the live-AI flag. Volumes are keyword-anchored with a conservative AI fan-out uplift, directional +/-50%. Modelled

The cluster map: ten territories, five you can dominate first
TerritoryDemand / moPromptsThe play
Fragrance dupes ★~465 + ~78k designer tail8A "smells like" asset per hero scent; 63 products ready to answer
Natural perfume & oils ★~28,5009The category head term; oils vs sprays education
Collagen ★~5,8006Marine vs bovine is the wedge; the NZ marine story
Skincare routine & serums~5,0009Calm-first routine and serum guides
Magnesium & sleep~3,4006Fastest-growing cluster, +86% year on year
Problem skin~2,9006Barrier support is literally your brand promise
Lifestyle & rituals~1,5003Ritual content that lifts order value
Natural deodorant ★~1,0003Objection content wins the AI answer here
Nutrition~6404Refresh the existing 2025 posts
Brand trust ★branded + roundups2Get cited in the "best NZ brands" lists that omit you today

★ = brand-strength territory, where Neat can genuinely win first. Answering the whole map takes roughly 160 content assets; the five starred territories alone take ~100. Full cluster maths in The data tab. Cluster model

4. The opportunity, in dollars

Put the three layers together and price them on your own numbers. The basket gives a click pool of roughly 28,000 organic clicks a month. We deliberately cut your proven 10.6–11.5% conversion to a cautious 4% for colder, non-branded visitors, and value each new customer at what yours actually spend: ~$300 a year on skincare and wellness, ~$130 a year on fragrance Shopify. At a 20% share of each territory, here is the money on the table, on top of the ~$389k the branded channel already earns.

AreaDemand / moNew customers / yrYear-one revenue
Fragrance (dupes + perfume & oils)~29,000~1,610~$210,000
Wellness (collagen, magnesium, nutrition)~9,900~550~$165,000
Skincare (routine + problem skin)~7,900~440~$130,000
Lifestyle & rituals~1,500~85~$17,000
Total~48,400~2,690~$530,000

Scale it to taste: a 10% share is roughly half these numbers, a 30% category-leader share roughly 1.5x (~$800,000). None of it counts the designer long tail. Modelled, plus or minus 40%

And we are probably underselling it. Every figure above assumes cold organic visitors convert at just 4%. Your store's organic conversion actually runs at 10.6–11.5%, so with your brand equity the realistic landing zone is 4–8%, possibly more. If it lands at 8%, every number in that table doubles. We would rather promise the floor and deliver the ceiling.
The compounding curve (a 20% share, arriving over twelve months)
$0k$100k$200k$300k$400k$500kYou own the systemTodayMonth 3Month 6Month 9Month 12$530K run-rate at a 20% share

One curve: a 20% share arriving over the year, and the percentages are shares of the new ~$492K, not growth rates. You are certified by month 3 with ~10% arrived, the system is handed over at month 6 with ~35% arrived, ~70% by month 9, and all of it by month 12, where the curve ends at ~$530K run-rate. The back-loaded ramp is deliberate: content ships from the first months, and rankings and AI citations follow with a lag. Hover any point to see the maths. Modelled, directional

Every dollar excludes repeat. The figures above are first-year customer value. In replenishment categories like skincare, fragrance and wellness that is only the start: two of every three of your organic orders already come from returning customers, so this same cohort is worth roughly $1.6M over three years, and a fresh cohort lands on top every year after.

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 schema layer, heading structure, AI-crawler access, your measurement baseline), owning your territories (the fragrance-dupe, collagen, magnesium and skincare hubs, roughly 100 to 160 pieces across the six months, roughly half refreshing pages you already have and half net-new, all written in your own warm, women-owned, made-in-NZ voice), and authority and compounding (the roundups, reviews and community presence AI trusts). We move into each phase as your site is ready for it rather than on a stopwatch. Everything is yours to keep: the foundations, the content, the links and your AI-Ready certification.

The full breakdown, the four pillars, your time commitment and how the six months run live in The offer tab.

Traffic, orders, revenue and conversion are from Neat's own Shopify and Search Console figures shared in our working session. Search volumes triangulated across Google Keyword Planner, Ahrefs and SEMrush (NZ), July 2026. Competitor footprints and Domain Ratings from Ahrefs Site Explorer, estimated. AEO score from our AEO Maturity Model against a live crawl. Revenue figures are directional, modelled from your own conversion and order value, and are not a guarantee of results.

Appendix A: your channels and current position

Neat is a women-owned, made-in-NZ natural beauty and wellness brand: skincare, fragrance (originals and a 63-scent dupe range), collagen and magnesium wellness, and lifestyle rituals, bought by 100,000+ Kiwis. Today it is a ~$3.9M online business grown almost entirely on paid media. The foundations are good. This is growth on a proven engine, not a rescue.

Your channels, last 12 months
ChannelSales / yrShareRead
Google Ads$1.39M~36%Strong, but rented; stops when spend stops
Meta (FB + IG paid)$833K~21%Converts at 1.5–2.5%, needs constant feeding
Email (Klaviyo)$806K~21%Your owned channel, working well
Organic search (all engines)$389K~10%Highest-quality traffic, smallest paid attention
TikTok Ads$44K0.35 ROAS$126K spent to make $44K back over the year

Your Shopify channel report, last 12 months, NZD. Organic converts at 10.6–11.5%, on par with your best channels and about five times your paid social. Bing organic alone did ~$52K last year at a 25% conversion rate. Shopify

Your search traffic is your name, not your market
Your top queries (last 3 months)ClicksBranded?
neat nz1,184Yes
neat591Yes
neat perfume324Yes
neat skincare223Yes
neat perfume nz · neat new plymouth · neat fragrance…~540All yes

Every top query already knows you. The non-branded category, where 48,400 monthly searches sit, is almost entirely uncaptured. GSC

Appendix B: the scorecard in detail

Here is what sits behind the 42, and who fixes each piece. It is content shape, schema and measurement, not a rebuild.

FindingStatusFix
robots.txt welcomes AI crawlersAhead of most peersDone
Shopify agentic-discovery feed liveStrongDone
Backlink authority (DR 21, 436 ref domains)Real asset for the categoryPoint it at content
Organization / Product / FAQ schemaThin on the themeCore technical fix
Non-branded ranking content151 keywords, all brandedThe core content work
Blog freshness (22 posts, early 2025)Stale, unanswerable shapeRefresh + cluster + schema
AI-visibility measurementNone yetBaseline + monthly panel
The phases flex, the destination does not. Your plumbing and authority bones are strong, so foundations move fast: the schema and heading fixes land in weeks, and the content engine starts as soon as everything is calibrated to your voice and range. What stays fixed is the destination: Certified 80/100 and compounding, inside the same six-month transformation.

Appendix C: who you are up against

The sites winning your searches are big retailers, not brands.

Who ranks in your spaceNZ organic / moDRRead
Chemist Warehouse587,90266Retail giant, wins on range not story
Pure Nature41,22243The local natural-category winner
HealthPost26,33957Natural health retailer
PerfumeNZ24,84341Fragrance retailer
Neat (you)~1,96721Almost entirely branded traffic

Ahrefs, New Zealand, July 2026, estimated. Ahrefs Site Explorer

No brand owns this answer space. Retailers rank on range; none of them can tell your story: women owned, made in NZ, 100,000+ local customers, IFRA-compliant dupes with 600+ five-star reviews. That story is exactly what AI engines reward when they choose who to recommend, and it is the lane sitting wide open.

Appendix D: where the numbers come from

SourceWhat it provides
Keyword volumes, triangulatedGoogle Keyword Planner, SEMrush and Ahrefs, taken as the median of three, plus keyword difficulty. Single tools undercount NZ badly (Ahrefs alone put this basket at ~20,600/mo; triangulated it is ~48,400), so we never rely on one source.
Head-term anchorsEvery AI prompt is anchored to the real, measurable keyword volumes underneath it, so the prompt demand is a floor, not a guess.
Query fan-out upliftOne AI prompt triggers many hidden sub-searches, roughly 95% of which carry zero traditional keyword volume. We apply a conservative 1.3 to 1.6x uplift rather than guessing top-down.
AI answer verificationWhich questions already trigger a Google AI answer, observed live in the results (they fire across your collagen, magnesium, skincare and perfume questions), not predicted.
Your own ground truthYour Shopify channel figures and Search Console queries shared in the working session. Every modelled figure reconciles back to these.

The full research behind every table here is in The data tab. Method