↓ Download PDF
Team Empathy  ·  Search & AI Strategy

Be the mattress brand
AI recommends in New Zealand.

A Search and AI strategy for The Flatpack Company

Getting The Flatpack found on ChatGPT

New Zealand   |   Prepared for Stephen  ·  July 2026

1. Executive summary

The short version. You have built a real organic engine already: around $1.2M a year across the organic engines in your own Shopify attribution, converting better than any paid social channel you run. Bed frames you win, because your brand name is the category. But you have moved to 70% mattresses, and mattresses is exactly where you go invisible the moment a Kiwi searches or asks AI. You already rank across the category, on 885,000 search impressions a quarter, and capture barely 1% of the clicks because you sit at position 8 to 20, not the top three. Closing that gap, and owning the mattress conversation, is the opportunity.
Three things to know
$1.2M
organic revenue a year, your best-converting acquisition channel
$425K–$735K
new revenue a year on the table (organic + AI)
34,800
non-branded searches a month in your focus categories
$21,325
already earned from ChatGPT in 12 months, unoptimised
50 / 100
AEO readiness today, Certified is 80
Where you focus, at a glance
Mattresses
~13,500 searches / mo
The prize
70% of your sales now, and invisible in AI. Ecosa owns the answer. This is the win.
Bed frames
~15,500 searches / mo
You win
Your beachhead. Cited first in AI because your name is the category. Defend and extend.
Toppers & headboards
~6,500 searches / mo
Expanding
The adjacent lanes you have just moved into. Easy, on-brand wins to fold in.

Triangulated NZ search volume per category, July 2026. We have left bedroom furniture out of the model: at roughly 2% of sales it is not where you are pointed. The full keyword, prompt and cluster research behind every number here lives in The data tab. GKP + Ahrefs + SEMrush

2. Where The Flatpack sits today

You started in 2019 as Otago students who could not find a decent, affordable bed that shipped flat and went together fast. Six years and 15,000-plus beds later you are a genuine NZ brand, doubling revenue two years running toward roughly $9M, with a 4.8 out of 5 rating across 569 reviews on your hero frame alone. The foundations are good. This is growth on a proven base, not a rescue.

Your channels, from your own Shopify attribution
ChannelSales / yrOrdersConversionOrder valueRead
Google (paid)$3,231,2772,9621.63%$1,091Your biggest engine, but rented; stops when you stop paying
Klaviyo (email)$1,143,3561,0093.44%$1,133Best converter, but retention, not new customers
Google Search (organic)$937,6928121.77%$1,155Your best acquisition channel after paid, and under-fed
Facebook (paid)$904,6278040.18%$1,125Huge sessions, weakest conversion
Direct$338,8073520.31%$963Your brand strength showing up
Instagram (paid)$238,9342110.12%$1,132High spend, low conversion
Bing (organic)$123,857985.33%$1,264Small but the most efficient channel in the store
ChatGPT$21,325201.7–3.1%~$1,066A brand-new channel, zero optimisation yet

Your Shopify attribution, last 12 months, last non-direct-click model, NZD. The organic engines combined (Google, Bing, DuckDuckGo, Yahoo) earn roughly $1.15M a year, second only to Google paid, on an order value steady around $1,100 across channels. Shopify attribution

One data note, so the numbers are honest. Your separate student business shares the same domain and analytics while converting off Shopify, which drags every blended conversion figure down (site-wide reads ~0.7% once corrected, and GA4 puts organic at 2.15% for the last 90 days versus Shopify's 1.77% over 12 months). We model conservatively inside that range, and the numbers sharpen further once we filter the student traffic out together.
You appear everywhere. You win almost nowhere.
Where you appearnon-branded impressions
96%
What you winnon-branded clicks
48%

Non-branded category search is 96% of your impressions but only 48% of your clicks. You own your name and rank across the category, but sitting at position 8 to 20 converts almost none of that visibility into visits. GSC

3. 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 50 today, with strong bones and a clear climb.

The Flatpack50/100
Developing · strong bones, no answer engine
Technical15/25
Content10/25
Authority15/25
Measurement10/25
+30
points to Certified. You are already halfway, carried by strong Technical and Authority. The climb is Content and Measurement.
Where you are  ·  where you need to be
The Flatpack
50 / 100
Today · Developing, 50 / 100 Where you need to be · Certified, 80 / 100

Here is what sits behind the score, and who fixes each piece. It is content, authority and a few schema fixes, not a rebuild.

FindingStatusFix
AI crawlers welcomed (GPTBot, ClaudeBot, Perplexity)Strong, all allowedDone
Agentic commerce (UCP) liveAhead of most peersDone
Product schema + reviews (4.8 / 569)Valid and strongAdd GTINs
Organization / WebSite schemaMissingHours
Mobile Core Web VitalsFailing (Lighthouse 54)Weeks
Content engine (blog, guides, FAQ)You have none yetThe main build
AI-visibility measurementNone yetBaseline + monthly panel

The single highest-leverage build is content: you have no blog and no buying guides, which is why you are invisible in mattresses. Building that engine is what turns the citation gaps below into revenue.

The phases flex, the destination does not. This scorecard is exactly what sets the phase mix. Your technical and authority bones are solid, so foundations move fast: the schema fixes land in weeks, and the content engine starts as soon as the kick-off deep dive calibrates it to your voice and positioning, mid-market affordable, not cheap. What stays fixed is the destination: Certified 80/100 and compounding, inside the same six-month transformation.
Measured the whole way. From day one we set up your AI visibility dashboard. We confirm your top keywords and priority prompts with you, track the keywords through Search Console and Google Analytics, track the AI prompts through our software, and you get live reporting on all of it. Your weakest pillar becomes a strength, and every month is measured against a baseline taken before we touch anything, on your numbers, not ours.
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, so we never rely on one.
Head-term anchorsEvery AI prompt is anchored to the real, measurable keyword volumes of the search terms that sit underneath it. The anchor is the floor of what the prompt is worth, so nothing in the model rests on a guess.
Query fan-out upliftOne AI prompt triggers many hidden sub-searches inside the engine, and roughly 95% of those carry zero traditional keyword volume. We apply a conservative 1.3 to 1.6x uplift to the anchor rather than guessing top-down, which is why our estimates are floors, not inflation.
AI answer verificationWhich questions already trigger a Google AI Overview or AI answer, verified in the live NZ results, and which brands get named. Every "AI live" flag in this document and the workbook is observed, not predicted.
Citation trackingA fixed panel of your priority category questions, run monthly across ChatGPT, Claude, Perplexity and Google AI Overviews, logging exactly which brands get named. The panel is confirmed with you at kick-off and the baseline is taken in week one, so every month is measured against day zero.
Your own ground truthSearch Console, GA4 and your Shopify attribution, including the ChatGPT channel already visible in your store. Every modelled figure reconciles back to these actuals.

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 here is in The data tab. Method

4. The market you are missing: search and AI

Mattresses is a high-research, long-consideration purchase. Buyers do not know what makes a good mattress, the market is deliberately confusing, and they ask the same questions in two places now, a search box and an AI chat. The demand is large, the difficulty is mostly low, and the AI layer is already answering.

4a. Validated demand, and where you rank today
TermSearches / mo (NZ)Your rank nowAI Overview live?
mattress topper nz3,600Page 2+-
bed frames nz2,9007-
bed base nz2,90011-
headboards nz2,9009-
mattress nz1,9003-
queen mattress nz1,600Page 2+-
mattress in a box nz3205-
best mattress nz590Page 2+Yes

Search volume triangulated across Google Keyword Planner, Ahrefs and SEMrush (median of three), NZ. Your rank is your current average Google position from Search Console. The pattern: decent on the generic terms, page two or worse on the growth terms (toppers, queen, "best mattress"), and even your strong spots leave clicks on the table (position 3 on "mattress nz" at under 3% click-through, with shopping and AI answers taking the rest). The full 53-term triangulation is in The data tab. GKP + Ahrefs + SEMrush GSC

4b. From category demand to dollars
Full categorysearches / mo, NZ
~83,000
Winnable poolorganic clicks / mo
~16,500
You todayyour share of the pool
~12%

The validated core (34,800) grows to a full category of about 83,000 searches a month once you add the long tail, which is unusually deep for mattresses given how much people research. Strip out ads, shopping carousels, AI Overviews and pure-research queries, and roughly one search in five survives as a winnable organic click, a pool of about 16,500 a month. AI prompts (another ~7,000 to 14,000 a month) are counted separately so nothing is double-counted. Modelled

4c. The AI questions, and who owns the answer

We built a library of 130 real questions Kiwis ask AI engines about your category, from Google's "People Also Ask", Reddit threads with hundreds of comments, review language and the head-term intent map. Here is a sample, with roughly how often each comes up a month.

Real questionClusterEst. monthly volume
What is the best mattress in a box in NZ?Mattress choice~900
What is the best affordable mattress in NZ?Mattress choice~700
Is a mattress in a box actually any good?MIB education~600
What is the best mattress for back pain in NZ?Firmness~450
What is the difference between a double and a queen bed?Sizes & fit~450
Are mattress toppers worth it, or should I buy a new mattress?Toppers~400
Do I really need a bed base, or will slats do?Bed bases~300
How do I furnish my first flat on a student budget?Student / first flat~350

A sample of the 130-question library, all of it in The data tab with per-question volume, priority and ownability. Volumes are directional floors (head-term anchor plus fan-out uplift). 122 of the 130 are questions you can uniquely or credibly own. Directional

And here is who the AI answers name today, cluster by cluster, versus where you show up.

Prompt clusterCited todayFlatpack?
Flat-pack bed framesThe Flatpack, Mocka, TargetWins
Best mattress / mattress-in-a-box NZEcosa, TSB Living, nzbeds, MoneyHubAbsent
Mattress buying (firmness, "do I need a base")Ecosa, Consumer, review sitesAbsent
Comparison & "best value" round-upsReview sites, Reddit, listiclesAbsent
Student / first-flat bedsWise Move, The Warehouse, StuffAbsent
The first-mover window. Almost no NZ mattress brand deliberately publishes the answers AI engines cite, and there are only a handful of mattress-in-a-box players here. You are cited where you have content and invisible where you do not. That is not a mystery, it is a to-do list, and the brand that answers these questions first becomes the default before the rest wake up.

5. Where to play: your content territories

The same demand, grouped into the territories a plan executes against, each scored on how much of it you can realistically own. Priority weighs demand, whether AI is live, and whether you can win it uniquely.

TerritoryCoverage todayDemandAIPriority /5What we build
Mattress choice & best-ofProducts only, no guidesHighYes5The comparison and best-of hub: best mattress in a box NZ, the NZ-owned angle, price-band guides, honest Ecosa comparisons. The single biggest win.
Mattress-in-a-box educationNoneMed + AIYes5The doubt-and-how-to set (expansion, downsides, durability, returns). Un-owned PAA territory; the fastest citation wins in the category.
Firmness & comfortNoneMed + AIYes5The firmness guide, sleep-position and back-pain content. Fixes the confusing-market problem in your honest voice.
Sizes & fitNoneMed + AIYes4The definitive NZ bed-size chart, room-fit and couple-sizing guides. Citable tables AI loves.
Bed frames & flat-packStrong (the beachhead)HighYes4Defend and extend: strength and weight ratings, the no-creak story, flat-pack vs traditional, storage beds.
Bed bases & slatsProducts onlyMed + AIYes4The do-I-need-a-base answer set and compatibility guides; routes the Reddit debate to your range.
ToppersNew category, no contentHighYes4Topper-vs-replace decision content on the biggest head term (3,600/mo); launch content for the new range.
Student & first flatStudent page + sub-brandMed + AIYes4The first-flat furnishing hub: checklists, budgets, moving guides. Your origin story and perfect ICP, currently answered by removal companies.
HeadboardsNew category, no contentMedYes3Attach and compatibility how-tos plus the renter angle for the new range.
Care, trial & lifespanSparse (assembly only)Med + AIYes3Care, warranty-in-plain-English, disposal and trial mechanics. Trust content that closes the online-buying doubt.

Priority is a 1 to 5 score, where 5 means lead with it now; it is not a count of pieces. Content requirement: roughly 50 to 70 pieces to own the core (mattress choice, mattress-in-a-box education, firmness and sizes), and 100 to 150 over the six months for full category coverage. Unlike most brands we audit, this is almost all net-new: your blog is empty, so every guide is a new asset, tightly interlinked with your collection and product pages. The full cluster maths is in The data tab. Cluster model

Lead with three. Mattress choice and best-of, the category-defining answers Ecosa holds on beatable content. Mattress-in-a-box education, the un-owned doubt-and-how-to questions with the fastest citation wins. Firmness and sizes, the education pair that fixes the confusing market in your voice. Bed frames stay defended behind them, and toppers launch with the new range.

6. The opportunity, in dollars

Every figure here is anchored to your own numbers: your organic conversion (2.15% in GA4 over the last 90 days, 1.77% on your Shopify attribution over 12 months, we model on the GA4 figure with the student-traffic caveat noted above), your roughly $1,100 order value, and a baseline of about $570,000 a year of non-branded organic revenue. Share and market size are directional at plus or minus 40%. The engine has two parts that compound together: organic taking a bigger share of the winnable pool, and AI search, where ChatGPT already sends you orders.

The compounding curve (the mid case, arriving over twelve months)
$400k$600k$800k$1M$1.2M$1.4MYou own the systemTodayMonth 3Month 6Month 9Month 12$1.31M run-rate, the mid case

One curve, one case: this is the mid case arriving. Roughly ~10% of the new revenue is live by month 3 (foundations), ~35% by month 6 (certified, content shipping), ~70% by month 9, and all of it by month 12, where the curve ends at $1.31M, exactly the mid-case run-rate in the table below (your ~$570K baseline plus $735K new). It keeps compounding past the engagement, and it does not switch off. Hover any point on the live version for the phase and figure. Modelled, directional

New revenue over today's baseline (per year)
CaseOrganic + AI new / yrWith repeat (×1.15)New / moRun-rate (baseline + new)
Conservative+$425,000~$490,000~$35,000~$995,000
Mid+$735,000~$845,000~$61,000~$1.31M
High+$1.01M~$1.16M~$84,000~$1.58M

Roughly two-thirds of each case is organic search (the climb in the curve above) and one-third is AI search, modelled from the AI traffic already converting in your store (ChatGPT alone: ~$21K and 20 orders in the last 12 months on your Shopify attribution) as you break into the mattress answers you are absent from. The repeat multiplier is a modest 1.15x, matched to your ~8% repeat rate and growing referrals. Run-rate is simply your ~$570K non-branded organic baseline plus the new revenue. Modelled, plus or minus 40%

$61,000
mid-case new revenue per month, organic and AI combined
$21,325
already earned from ChatGPT in 12 months, before any optimisation
~83,000
category searches a month in your focus categories

7. Who you are up against, and the authority play

BrandOrganic visits / moKeywordsTop-3DRRead
Mocka~136,0006,6482,10847Broad furniture leader, the long-horizon ceiling
Ecosa~42,8002,67876730Owns mattresses, the nearer target
Sleepyhead~5,60089941032Legacy brand, mostly branded traffic
The Flatpack~8,50047916821Strong on bed frames, thin on mattresses

Ahrefs, NZ, July 2026, estimated organic traffic. Ecosa is the brand to catch in mattresses, on beatable content. Ahrefs Site Explorer

Authority is where the mattress game is actually won. Roughly two-thirds of what AI engines cite comes from third-party sites, not your own. For a high-research buy like a mattress, that means Reddit threads, "best mattress NZ" round-ups, Consumer and MoneyHub reviews, and comparison pages. Ecosa and the review sites hold those today; you hold none of them. That is the highest-leverage, longest-compounding lever you have, and it is exactly the local, honest, New-Zealand-owned story that plays well against a big Australian brand. There is not even an NZ mattress-buying community on Reddit yet. That is a lane to own, not just enter.

Our honest sequencing: the foundations and content phases get you cited on your own answers and lift you from page two to the top three. Authority, the digital PR, review-site placements and Reddit presence, is the engine that runs alongside and then compounds for years after. It is a core focus we are building deep, done-for-you capability in specifically for Shopify brands, and it is where most of the long-term share-of-voice sits.

8. 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 (Core Web Vitals, schema and GTINs, the student-traffic measurement fix, your baseline), owning your territories (the mattress hubs, the education pair and around 100 to 150 pieces across the six months, with content production starting as soon as everything is calibrated to your voice and positioning, mid-market affordable, not cheap), and authority and compounding (the review-site placements, round-ups and Reddit presence AI trusts). We move into each phase as your site is ready for it rather than on a stopwatch. Behind it, we are building the platform you will own, exclusively for Shopify brands: an early version is only weeks away, and by the end of your six months it is a comprehensive system your team can log into and run. Everything is yours to keep: the foundations, the content, the links and your AI-Ready certification.

The full breakdown, deliverables, your time commitment and pricing live in The offer tab.

Traffic, orders, revenue and rankings are from The Flatpack's own Google Analytics, Search Console and Shopify (attribution last 12 months, last non-direct-click model; GA4 last 90 days), online store only, excluding the separate student business where noted. 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. Pricing is set out separately.