Paid search is greenfield with a genuinely attractive, cheap, high-intent demand pool — led by "refurbished". TYDI currently runs no Google Ads (0 paid keywords; a brief, tiny paid burst ran Dec 2025–May 2026 then stopped). Unlike the broken organic channel, paid is independent of the migration mess and can be switched on quickly once tracking + a clean product feed are in place. Recommended 6-month build: ~NZD $27,500, ramping $2k → $6.5k/mo, test → prove → scale. Breakeven ROAS ≈ 2.2× (46% GM); the keyword economics suggest a realistic 3–6× is achievable.
NZ search demand for refurbished consumer tech is large, low-difficulty and commercially intent-rich — and it's exactly what TYDI sells.
| Cluster | ~Searches/mo (NZ) | CPC (US$ est.) | Why it matters |
|---|---|---|---|
| Refurbished Apple/mobile (iPhone, iPad, MacBook, Watch, Samsung) | ~12,000+ | $0.20–0.45 | Highest-intent, lowest-difficulty, on-brand. The headline wedge. |
| Dyson (vacuum, v8, v11, v15, refurbished dyson) | ~13,000 | $0.25–0.90 | Lead product + a genuine supplier moat (Dyson exclusivity). |
| Category (vacuum cleaner nz, robot vacuum nz) | ~8,000 | $0.09–0.20 | Cheap clicks; best via Shopping/PMax feed. |
| Laptops (refurbished/cheap/gaming laptop nz) | ~2,400 | $0.20–0.30 | Solid secondary line. |
Standouts: refurbished iphone nz 3,200 (KD 0), dyson vacuum 6,500, dyson v8 3,300, vacuum cleaner nz 5,000, robot vacuum nz 3,000, refurbished phones 1,000, refurbished ipad 800, apple refurbished nz 700, refurbished macbook 500. Most carry difficulty 0–19 and US$0.20–0.50 CPCs — cheap, motivated traffic.
Naming follows the brain default [Brand]_[Network]_[Geo]_[Funnel]_[Theme].
| Campaign | Type | Purpose | Day-1 bidding |
|---|---|---|---|
Tydi_Search_NZ_Brand_Core | Search | Brand defence | Max clicks → Impr. share |
Tydi_PMax_NZ_RefurbApple | Perf. Max | Feed-led refurbished Apple | Max Conversion Value |
Tydi_PMax_NZ_Vacuums | Perf. Max | Feed-led Dyson + robot vacuums | Max Conversion Value |
Tydi_Search_NZ_BOFU_RefurbApple | Search | High-intent refurb Apple | Manual / Max Conv |
Tydi_Search_NZ_BOFU_Vacuums | Search | Dyson model + refurb | Manual / Max Conv |
Tydi_Search_NZ_BOFU_Laptops | Search | Refurb/cheap/gaming laptops | Manual / Max Conv |
Tydi_DemandGen_NZ_Retargeting (mth 5+) | Demand Gen | Re-engage visitors/cart | Max Conv |
Tydi Auto-Negatives: repair, parts, manual, fix, "battery replacement", review, vs, rental, hire, jobs, free, "how to". Refurb retail bleeds budget on repair/parts/info queries — block from day 1.For e-comm, Shopping/PMax is the workhorse — the feed captures the long tail of model+condition queries (e.g. "iphone 13 refurbished", "dyson v8 animal") that you'd never build out as keywords. Search campaigns cover the high-value head terms and the "refurbished [x] nz" intent.
dyson v8 needs the trademark / IP position confirmed first — and here it's not abstract: the trademark owner is also the supply-contract counterparty, so a keyword dispute can put the supply line at risk. Confirm the position before any brand-keyword spend goes live. See brand-strategy.html for the full trademark/reseller analysis and the NZ IP counsel sign-off it calls for.RefurbiPhone_Exact: [refurbished iphone nz], [refurbished iphone], [iphone refurbished nz], [refurbished iphone 13], [refurbished iphone 14], [refurbished iphone 15]RefurbMacBook_Exact: [refurbished macbook], [refurbished macbook nz], [macbook air refurbished], [refurbished mac]RefurbiPad_Exact: [refurbished ipad], [refurbished ipad nz], [ipad refurbished nz], [refurbished ipads]DysonV8_Exact / DysonV11_Exact: [dyson v8], [dyson v8 nz], [dyson v11], [refurbished dyson]RefurbLaptops_Exact: [refurbished laptops], [refurbished laptops nz], [refurbished laptop nz], [used laptops nz]AppleRefurb_Phrase: "apple refurbished nz", "refurbished apple watch", "refurbished samsung phones"RSAs per ad group via google-ads-creation (15 headlines / 4 descriptions; lead with Refurbished · NZ Warranty · Trade-in Welcome · NZ Owned). Sitelinks: Refurbished iPhones, Dyson Deals, Refurbished Laptops, Warranty & Returns.
| Month | Phase | Budget (NZD) | Focus |
|---|---|---|---|
| 1 | Test | $2,000 | Brand + PMax RefurbApple + BOFU RefurbApple. Tracking shakedown. |
| 2 | Test | $3,000 | Add PMax Vacuums + BOFU Vacuums. Mine search terms → negatives. |
| 3 | Prove | $4,500 | Scale winners; add BOFU Laptops; first tROAS test on PMax. |
| 4 | Prove | $5,000 | Graduate PMax/Search to Target ROAS; expand asset groups. |
| 5 | Scale | $6,500 | Add DemandGen retargeting; broaden converting themes. |
| 6 | Scale | $6,500 | Optimised scale; lock the efficient frontier. |
| Total | ≈ $27,500 | avg ≈ $4,600/mo (≈ 5–8% of ~$66k/mo revenue) | |
Plus a one-off setup (conversion tracking + Merchant Center feed + build): your time if DIY, or ~NZD $1,500–3,000 if outsourced.
Assumptions (to confirm in DD): AOV NZD $300, gross margin 46% (from dossier FY26A), breakeven ROAS ≈ 2.2×.
| Scenario | Blended CPC | Conv. rate | ROAS | Read |
|---|---|---|---|---|
| Conservative | $1.50 | 1.5% | 2.8× | Just above breakeven — fine while learning. |
| Target | $1.10 | 2.0% | 5.5× | Realistic once feed + negatives mature. |
| Stretch | $0.85 | 2.5% | 9.4× | If feed quality + CR are strong. |
ROAS = (Conv. rate × AOV) ÷ CPC. At the Target scenario, $5,000/mo → ~90 orders → ~$27k revenue → ~$12.5k gross profit − $5k ad spend = ~$7.5k/mo contribution, CPA ≈ $55 vs gross profit/order ≈ $138. The model only fails if real CPCs are 2–3× Ahrefs and CR is <1.5% and AOV is low — hence the DD validation.
condition = refurbished, GTIN/MPN, price, availability, images. The single biggest lever for refurb e-comm.google-ads-new-account)Recommendation: budget ~NZD $27,500 over 6 months ($2k→$6.5k ramp), refurbished-Apple-led, PMax + tight high-intent Search, on a clean feed with verified tracking. Self-funding once ROAS clears ~2.2× (likely month 2–3). Treat the spend as growth capital that should pay for itself, not a sunk marketing cost — and it's a channel a generic buyer couldn't stand up as fast.
Confirm in due diligence (add to broker list):