# Koa — fact-check & pitch guide Built 2026-08-09. Everything below is either **verified against a primary source** or **explicitly marked unverified**. Nothing here is asserted from memory. --- ## 1. The one thing that loses you the room **The deck never states an SDG target by number.** "SDG" appears three times and every one is *the question* — *"Which UN SDG targets does it hit, by number?"* — never the answer. At an SDG hackathon that is the single question you are guaranteed to be asked. ### The answer, verified against sdgs.un.org | Target | Official wording (verbatim) | Fit | |---|---|---| | **3.4** | "…reduce by one third premature mortality from non-communicable diseases through prevention and treatment and **promote mental health and well-being**" | **STRONG** | | **10.2** | "…empower and promote the social, economic and political inclusion of all, **irrespective of age**, sex, disability, race, ethnicity, origin, religion or economic or other status" | **STRONG — lead with this** | | **11.7** | "…provide universal access to safe, inclusive and accessible, green and public spaces, in particular for women and children, **older persons** and persons with disabilities" | MODERATE | | **3.8** | Universal health coverage… access to quality essential health-care services | MODERATE | | **16.7** | "Ensure responsive, inclusive, participatory and representative decision-making at all levels" — indicator **16.7.2** is disaggregated **by age** | STRETCH — but your best SDG-16 hook | | 16.6 | "…effective, accountable and transparent institutions" | **WEAK — do not lead with it.** Koa is not an institution | **Do NOT claim 3.d.** Its "early warning" means national epidemic capacity, not individual decline detection. A judge who knows the framework will catch it. ### Say this when asked > "Our primary targets are **3.4 and 10.2** — and 10.2 names it directly: the social, > economic and political inclusion of all, *irrespective of age*. On SDG 16 we'll be > straight with you: our fit is at the **indicator** level, not the institutional one. > Indicator **16.7.2** measures the proportion of people who believe decision-making is > inclusive and responsive, **disaggregated by age** — and someone who hasn't left their > house in eleven days has dropped out of that measurement entirely. Koa keeps older > New Zealanders present enough to be counted and heard. That's the precondition SDG 16 > is built on." Owning the imperfect fit reads as rigour. It also disarms the judge who was going to ask. --- ## 2. Global framing you are not using — verified **WHO Commission on Social Connection** (est. Nov 2023; co-chairs Chido Mpemba and Dr Vivek Murthy). Flagship report *"From loneliness to social connection: charting a path to healthier societies"*, **30 June 2025**: - **1 in 6 people worldwide** is affected by loneliness - **~871,000 deaths a year** are linked to it — about **100 every hour** - Social isolation affects **up to 1 in 3 older adults** ← your stat - Linked to stroke, heart disease, diabetes, cognitive decline, premature death Also: the **78th World Health Assembly (May 2025)** adopted the **first-ever WHA resolution on social connection**. Cite it by title and "WHA78, May 2025" — **do not quote a resolution number**, it could not be verified from a primary source. This turns "a nice idea for lonely nanas" into "a named global health priority". Use it. --- ## 3. Numbers in the deck — status | Claim | Status | |---|---| | Loneliness → **+29% heart disease**, **+50% dementia**, **+26% premature death** | **VERIFIED** — CDC, *Health Effects of Social Isolation and Loneliness* (2024-05-15) | | Loneliness → **+33% depression** | **UNVERIFIABLE — no primary source exists.** Replaced with CDC's verified **+32% stroke**. If you want depression, the US Surgeon General (2023) says lonely people are **more than twice as likely** to develop it. | | **"1 of only 7 things we said we'd track"** | **WRONG AS WORDED.** The strategy has **five key areas**, not seven outcomes, and the current framework has **60 measures**. The "seven" is real but different — the Action Plan says "a set of **seven indicators** to understand [COVID-19's] impact on older people… income adequacy, unemployment, benefit receipt, **loneliness**, elder abuse, discrimination, and housing." They were a **COVID-impact set**, not the strategy's permanent tracked outcomes. Say "one of seven indicators the Government used to track the impact on older people" — not "one of only 7 things we said we'd track" | | Strategy name & dates: **Better Later Life – He Oranga Kaumātua 2019–2034** | **VERIFIED** | | Vision: *"Kia noho ora tonu ngā kaumātua / Older New Zealanders lead valued, connected and fulfilling lives"* | **VERIFIED** — note the deck drops **"and fulfilling"** | | **~192,000 older NZers live alone** | **VERIFIED — and better than you thought.** **192,201** aged 65+ lived alone at the **2023 Census** = **25.3% of all over-65s and 49.4% of everyone in NZ who lives alone.** Say "at the 2023 Census", not "today". | | **6 in 10 felt lonely in last four weeks** (Age Concern NZ) | **WRONG — THE MOST DANGEROUS LINE IN THE DECK.** Wrong body (Age Concern **Auckland** + CHT Age Friendly Fund, not Age Concern NZ), wrong wording ("recently felt lonely and/or socially isolated", not "in the last four weeks"), and an **Auckland/Tauranga convenience sample presented as national**. The national four-week figure is **21.2%**, and only ~2% feel that way most/all of the time. A judge who checks finds a 59-vs-21 gap. **Removed from all three decks.** | | 65+ = 15% (2018) / 17% / 20% (2034) | **VERIFIED — and you undersold 2034.** 715,200 = **15.2%** (2018); **16.6%** at 2023 Census; and the Office for Seniors' own source says **21%** by 2034, not 20%. Corrected to 21% in all decks. | | Age bands 2018→2034 (six figures) | **VERIFIED — exact match** to Stats NZ 2018-base projections. ⚠ They sum to **747,800**, not the 715,200 Census count — don't put both on one slide without a footnote. | | Risk attributions | **Re-attribute.** +26% premature death = **Holt-Lunstad 2015**. +29% heart disease = **Valtorta 2016, *Heart***. +50% dementia = **CDC / National Academies 2020** (social **isolation**). Crediting all of them to Holt-Lunstad is an error on the slide. | --- ## 4. Deck fixes, in order of payoff 1. **State the SDG targets.** Free. Biggest single gain. See §1. 2. **Fill in the Impact Statement slide.** It is currently a blank template asking *"What is the number that makes it fundable? What is the closing line?"* — that is the slide judges score against. 3. **Delete the two judging-criteria slides.** They are your briefing notes. Leaving them in says you didn't review the file. 4. **Fix the order.** Margaret arrives at 17, after the product at 6. "Thank you" sits at 12 with six narrative slides behind it. → use **Version B**. 5. **Upgrade the 192,000.** It's verified — say **192,201 at the 2023 Census**, and add the punch: that's **a quarter of all over-65s and half of everyone in NZ who lives alone**. 6. **Say who pays.** You *do* have pricing — NZ$65 per resident per month — but it's buried. Put one line per phase: whānau pays / provider pays / who pays for Guardian. 7. **Name a competitor.** ElliQ, GrandPad, Joy for All exist. One line on why you're different beats being caught without one. 8. **Substantiate "tikanga-led."** Right now it is a claim with no evidence behind it — no named principles, no Māori partners credited, no cultural governance. **The Kete** is your genuine proof and it is currently only visible inside an image. Put it in the words. --- ## 5. Honest notes on the imagery - The new photography is **generated**, not real. It is consistent and unmistakably Aotearoa, but if anyone asks whether Margaret is a real person, say plainly that she is a composite persona illustrated with generated imagery. - For a **tikanga-led** pitch, a real photo of a real kaumātua, used with permission, beats any generated image. I did not generate Māori faces — AI renders moko and cultural markers badly, and getting that wrong in this room costs more than having no face at all. - The data slides are now **native PowerPoint shapes**, not pictures. Every number is editable text, so you can fix a figure in seconds and it stays sharp on a projector.