Every rule. Nothing hidden.

Contents

This Charter is the single authority for every Mature Miles Index score. Book One sets out what we stand for; Book Two sets out exactly how a score is made. We publish both in full, so that a traveller, a tourism board or a sceptic can see exactly how a score is made and check it for themselves.

1. ·Foreword

Book One · The Charter

1. IMission and Philosophy
2. IIThe Traveller and the Destination
3. IIIEvidence
4. IVWeighting, Scoring and Verification
5. VKnow Before You Go
6. VIIndependence and Integrity
7. VIIOur Standard and Our Promise

Book Two · The Rulebook

1. IAbout this Rulebook
2. IIDesign principles and shared definitions
3. IIIArchitecture at a glance
4. IVWhy five pillars and these weights
5. VPillar 1 — Health & Safety (35%)
6. VIPillar 2 — Getting Around (25%)
7. VIIPillar 3 — Cost (20%)
8. VIIIPillar 4 — Climate (10%)
9. IXPillar 5 — Things to Do (10%)
10. XWhat moved to Know-Before-You-Go
11. XIUniversal scoring rules
12. XIIResearch procedure
13. XIIIVerification and publication
14. XIVCost-band calibration and freeze
15. XVThe destination workbook
16. AAI Reviewer Brief
17. BWhat changed from v1.0
18. CConfirmed decisions (28 October 2026)

Version 2.0 · Adopted 28 October 2026

Mission and Philosophy

§1 Why the Mature Miles Index exists

Short version: travellers over 60 deserve better information than they get.

Most travel rankings are written for someone else, usually a younger traveller with a backpack, or a reader with a luxury budget. They rarely tell you what matters once you’re past 60. How far is the nearest hospital? Can you get from the airport to your hotel without a taxi? Will the walk to the cathedral involve 200 steps?

The Mature Miles Index sets out to answer those questions for the world’s most popular destinations, using evidence anyone can check. The aim is a score mature travellers can trust, so they can plan with confidence and see more of the world.

§2 What we mean by “evidence”

Every MMI score is built from published, public information: journey times from Google Maps, the World Health Organization’s figures, official travel advice, water authorities, weather services, published prices and the big booking sites. Anyone should be able to open a source and confirm the figure in a few minutes.

Everything comes from sources open to the public, so we don’t use paid databases or our own inspection visits.

An indicator is scored only when reliable public evidence exists, and a destination receives its overall score once all 11 indicators are filled.

§3 How we treat evidence

Ten working rules apply to every piece of evidence we use.

3.1 Use what already exists. Before we consider measuring anything ourselves, we look for a credible source that has already done the work.

3.2 One source per fact. Each figure comes from one named source, fixed before the research starts, so two researchers will always use the same one. The score as a whole draws on many sources.

3.3 It must help the traveller. A source can be perfectly reliable and still be no use to us. The question we ask is whether it tells an older traveller something they’d want to know.

3.4 Like with like. We only compare figures that measure the same thing in the same way.

3.5 Everything is dated. Each figure records the day it was captured. All the evidence for one destination is collected within 14 days, and the score stands for 12 months.

3.6 Limitations are stated. Where a source has a known weakness, we say so in the evidence record.

3.7 Gaps stay gaps. An indicator without reliable evidence is left blank, and the destination has no published score until it’s filled.

3.8 Ratings are used as filters. Some of our sources carry traveller ratings, Tripadvisor and GetYourGuide for example. We use those ratings only as fixed cut-offs, such as “4.5 stars or better”, applied the same way to every destination. We add no opinions of our own.

3.9 Published sources only. If a figure isn’t published where anyone can see it, we don’t use it.

3.10 Every score can be traced. Any published score can be followed back, figure by figure, to the page it came from.

§4 What the score is for

The MMI score is there to help travellers over 60 make better decisions about where to go.

Plenty of lists already name the world’s “best” cities. The question we answer is more practical: if you’re an active traveller over 60 and you choose this destination, what are you likely to find when you get there?

And just as important: how much of that can you know before you book? The more you know in advance, the fewer unpleasant surprises.

§5 Our approach

We start with the destination and the evidence, and let the score come out at the end. Our own likes and dislikes are left at the door. If we happen to love a city and it scores poorly, it scores poorly.

In practice, every part of the method was settled in this order:

1. What matters to active travellers over 60.
2. What of that can be measured reliably.
3. What evidence exists.
4. How that evidence can be compared from place to place.
5. How it turns into a score.
6. How much weight each part carries.

Where the evidence supports no clear conclusion, we say so. Our motto sums it up: every score earned.

The Traveller and the Destination

§6 Who we have in mind

We picture an active, independent traveller of about 60 or older: someone who travels without a tour guide holding a flag, but who appreciates the nearest hospital being close, public transport that works, and not having to haul a suitcase up a hill.

Those are the things the score covers: safety and healthcare, arriving easily, getting around without stairs or steep climbs, what a comfortable day costs, the weather, and having plenty worth doing.

Specialist accessibility for travellers with significant mobility, sight, hearing or other disabilities is outside our scope. It is a serious subject that needs its own method and its own experts. Where accessibility helps every older traveller, step-free routes for example, it is part of our score.

§7 What counts as a destination

We score three kinds of destination:

7.1 Cities, such as Paris, New York, Tokyo and Vienna.

7.2 Regions, such as Champagne, Tuscany and the Italian Lakes.

7.3 Islands and island groups, such as Crete, Mallorca and the Greek Islands.

We keep it to these three on purpose. How we choose which destinations to score is set out in §26.

§8 What we measure

The score looks at what a visit is actually like for a mature traveller.

There are five pillars, made up of 11 indicators:

1. Health & Safety (35%): the nearest emergency hospital, the strength of the health system, official travel advice, and whether the tap water is safe.
2. Getting Around (25%): getting from the airport without a taxi, reaching the main sights without a car, and how much climbing and walking that involves.
3. Cost (20%): what a comfortable day costs.
4. Climate (10%): how many months of the year are comfortable.
5. Things to Do (10%): the range of experiences on offer and the choice of day trips.

Each indicator had to pass the test in §16 to earn its place, and the weights follow what older travellers say matters most (§17). Hotels and restaurants are covered as advice instead of being scored (§19).

§9 The same measure for every destination

Every destination is measured with the same 11 indicators, the same score bands and the same rules. No rule names a particular place.

Cities, regions and islands differ, of course, and we deal with that through one fixed starting point: the official tourist information office in the best-connected town, which we call the Base. Every journey is measured from there. If a region or island really does need a car to see it properly, it scores lower on Getting Around, which is the right answer for a visitor who’d rather not drive abroad.

§10 Starting with the traveller

The method was built around one question: what would an older traveller want to know before spending money, time and energy on a trip?

To answer it, we walked through a whole journey, step by step. Planning and booking, the flight and the arrival, finding your feet, getting about, meals and rest, sightseeing, dealing with something going wrong, and finally getting home. At each step we asked what could make the trip harder, more uncomfortable, more expensive or less safe than expected.

The data came afterwards, chosen to answer those questions. That is also why every indicator in the Rulebook is written as a plain question, such as “If I fall ill, how quickly can I reach an emergency department?”

§11 Knowing before you go

A good MMI assessment should give you a fair picture of a destination before you leave home. Between the score and the Know Before You Go notes (§21), it should help you answer questions like these:

• How easy will it be to get around, and how easy is arrival?
• What will the weather be like, and when is the best time to go?
• How physically demanding is it?
• What happens if I need a doctor?
• Is it safe, and can I drink the water?
• What will a comfortable day cost?
• What is there to do, and where should I stay?

Evidence

§12 The evidence record

Every figure behind a score is recorded with:

• the indicator it belongs to;
• the source and a direct link to it;
• the figure itself;
• the date it was captured;
• a screenshot in which the figure can be read, or a dated copy of the page text;
• the score band it falls into.

Each destination’s evidence is kept in its workbook and a numbered evidence pack, so we can always show exactly which files produced a score.

§13 When evidence is missing

Some things that matter to travellers simply aren’t published in a form we can compare. Those indicators are left blank, and the record says: “Insufficient comparable evidence: not scored.” The destination then has no published score until the gap is filled. It is the honest way to deal with it, and readers should be able to see that we’ve done so.

§14 Contacting destinations

Now and then we may contact a tourism office, a municipality or an authority when we can’t find an important public source. We ask one thing: where is this information published?

Anything a destination tells us is used only to find the public source.

Our certification questionnaire, which destinations and businesses complete when they apply for MMI certification, is a separate matter altogether. It is used for certification only.

§15 Corrections from destinations

A destination may point us to a published figure or tell us about a factual mistake. We check it against the public source and, if we got it wrong, correct the score and record the change.

§16 How an indicator earns its place

Before an indicator becomes part of the method, it has to pass six questions:

16.1 Does it matter? Does it make a real difference to a senior traveller’s trip?

16.2 Is there good evidence? Can reliable public information be found?

16.3 Can it be compared? Can it be measured the same way in every destination?

16.4 Does it help decide? Will the result help someone choose where to go?

16.5 Is it new information? Does it tell us something the other indicators don’t?

16.6 Does it predict the experience? Does it help forecast what the trip will actually be like?

Weighting, Scoring and Verification

§17 How the pillars are weighted

Some things matter more than others. Visitor safety and the price of a cup of coffee can both be measured, but nobody would give them equal weight.

The weights also had to come from somewhere other than our own preferences. They follow published surveys of senior travellers in the UK, the US, Canada, Australia and New Zealand: what they say matters, and how consistently they say it.

The result is Health & Safety 35%, Getting Around 25%, Cost 20%, Climate 10% and Things to Do 10%.

The full reasoning is in Book Two. The weights are reviewed once a year, and any change counts as a new version of the method, with every affected score recalculated.

§18 How a score is calculated

The arithmetic is published and fixed:

• each indicator scores a whole number from 1 to 5, against score bands set before any research starts;
• an indicator without evidence stays blank and is not treated as a zero;
• a pillar’s score is the weighted average of its indicators;
• each pillar then contributes its share of the 100 points, so a pillar scoring 4 out of 5 earns four-fifths of its share;
• the five shares add up to the MMI score out of 100.

Two researchers working from the same evidence and the same Rulebook should arrive at the same score.

§19 Hotels and restaurants

Where to stay, where to eat and which discounts apply all matter on a trip. But they’re largely personal, they change constantly, and surveys show they rarely decide where older travellers go.

So hotels and restaurants are covered as advice. Each destination’s Know Before You Go notes (§21) cover the best area to stay in, dining tips, senior discounts and any official food-safety warnings, all with their sources.

§20 The second check

A finished assessment stays a draft until someone else has checked it.

That second checker, a person or an AI working only from the destination’s workbook, its evidence pack and a written brief, opens every source and screenshot, confirms every figure and score band, and lists anything they disagree with. Disagreements are settled by capturing the evidence again.

A score is published only once all 11 indicators are scored and the second check is complete.

Know Before You Go

§21 Know Before You Go

Alongside every score we publish practical notes on the things a traveller should know before booking. They are sourced like everything else, and they don’t count towards the score. They may include:

Best time to visit: the comfortable months, and any months flagged for heat.

Best base: the most practical area to stay in, and why.

Watch out for: steep districts, official warnings and anything else that could catch you out.

Senior discounts: concessions on transport and at the main sights.

Dining notes: usual meal times, and how easy vegetarian or gluten-free eating is.

Evenings out: where the pubs, bars and live music are, and whether there’s an organised pub crawl.

Emergency numbers: who to call if something goes wrong.

§22 Personal taste

There is no perfect destination for everyone. One traveller wants museums, architecture and good restaurants; another wants beaches, nature and quiet streets.

The score tells you how well a destination works for a mature traveller. Whether you’ll love it is up to you. That is why Things to Do carries the lightest weight, and why our Find Your Best Fit tool lets you set your own priorities.

§23 Things to Do

Counting attractions tells you very little. A long list can still be a poor one for a mature traveller.

So the pillar looks at two things. The first is variety: how many of eight kinds of experience are available and well reviewed, including two easy ways to see the sights, such as a hop-on hop-off bus or a boat trip. The second is day trips: how many different places organised day trips actually go to. In Barcelona, for instance, ten of the first 30 day trips listed were to Montserrat, so it counts once.

Nightlife, pub crawls and extreme sports are left out of the score. Evenings out get their own note in Know Before You Go (§21).

§24 Cost

The Cost pillar measures what a comfortable day costs per person: a night in a 3- or 4-star hotel, a mid-range dinner, a simple lunch, a day’s public transport pass and entry to the main sights. It is scored against fixed dollar bands, so one destination’s score stays the same however many others are added.

An expensive destination can still be excellent value, and a cheap one can disappoint. The score tells you what a comfortable day will cost; whether that’s good value is your call.

§25 The Base

Nobody experiences a whole city or region at once, so we measure every journey from one fixed point: the official tourist information office in the Base town.

Advice on which neighbourhood or town to stay in goes into Know Before You Go as the “best base”. A destination can score well overall and still come with a note to choose your base carefully.

Independence and Integrity

§26 How destinations reach the Index

The Index is published every month as The Mature Miles Index · [Month Year]. The first monthly Index appears on 1 January 2027. Each month’s Index carries last month’s destinations forward and adds the newcomers. It shows each destination’s name, its country, its score, and whether it is new or has moved up or down.

A destination reaches the Index in one of two ways.

Certified. A destination applies, is assessed, meets the MMI standard and pays its assessment fee. It joins the next monthly Index and stays on it for as long as its certificate is valid.

Stowaway. Overlooked by the popular lists, spotted by us. Each month we choose a small number of cities and towns that do not appear on the world rankings of four well-known lists (Tripadvisor Travellers’ Choice, Time Out’s best cities, Travel + Leisure World’s Best Awards and Condé Nast Readers’ Choice), using their latest editions. A place is excluded if it appears in the top 25 of any of them, or anywhere on a list that publishes fewer than 25. Regional and national rankings are not used. To make sure travellers have actually heard of them, the most-reviewed of its five Key Sights must have at least 5,000 Tripadvisor reviews. Stowaways are scored under exactly the same rules, and only those that meet the MMI standard are listed. Each is shown, marked as a Stowaway, for the three months after its assessment. Stowaways never make up more than half of any month’s Index.

If official travel advice says to avoid a destination, it is not assessed.

§27 How MMI is funded

MMI has to earn its keep, through certification fees and, in time, sponsorship. Certification is awarded on the published score, and every destination, certified or not, is scored under the same published rules.

§28 Transparency

The whole method is published in this Charter, so that anyone can see what we measure and why, where the evidence comes from, how it is scored and weighted, and where its limits are.

Every published score states the method version, the dates the evidence was captured, and any warnings a traveller should know about. Where it is reasonable and lawful, we make the underlying evidence available to readers, journalists and researchers.

§29 When sources disagree

Sources do sometimes disagree. We deal with that by fixing one named source for each fact before research begins, so there is nothing to choose between.

If the second check finds a figure that doesn’t match the evidence, the evidence is captured again.

Where an uncertainty genuinely can’t be resolved, we report it as an uncertainty.

§30 Keeping the method up to date

No method is perfect on its first outing, and ours will improve as better sources appear, as travellers’ needs change and as real assessments show up weaknesses. The method is reviewed every year, including the cost bands against inflation.

Every change goes into a dated decision record that names the rule it replaces.

If we find a factual mistake after publication, we correct it, recalculate the score and log the change with its date.

§31 Testing a new indicator

Before any new indicator joins the method, we put it through an evidence audit. Who publishes this information, and how reliable is it? How current is it? Can it be compared between destinations? What are its limits? Can we get it the same way every time? And does it genuinely help predict what a traveller will experience?

Our Standard and Our Promise

§32 Testing the method

Before publishing any scores, we tested the method on real destinations of different kinds. The first round covered Barcelona, Vienna and the Italian Lakes: two great cities and a region. The point was to find out whether the method itself held up, and every lesson learned went into the decision record. Barcelona taught us, for example, that the first hospital Google lists isn’t always the nearest one.

The cost bands are calibrated on the first ten destinations scored and then fixed.

The first monthly Index, covering at least 20 destinations, is published on 1 January 2027 and updated every month after that.

§33 The test we hold ourselves to

Could a journalist, a traveller, a tourism board or a researcher look at our evidence and understand why a destination received its score? If the answer is no, the assessment isn’t finished.

We expect to be challenged, and we welcome it. A good index should be able to answer the questions put to it.

§34 What we will never do

We will never:

• invent evidence, or pretend missing evidence doesn’t matter;
• present opinion as fact;
• adjust scores to produce a more attractive ranking;
• hide a known weakness in the method;
• treat a traveller’s anecdote as if it were an official statistic;
• treat AI-generated text as evidence;
• change the method to get a result we want.

§35 The question we ask ourselves

Every decision about the method has to pass one question: would we make the same decision if the result went the other way?

If the honest answer is no, the decision gets looked at again.

§36 What travellers can expect from us

Nobody can promise a trip without surprises. What we can offer is a well-researched, openly checked picture of a destination before you spend your money and board the plane.

If, after reading an MMI assessment, a traveller can say, “I’ve got a fair idea of what I’m letting myself in for,” we’ve done our job.

§37 Every score earned

Everything in this Charter comes back to our motto: every score earned.

It means each score rests on evidence anyone can check, gathered under rules published in advance and checked by a second pair of eyes.

The real measure of success is the day a traveller says, “Before we book, what’s the MMI score?”

About this Rulebook

This Rulebook is the single authority for scoring any destination on the Mature Miles Index (MMI): five pillars, 11 indicators, one destination workbook. Where the workbook and this Rulebook ever disagree, the Rulebook governs until the workbook is corrected through the decision record.

The method is destination-neutral: no destination is named in any rule. Destination facts live only in each destination's workbook.

Design principles and shared definitions

Every indicator answers a question a 60+ traveller would actually ask, using a source anyone can open and check in minutes.

Principles

• Traveller first. Each indicator states the plain question it answers ("If I fall ill, how quickly can I reach an emergency department?").

• One source per fact. Each input has one named primary source, so two researchers cannot pick different ones.

• Public and checkable. No paid databases, no site visits, no specialist audits. A complete destination should take about one working day to research.

• Built to separate destinations. Score bands are set so that ordinary destinations spread across 1–5 rather than all scoring 5.

• Measured once, reused. The same Base and Key Sights feed several pillars.

Shared definitions (frozen before research starts)

Architecture at a glance

Five pillars, 11 indicators, every indicator scored 1–5, pillar weights totalling 100.

Why five pillars and these weights

Surveys of older travellers across the Anglosphere point to the same five concerns; hotels and restaurants do not appear as decision factors. The weights follow how consistently and how often each concern is cited (Charter §17).

Sources: AARP 2026 · Luggage Forward 2026 · Global Rescue 2026 · Safer Tourism Foundation · ABTA · MMGY/ENAT/WTTC 2025 · Australian Seniors 2022 · THIA Canada 2025 · NZ Seniors 2024 · Atmosphere 2016

Pillar 1 — Health & Safety (35%)

Four equally weighted indicators: two measured at the destination itself, two at country level. The Global Peace Index is dropped, because Canada's travel advice already reflects national security and also names local petty crime.

Why Canada's advice: it is in English, uses four fixed risk levels, sets levels by region within a country, and names specific places in its petty-crime warnings. Using one government keeps the measure consistent.

In plain English — why several hospitals? A traveller who falls ill asks the hotel or a shopkeeper for the nearest hospital and goes there. Indicator 1.1 measures exactly that one hospital: how long it takes to reach the nearest 24-hour emergency department. The researcher compares several search results only to be sure of finding the nearest one, because Google does not list hospitals in order of distance. In Barcelona, Google listed Hospital del Mar first, but Hospital Clínic is closer.

Pillar 2 — Getting Around (25%)

Three equally weighted indicators covering the whole journey: arriving, reaching the sights, and doing it without stairs or steep walks. All are measured in Google Maps from Base point 1, so a reviewer can repeat every check in minutes.

Pillar 2 measures from Base point 1 only, even where a second Base is used: Base 1 is the main hub by definition, and measuring twice would double the research. For Regions and Islands, 2.2 naturally scores lower where a car is needed; that is the correct result for a traveller who does not want to drive abroad.

In plain English — what about taxis? Pillar 2 measures how easily a visitor gets around independently by public transport, at the day-ticket cost counted in Pillar 3. A taxi can reach almost any sight, but at a price: if Getting Around assumed taxis, Cost would have to count taxi fares, and the destination would lose the points there instead. Taxis are already taken into account: 2.2 compares each public-transport trip with the driving (taxi) time, and a sight only fails when public transport is much slower. Practical taxi tips for sights with a steep or long last stretch belong in Know Before You Go.

Pillar 3 — Cost (20%)

One indicator: what a comfortable day actually costs, scored against fixed dollar bands, so adding or removing a destination never changes another destination's score.

Hotel price sample (frozen): search the Base town on Booking.com for the travel-window dates (2 adults, 1 room, 3 nights). Filters: property type Hotels, 3 and 4 stars, "Less than 3 km" from the centre. Sort by "Distance from city centre" and record the total price of the first 10 hotels, and the number of properties the search reports. The evidence must show every hotel's name and total. Sorting by distance avoids Booking.com's paid ranking.

The dollar bands are provisional until the first scoring round, then frozen (see Cost-band calibration and freeze).

Pillar 4 — Climate (10%)

One indicator: how many months of the year are comfortable for an older traveller. That replaces the annual temperature average and the Specialist profiles, which rewarded heat.

Two unscored outputs go to Know-Before-You-Go: the list of Comfortable Months (the best time to visit) and any month with a mean daily maximum of 32 °C or more, flagged as a heat-risk month.

Pillar 5 — Things to Do (10%)

Two equally weighted indicators from one platform, GetYourGuide, with a fixed search depth so the score never depends on how long the researcher keeps looking. A listing counts only if it is rated 4.5 or higher with at least 50 reviews, and is something an active 60+ traveller would realistically do (no nightlife, pub crawls or extreme sports).

Two of the eight types are low-effort sightseeing, so a destination that offers ways to see the highlights without long walks scores higher. One listing counts towards one type only; borderline inclusions get a one-line reason in the evidence log.

In plain English — why 30 day trips? No visitor weighs up 30 day trips; most choose one or two. Indicator 5.2 does not measure the choice a visitor makes, but how much real choice there is. Thirty listings show whether a destination offers many different worthwhile places, or one famous trip sold many times over. Barcelona's first 30 listings name 10 different places, although Montserrat alone fills 10 of them.

What moved to Know-Before-You-Go

Accommodation and Food are no longer scored. What a traveller needs from them is advice, which each template's destination-specific tab records as evidenced but unscored notes.

Universal scoring rules

Six rules apply word-for-word in every pillar and template.

1. One raw-score scale. Every indicator scores a whole number from 1 to 5. Zero is not a valid score.

2. One missing-evidence rule. Insufficient evidence is not a score, not a zero and not automatically a failure. Leave the score blank and state what remains unverified. A pillar with a blank indicator has no pillar score, and the destination has no published overall score until the gap is closed.

3. One aggregation rule. Pillar score = the weighted mean of its indicator scores using the internal weights above, which add up to 100% in each pillar. Where weights are equal, use the arithmetic mean.

4. One contribution formula. Pillar contribution = pillar score ÷ 5 × pillar weight, in percentage points. Overall MMI score = the sum of the five contributions, out of 100.

5. One evidence standard. A frozen destination boundary and Base; one named source per input; direct evidence (not search pages or AI-generated text); a capture date; and the exact score band applied.

6. One version-controlled decision record. Every change identifies the rule it replaces and the templates and tabs that need updating.

Rounding: keep full precision in calculations; display pillar scores and contributions to two decimal places. Thresholds are applied to the source value exactly as published, before any rounding.

Research procedure

One researcher completes one MMI Destination Workbook per destination, in about 4 to 4½ hours, within a 14-day capture window.

1. Save a copy of the workbook named after the destination and the capture month.

2. Start Here: enter the destination profile and the 5 Key Sights once. Every step reuses them.

3. Step 1 Climate (sets dates): done first only because it works out the travel-window date used for driving times and the hotel search. Climate is one of the two lightest pillars.

4. Steps 2–5: Health & Safety, Getting Around, Cost, Things to Do. Each row says what to find, how, and links to the source.

5. Evidence: for every answer, record the link used, the date and a screenshot named with the row ID in which the figure is readable (plus a dated text capture for lists). Mark the row Verified only when all three exist.

6. Know Before You Go: add the destination-specific notes a 60+ traveller needs (best base, senior discounts, emergency numbers, steep areas). Evidenced, never scored.

7. Never guess. A row that cannot be evidenced stays 'Not yet', with what is missing written in its Note.

The Start Here tab lists the pillars in order of weight and shows the step order beside them, so the working order never implies importance.

Verification and publication

A completed workbook is a draft until a second checker, a person or an AI working only from the issue-stamped workbook, its evidence zip and the AI Reviewer Brief, has confirmed every Verified row.

Cost-band calibration and freeze

The Cost bands (≤$100, $101–150, $151–200, $201–275, >$275) are provisional and will be frozen after the first real scoring round.

1. Complete workbooks for the first 10 destinations, deliberately spread from inexpensive to expensive and covering at least one Region and one Island.

2. Check the spread of Comfortable Day costs. The bands pass if the 10 destinations use at least four of the five bands and no single band holds more than four.

3. If they fail, move the band edges once, to round dollar figures, so that they pass; record the change in the decision record.

4. Freeze the bands for v2.0. They are reviewed annually against inflation, and any change is a new version with all affected scores recalculated.

Indicative check (28 October 2026): Numbeo meal prices for one European capital (€55 mid-range dinner for two, €15 lunch), plus transport, sights and a typical mid-range double room, give roughly $150 a day, the middle of the scale, as intended. The hotel samples needed for a full calibration require a live Booking.com search.

The destination workbook

One Excel workbook per destination replaces the seven separate v1.0 workbooks. Only yellow cells are typed in; green cells and the Rules tab are locked.

AI Reviewer Brief

Paste the text below into any AI (Claude, Gemini, ChatGPT) together with this Rulebook and a completed workbook. It turns the AI into the second checker, not a second researcher.

MMI AI REVIEWER BRIEF — v2.3 (29 Oct 2026)

You are the independent second checker for one Mature Miles Index (MMI) destination score. You CHECK the evidence and TEST the researcher's judgement. You do not research the destination again, and you do not rescore it from your own knowledge.

WHAT YOU RECEIVE (all three, from the same handover message)
1. The MMI Master Methodology & Rulebook v2.0 (PDF or text).
2. The destination workbook: MMI_<Destination>_<ISSUE>.xlsx
3. The evidence pack: MMI_<Destination>_<ISSUE>_evidence.zip, which contains MANIFEST.txt.

STEP 0 — CONFIRM YOU HAVE THE RIGHT PACKAGE. DO THIS BEFORE ANYTHING ELSE.
a. Read the ISSUE code (e.g. BCN-04) in three places: the workbook file name, the workbook's Start Here tab (row 'Workbook issue'), and the first line of MANIFEST.txt.
b. Compute the SHA-256 of the workbook file and compare it with the hash in MANIFEST.txt.
c. If any of these do not match, or MANIFEST.txt is missing, STOP. Reply only: "WRONG OR MIXED PACKAGE — <what does not match>". Do not review an older or different copy. Ignore any other MMI files that may be in your workspace (older workbooks, v1.0 category files): review only the files named with this ISSUE code.

STEP 1 — EVIDENCE CHECK, EVERY ROW MARKED 'Verified'
For each Verified row on Start Here and Steps 1–5:
1. Find its evidence: the link (Link / Link you used / Note) AND the evidence file (Screenshot file / Note): a screenshot, or a dated text capture (.txt) of the page. Every file named must exist in the zip. List any that do not.
2. Open the evidence file and compare it with the value typed in the workbook: same figure, unit, date, place and name.
3. Open the link if you can.
   - If the live page shows the same value: CONFIRMED.
   - If the live page now shows something different, but the evidence file (dated inside the capture window) shows the workbook value: SOURCE CHANGED. The evidence file is the record; report both values. Live prices, exchange rates, review counts and search results change daily, so SOURCE CHANGED is expected and is NOT a failure: the row counts as confirmed.
   - If you cannot open the link, use the evidence file alone and say so: CONFIRMED FROM FILE.
   - If neither the link nor the file shows the value: UNCONFIRMED (not a fail, not a zero). Say what is missing.
   - If the evidence shows a different value: DISAGREE. Give the value you see.
4. Check the date checked lies within the 14-day capture window on Start Here.
5. Check the score band applied matches the Rules tab thresholds.
6. Report what you see in the evidence, not what you expect. Never describe a rule or method that is not in the Rulebook or the workbook (for example, an age bracket, discount or filter that was never used). If you are unsure how a value was found, say UNCONFIRMED and ask.
COMPLETENESS: Start Here shows 'Verified rows' beside 'Methodology version': the number of rows marked Verified across all tabs, with the count per tab. Your evidence table must have exactly that many rows, one per row ID, in tab order, including every row on Steps 3, 4 and 5. Do not group, summarise or skip rows ("all others confirmed" is not allowed). If you cannot finish every row, say so in section 1 and give the number you did check; a partial table is reported as INCOMPLETE REVIEW, never as a pass.
Files named SUPERSEDED_<file> are earlier captures kept for the record (rule MMI-2.0-06). No workbook row cites them; do not check workbook values against them.
Month rows on Step 1 take their evidence from rows 4a–4c. The 30 day-trip rows on Step 5 share one category page, one screenshot and one text list of all 30 listings.

STEP 2 — TEST THE JUDGEMENT CALLS. THIS IS REQUIRED, NOT OPTIONAL.
A real checker questions every decision where a reasonable researcher could have chosen differently. For each item below, say AGREE or QUESTION. For every QUESTION give: the rule it depends on, the alternative choice, and whether it would change an indicator score (and by how much).
- Base town and Base point: is this the official tourist office the rules point to? Do the journeys start at that office's own Google Maps place pin (not a typed address)? For a Region, is the base reason recorded in the Change Record?
- Key Sights: are these the first 5 actual places (not tours or tickets) in Tripadvisor's 'Top Attractions in <destination>' list, in default order?
- 1.1 Hospital: is it the nearest 24-hour emergency department by drive time? Within the first 6 search results, was each hospital counted once and were non-hospitals (primary-care or urgent-care centres, clinics) skipped? Were they measured to their emergency-department entrances? Was the higher figure of the range used?
- 1.3 Travel advice: correct level? Is the destination itself (not just the country) named for petty crime?
- 1.4 Tap water: is any official local caveat being missed?
- 2.1 Airport: is it the airport the official tourism board lists first (evidence: the board's own page)? Fastest public transport, correct number of changes? If Google warned its timetables were out of date, was the operator's planner used?
- 2.2 / 2.3 Each Key Sight: walk-only decision (≤10 min); public transport = the shortest of all options Google lists, not the first listed; public-transport vs 2× drive time; changes; step-free classification; last-stretch metres and climb; use of the 'Mostly flat' = 0 m rule.
- 3.1 Cost: hotel search filters and dates as specified; median of the first 10; Numbeo items are the specified ones; transport ticket type; adult Key Sight prices, use of official 'from' prices, free sights = 0; one exchange rate.
- 4.1 Climate: right station and 1991–2020 period (if several tables qualify, the station nearest the Base point; tables rejected and why); any borderline month (within 0.5 °C or 0.5 rain days of a limit).
- 5.1 Variety: were the fixed search terms used, and is each listing the first qualifying one for its type (≥4.5 rating, ≥50 reviews, suitable for an active 60+)? Is 'None found' justified?
- 5.2 Day trips: main place chosen for each listing; exclusions (in-city tours, unsuitable activities, under 50 reviews); places with no name; duplicates counted once.

STEP 3 — WHAT YOU MUST NOT DO
- Never replace a value with one from memory, another website or an estimate.
- Never change weights, thresholds or definitions. Put disagreements with a rule under 'Method comments'.
- Never add labels, tiers or rankings the Rulebook does not define (e.g. 'High-Quality Destination'). The result is a score out of 100, nothing more.
- Never give an overall verdict such as 'ready for publication', 'approved', 'clean' or 'compliant'. Your job ends at the six sections below. Whether the score is published is decided by Mature Miles, not by you.
- Do not report a column heading such as '#' as a formula error. Report only real Excel errors (#VALUE!, #REF!, #DIV/0!, #NAME?, #N/A).

WHAT TO RETURN
1. Package check: ISSUE code, hash match yes/no, and the 'Verified rows' count from Start Here versus the number of rows in your evidence table (they must be equal).
2. Evidence table, one row per Verified workbook row:
   Tab | Row ID | Workbook value | What the evidence shows | Live source (if opened) | Result (CONFIRMED / CONFIRMED FROM FILE / SOURCE CHANGED / UNCONFIRMED / DISAGREE) | Note
3. Missing evidence: every named file not in the zip, and every Verified row with no link or no file.
4. Judgement calls: the Step 2 list, each AGREE or QUESTION with the details asked for.
5. Recalculated scores: each pillar and the total from CONFIRMED, CONFIRMED FROM FILE and SOURCE CHANGED values (the dated evidence value). Leave out only UNCONFIRMED and DISAGREE rows; if that leaves an indicator without a value, say which and why. State whether any QUESTION would change the total if accepted, and by how much.
6. Method comments (optional).

What changed from v1.0

The six universal rules are kept; the structure, weights and most indicators are replaced by fewer, checkable measures.

Confirmed decisions (28 October 2026)

All five points were approved, one with an amendment.

• Single sources: Government of Canada for travel advice, Numbeo for meal prices, GetYourGuide for activities, Booking.com for hotels, Tripadvisor for Key Sights and restaurants.

• Hotel facilities (6.2): approved with a fallback to the hotel's website and Tripadvisor; superseded later the same day when hotel facilities moved to Know-Before-You-Go. Everywhere, unverified stays blank.

• Provisional bands for Cost (3.1), frozen after the first scoring round (Dietary options later moved to Know-Before-You-Go).

• Travel window tied to the first Comfortable Month.

• Template format: Excel files (.xlsx) with working formulas (later the same day: one workbook per destination, below).

Confirmed 28 October 2026 (later the same day):

• Five pillars replace seven categories; Accommodation and Food become Know-Before-You-Go advice.

• Weights 35 / 25 / 20 / 10 / 10, based on US, UK, Australian, Canadian and New Zealand survey evidence.

• Pillar 2 measured from Base point 1 only.

• One destination workbook replaces the separate templates (novice-friendly).

• Working order kept (Climate first, because it sets dates); the pillars are shown in weight order everywhere else.

• Wikipedia climate table first where it cites the national weather service for 1991–2020.

Clarifications from the Barcelona pilot (approved 28 Oct 2026)

• MMI-2.0-P2-02 — Public-transport journeys use 10:00 on the first Tuesday after the capture start; driving times and hotel prices keep the travel window (Google can't plan public transport months ahead).

• MMI-2.0-P2-03 — Where Google says a walk is 'Mostly flat' and gives no figure, the climb is recorded as 0 m.

• MMI-2.0-P3-02 — An attraction with dynamic pricing uses its official 'from' price.

• MMI-2.0-P5-02 — Day-trip listings that stay inside the Base town (city tours, tuk-tuks, hop-on buses) are recorded but marked not suitable.

• MMI-2.0-P5-03 — A day-trip title naming no place counts as one place, once; a title starting with the home country uses the next place named.

Core pillar designs (approved 28 Oct 2026, no changes)

• MMI-2.0-P1-01 — Health & Safety: hospital access, WHO health system strength, Canada travel advice, safe tap water.

• MMI-2.0-P1-02 — Base point = official tourist office; nearest 24-hour emergency department; 10:00 travel-window Tuesday; higher figure of any range.

• MMI-2.0-P1-03 — WHO UHC latest year; bands 85/75/65/55.

• MMI-2.0-P3-01 — Comfortable Day cost in USD against fixed bands (provisional until 10 destinations).

• MMI-2.0-P4-01 — Count of Comfortable Months (16–27 °C, ≤10 rain days); heat-risk months flagged.

• MMI-2.0-P5-01 — GetYourGuide only: 8-type variety and distinct places in the first 30 day trips; ≥4.5 rating, ≥50 reviews.

Further decisions (approved 28 Oct 2026)

• MMI-2.0-P2-04 — If Google warns it doesn't have the latest timetables, use the train or bus operator's own journey planner and note which.

• MMI-2.0-03 — Region Base town = the best-connected town with an official tourist office from which the main sights are easiest to reach by public transport (may be the hub just outside the region). Italian Lakes base: Milan.

• MMI-2.0-04 — Every handover is issue-stamped: workbook MMI_<Destination><ISSUE>.xlsx, evidence MMI<Destination>_<ISSUE>_evidence.zip with MANIFEST.txt (issue code, workbook SHA-256, file list). Reviewers stop if the three don't match.

• AI Reviewer Brief v2.1 — package check, row-by-row evidence check, required judgement-call test.

Bullet-proofing after the ChatGPT cross-review (approved 28 Oct 2026)

• MMI-2.0-05 — A screenshot counts only if the figure it supports is readable in it; lists also get a dated text capture; where a site cannot be screenshotted, a dated text capture of the official page is the evidence.

• MMI-2.0-02b — Key Sights = the list headed 'Top Attractions in <destination>' on tripadvisor.com Things to Do (logged out, English, USD), default order.

• MMI-2.0-P1-04 — 1.1: search 'hospital emergency room' around the Base point; compare the first 6 results to their emergency-department entrances; nearest = lowest higher-figure; confirm 24-hour opening.

• MMI-2.0-P2-05 — 2.3: last stretch measured on the step-free route if one exists, otherwise the fastest route, to the sight's Google Maps pin; screenshot shows metres and climb.

• MMI-2.0-P3-03 — Free entry needs evidence: the official site says free, or Tripadvisor lists the sight under 'Free Entry'.

• MMI-2.0-P3-04 — 3.1a records the number of properties the search reports; the sample uses the first 10.

• MMI-2.0-P5-04 — 5.1 uses one fixed search term per type; the first listing whose title names a thing of that type and meets the tests; up to 20 listings checked.

Second cross-review fixes (29 October 2026, after the BCN-06 ChatGPT and Gemini reviews)

• MMI-2.0-06 — each value comes from one capture; re-checked evidence is kept in the zip renamed SUPERSEDED_<file> and is no longer cited.

• MMI-2.0-07 — every journey and distance starts at the tourist office's own Google Maps place pin, never a typed address.

• MMI-2.0-P1-05 — 1.1: within the first 6 results, each hospital counts once; non-hospitals (primary-care or urgent-care centres, clinics) are skipped.

• MMI-2.0-P2-06 — fastest public transport = the shortest of all options Google lists, not the first listed; last-stretch metres = the final walk in that route's Details.

• MMI-2.0-P4-03 — 4.1: if several Wikipedia tables cite the national weather service for 1991–2020, use the station nearest the Base point and record the tables rejected.

• AI Reviewer Brief v2.3 — exact Verified-row count, no invented methods, no overall verdicts, SOURCE CHANGED counts as confirmed, SUPERSEDED files ignored.

Clarifications from the first Vienna scoring (29 October 2026)

• MMI-2.0-P1-06 — where Google lists more than one driving route, use the lowest higher-figure among them.

• MMI-2.0-P1-07 — a tap-water caveat is an official warning about the public supply; advice about private plumbing in buildings (e.g. old lead pipes) is not a caveat and goes in Know Before You Go.

• MMI-2.0-P2-07 — if the Base point lies inside an area Key Sight (old town, district), that sight is walk-only with 0 m and 0 climb.

• MMI-2.0-P2-08 — if Google's wheelchair option returns only a walking route, the sight is step-free and the last stretch is the final walk of the fastest public-transport route.

• MMI-2.0-P3-05 (approved 29 Oct 2026) — hotel totals use the price a logged-out visitor pays; members-only prices are ignored.

• MMI-2.0-P5-05 (approved 29 Oct 2026) — a day-trip title that names the Base town first uses the next place named.

• MMI-2.0-C1 (approved 30 Oct 2026) — certification standard, fixed and published before any scoring: a destination is MMI Certified when its total score is 75 or more out of 100 and no Health & Safety indicator scores below 3. A total of 75 means the destination averages between 3.5 and 4 out of 5 across all 11 indicators. The standard is reviewed openly once a year.

• MMI-2.0-C2 (approved 30 Oct 2026) — a destination below the standard receives its full report, showing what held it back, and may apply for reassessment no sooner than 3 months after its result, at 75% of the original fee.

• MMI-2.0-C3 (approved 30 Oct 2026, revised the same day) — the certificate year is set by the date of payment. Payment received January to September gives a certificate for that year; payment received October to December gives a certificate for the following year. The fee is the same all year. A certificate may be displayed indefinitely, since its year is printed on it. Launch exception: certificates issued in November and December 2026 are 2027 certificates.

• MMI-2.0-C4 (approved 30 Oct 2026) — the Index is published every month as The Mature Miles Index · [Month Year], from 1 January 2027, showing each destination's name, country, score and movement (new, up or down). A Certified destination joins the next monthly Index once it has met the standard and paid its assessment fee, and stays while its certificate is valid. Applicants who pay from October to December choose whether to appear on the Index straight away or from the following January.

• MMI-2.0-C5 (approved 30 Oct 2026) — Stowaways (“overlooked by the popular lists, spotted by us”): cities and towns that do not appear in the top 25 of the latest world rankings of Tripadvisor Travellers' Choice, Time Out's best cities, Travel + Leisure World's Best Awards and Condé Nast Readers' Choice, or anywhere on a list that publishes fewer than 25 (regional and national rankings are not used), and whose most-reviewed Key Sight has at least 5,000 Tripadvisor reviews. Scored under the same rules; listed only if they meet the standard, marked as Stowaways, for the three months after assessment; never more than half of any month's Index. Failed Stowaway scores are not published. Stowaways receive no certificate, badge or report unless they apply and pay, when they become Certified and the Stowaway mark is removed.

• MMI-2.0-C6 (approved 30 Oct 2026) — destinations certified before 1 January 2027 carry the title Early Boarder permanently.