The Oh-No Travel Index scores a trip plan from 0 to 20, and lower is better. Each point is a moment a traveler finds out the hard way: a passport with five months left, a museum shut on a Monday, seven stops in one day. A plan earns a point for each of 15 must-dos it leaves unanswered and each of 5 faults it makes on the day.
- 0 is a plan that answers every must-do and makes none of the day faults.
- Must-dos count only when the plan answers them, not when it names the topic.
- The sources are 115 Reddit threads about trips that went wrong and a study of 356 AI-generated itineraries.
- No day plan means the 5 day faults are "not scored", never scored as 0.
The 20 items
The items run in the order a trip does, from before you fly to the last day.
| # | Before you go | The oh-no when a plan misses it |
|---|---|---|
| 1 | Entry papers: visa, passport validity, travel authorization | Turned away at check-in with five months left on a passport |
| 2 | Airport to the city | You work out the trains in the arrivals hall |
| 3 | Where to stay | A hotel an hour from everything you planned |
| 4 | The transit card and how trains work | A queue for paper tickets at every gate |
| 5 | Whether a rail pass is worth it | A national pass for a city-only trip |
| 6 | Cash or card | A cash-only lunch counter, and no cash |
| 7 | SIM, eSIM or pocket Wi-Fi | No maps the moment you leave the hotel Wi-Fi |
| 8 | Etiquette and tipping | A tip handed back, or shoes on where they come off |
| 9 | What to book ahead | The timed-entry museum sold out a month ago |
| 10 | Closures and holidays on your dates | The garden shut on Mondays, or a national holiday crowd |
| 11 | Weather and what to pack | Flip flops in a 12°C November |
| 12 | Scams and tourist traps | A bar tab many times the posted price |
| 13 | A day trip | A free day spent on the hotel's printed map |
| 14 | The must-do sights | Home without the one place everyone asks about |
| 15 | The leaving day | The right flight time, the wrong airport |
| # | On the day | The oh-no when a plan makes it |
|---|---|---|
| 16 | Arrival day planned like a full day | A museum at 2 PM after a night flight |
| 17 | Too much in one day | Seven stops, and you remember none of them |
| 18 | A stop closed when you arrive | The market stalls packed up two hours before your slot |
| 19 | A day that doubles back | Two train rides to undo the order of your own day |
| 20 | The same place twice | Day 6 sends you back to day 2's temple |
The specifics change by destination: item 4 is Suica or Pasmo in Tokyo and an Oyster card in London, and item 5 is the JR Pass in Japan and a Eurail pass in Europe.
How is a plan scored?
A must-do scores a point when the plan leaves it unanswered or answers it wrongly. Recommending a nationwide JR Pass for a Tokyo-only week is wrong, and so is a packing list that does not fit the forecast. Naming a topic is not answering it: "check the opening hours" does not answer item 10.
A day fault scores a point when it occurs at least once in the trip:
- Item 16: the arrival day carries more than half a normal day's stops for the chosen pace.
- Item 17: a day carries more named stops than the chosen pace allows. A comfortable pace means at most 4 a day, not counting meals, airports or the hotel.
- Item 18: a stop falls outside the venue's official opening days and hours for its scheduled day and time.
- Item 19: a day's order adds a train ride it did not need; a walkable loop does not count.
- Item 20: the same named place appears on two days.
A planner that produces no day-by-day schedule is "not scored" on items 16 to 20, and its total says so.
Try free
Plan a trip in about 2 minutes
Tap through 8 quick cards for a complete itinerary with budget, maps, and packing lists. No account required.
Plan a trip →Eleven AI trip planners on one Tokyo brief
Eleven planners received the same brief, six on October 19, 2026, three on October 21 and two on October 22: 7 days in Tokyo from November 16 to 22, a couple, USD 1,500 excluding flights, a comfortable pace, a first visit to Japan. ChatGPT ran logged out. Gemini ran in its 3.6 Flash mode after its default Thinking mode returned an error twice. Mindtrip ran signed in, because it now asks for an account before it plans. Layla blocks visitors from India, so it ran through a German VPN; it confirmed the brief before planning and added flights from Frankfurt that the brief excluded. Airial ran signed in with Google, needed two taps before it would plan, and added flights from Bengaluru. Each plan was scored on all 20 items from its own text, and each closed-stop verdict was checked against the venue's official site.
| Planner | Must-dos left (of 15) | Day faults (of 5) | Oh-No Travel Index |
|---|---|---|---|
| Airial | 12 | 1: lunch at a soba shop on its closing day | 13 |
| ChatGPT | 8 | 1: 3 stops on arrival day | 9 |
| Gemini | 6 | 2: 5 stops on day 3, the Unicorn Gundam statue after its display ended | 8 |
| iMean | 4 | 1: 5 stops on day 3 | 5 |
| Layla | 12 | 0 | 12 |
| Mindtrip | 12 | 1: a day that doubles back to Shinjuku for dinner | 13 |
| MonkeyEatingMango | 2 | 2: Tsukiji Outer Market after the stalls close, a doubled-back last day | 4 |
| Plantrip | 6 | 5: 3 stops on arrival day, 5 on its busiest days, Shinjuku Gyoen on a Monday and Miraikan during its refurbishment, a doubled-back day, an observation deck twice | 11 |
| Stardrift | 9 | 1: Tsukiji Outer Market on a Sunday | 10 |
| Stippl | 11 | Not scored: no day plan | 11, plus 5 not scored |
| Wanderlog | 7 | 1: 5 named stops on day 3 | 8 |
The table is alphabetical, not ranked. MonkeyEatingMango is the product behind this index, so read its row with that in mind: its two unanswered must-dos (the JR Pass and Monday closures) and both of its day faults are listed above. Each linked review has the full per-item comparison with MonkeyEatingMango.
A re-check of every plan at one depth on October 22, 2026 moved one score: Gemini from 7 to 8, because the Unicorn Gundam statue it scheduled ended its display on August 31, 2026. These are single runs of one brief. A planner's score can change from run to run and from version to version, so the date belongs with every score.
Where do the 20 items come from?
The items come from what travelers report going wrong and what plans measurably get wrong. A review of 115 Reddit threads in October 2026, across 8 travel subreddits, gave the must-dos and two of the day faults: too much in one day was the most common planning complaint, and entry papers and the leaving day were the costliest mistakes in threads asking for travelers' biggest mistake. The last three day faults are the most common in 356 AI-generated itineraries covering 2,735 planned days, where 43% of days carried at least one verifiable fault before correction.
Budget is left out on purpose. Whether a trip fits a budget depends on hotel prices for the dates, which most planners do not have, so cost is compared separately on each review page.
To see how a plan scores before you travel, plan a trip on MonkeyEatingMango and check it against the 20 items yourself.
Disclosure: MonkeyEatingMango, the product behind this blog, is one of the planners scored above.
Last updated October 2026.
How to cite
“The Oh-No Travel Index scores a trip plan from 0 to 20, lower is better: one point for each of 15 must-dos the plan leaves unanswered and each of 5 faults it makes on the day.” with a link to monkeyeatingmango.com/research/oh-no-travel-index/. Data and charts are licensed CC BY 4.0: free to republish with attribution.
Written by Shobhit Shrivastava · Founder
Related reading
Google Trip Planner (2026): The 5 Tools Explained
Google has had no single trip planner since Google Trips closed in 2019. AI Mode in Search, Gemini, Maps lists, Google Flights and Calendar each cover a piece; Gemini scored 8 of 20 on a 7-day Tokyo test.
Stardrift AI Review 2026: Is It Free, and Is It Any Good?
Stardrift turned one message into a dated 7-day Tokyo plan in about 90 seconds, with real hotel prices and an honest budget warning. It left the practical part of the trip to you.
Plantrip AI Review 2026: Is It Free, and Is It Accurate?
Plantrip turned one typed brief into a full 7-day Tokyo plan in 92 seconds, with the clearest train directions of six planners tested. It also packed the days and scheduled a garden on the day it closes.