121 or 97? Two Travel-Rule Trackers Counted the Same Chaos and Disagreed by a Quarter


Executive Summary: Two Numbers, Two Methods, One Unmeasured Interval

Between 30 September and 19 October 2026, two commercial travel-rule trackers published counts of how many entry requirements changed, and the two numbers do not reconcile. EntryCheck's second issue of its Digital Borders Report, published 30 September 2026, states in its own headline that it counted 121 entry-rule changes in the third quarter of 2026. [source-1] Sixteen days later, on 19 October, iVisa announced a free Global Policy Tracker and stated that its team had documented and verified at least 97 visa, electronic travel authorisation and arrival-card changes across 61 countries since October 2025. [source-2]

The 24-change gap between the two figures is not a contradiction. The two trackers are measuring different intervals over different populations with different inclusion rules, and neither publishes a method detailed enough for a reader to reconstruct its count. [source-1] [source-2] That is precisely the problem. When entry rules change at the pace these two trackers describe, the traveller's practical question is not "how many rules changed" but "is the tracker I am reading counting the rules that will affect me", and a count without a published inclusion rule cannot answer that question. [source-3]

What can be established from primary sources is far more useful than either headline number. The European Union's own migration authority confirms that the Entry/Exit System went live on 12 October 2025 and reached deployment across 29 Schengen states in April 2026, and that nine member states asked Brussels to pause the rollout. [source-4] [source-2] Kazakhstan's state news agency states that the country will require an electronic travel authorisation from July 2027. [source-5] The United Kingdom's ETA system became mandatory for 85 nationalities in February 2026, and by peak summer travel an outage left some travellers facing delays above 60 hours, with carrier fines reaching £2,000 per passenger. [source-2]

1. The Two Counts, Stated Exactly

It matters that the two numbers are reported in the sources with their own qualifiers intact, because the qualifiers are what makes the comparison possible at all.

Tracker	Count	Interval	Population	Method published	Update cadence
EntryCheck	121	Q3 2026 (1 Jul - 30 Sep 2026)	Global digital entry systems, country-level	Summary only; no sample selection, classification or verification procedure disclosed	Quarterly
iVisa	97 (at least)	Oct 2025 - 19 Oct 2026	Visa, ETA and arrival-card changes, 61 countries	Cites official government or other verified sources for every entry	Weekly
Difference	24	Not comparable intervals	Not comparable populations	-	-

Sources: [source-1] [source-2]

Two features of the EntryCheck figure deserve emphasis. First, the report is described as measuring what moved inside the world's digital entry systems, and the published summary does not disclose its sampling frame, its classification method, or a per-item reference table for all 121 entries. [source-1] Second, the entries listed under the quarterly heading are not uniformly in-force rules. Several are explicitly proposals or intentions, described with the language of consideration rather than enactment, and several carry a source that is a media outlet rather than a government domain. [source-1] A count that mixes enacted rules with proposals measures activity in the announcement space, which is a legitimate thing to count but is not a count of rules that a border officer is applying.

The iVisa figure carries a different and more unusual qualifier: the tracker states outright that it is not a complete record of every travel policy change, and that coverage leans toward high-traffic routes and English-language sources. [source-2] That is a disclosure most trackers in this category do not make. It is also, on inspection, the more important number for a traveller, because it is the one that describes how the record was assembled. A tracker that says "here is what I checked and here is what I did not check" is more useful than a tracker that states a total without a method.

2. What the Count Conceals: A Calendar of Concrete, Dated Changes

The individual changes enumerated inside the EntryCheck report are far more actionable than the total, and several carry hard dates and named government sources. The sampling below is drawn from the entries published under the quarterly heading. [source-1]

Date	Authority	Change	Instrument
2026-09-01	India	International departure no longer stamped on boarding passes	Paperless exit
2026-09-02	India	e-Group Visit Visa for cruise operators, multiple entry, 30 days	eVisa
2026-09-06	EU	EES biometric temporary border-check grace period ends	EES
2026-09-07	UAE	Click and Travel mobile passport check at Dubai International	Mobile processing
2026-09-09	Mexico	Default 180-day visitor permit no longer granted; stay set by purpose of travel	Permission model
2026-09-15	Thailand	New entry rules shorten visa-free stay for many nationalities, restrict visa-on-arround eligibility	Visa waiver
2026-09-19	Thailand	Suvarnabhumi eGate opened to 33 visa-exempt nationalities	eGate
2026-09-19	Hong Kong	Go Hong Kong app allows eligible travellers to register for e-Channel before arrival	Arrival app
2026-09-19	Laos	eVisa approval letter validity temporarily extended from 60 to 180 days	eVisa
2026-09-21	China	Nationwide online registration extended to foreign nationals lodging outside hotels	Registration
2026-09-25	Japan	JESTA advanced; targeted for fiscal year 2028	Digital arrival
2026-10-03	EU	ETIAS expected to slip to 2027 while EES continues to cause airport delays	Deferred launch

Sources: [source-1] [source-4]

Three of these deserve a traveller's attention beyond the table. The Mexico change on 9 September 2026 inverts the assumption most visitors carry: the default 180-day permission is no longer the fallback, and the permitted stay is set by the stated purpose of travel instead. [source-1] The Laos change on 19 September is a temporary extension, from 60 days to 180, of eVisa approval letter validity, which is the kind of rule that changes silently and is absorbed into a booking cycle without anyone re-reading it. [source-1] The Hong Kong Go Hong Kong app and Thailand's 33-nationality eGate expansion are both convenience changes that expand access rather than restrict it, and they are the kind of entry that a tracker counting only "restrictions" would systematically under-report. [source-1]

3. The System-Level Failures Are the Part That Actually Bites

Individual rule changes are the visible layer. The more consequential pattern in the 2026 record is what happened when the new digital systems met real traffic, and here the primary sources are more specific than either tracker headline.

The United Kingdom's ETA became mandatory for 85 nationalities in February 2026. By peak summer travel, a system outage left some travellers facing delays of more than 60 hours, and carriers were fined up to £2,000 per passenger for boarding an ineligible passenger, according to UK Home Office figures. [source-2] That is a distinctive failure mode. It is not a rule being enforced wrongly. It is a rule being enforced correctly against a system that was not available, with the cost falling on the traveller in the form of carrier penalties.

The European Union's Entry/Exit System went live across 29 Schengen states in April 2026, and an easyJet flight left 122 passengers behind in Milan within days of the launch. By summer, some border crossings reported wait times of five hours, and nine member states asked Brussels to pause the rollout. Brussels declined, and the EU has since pushed its planned €20 ETIAS authorisation, originally set to launch alongside the Entry/Exit System in late 2026, back to sometime in 2027 without confirming a new date. [source-2]

Sri Lanka's pre-departure ETA became mandatory on 15 October 2025, was revoked on 30 October, and after a February 2026 system disruption delayed approval notices, Sri Lankan missions told travellers who had already applied that they could fly before receiving approval. The pre-departure ETA has been required again since 25 May 2026. [source-2] A requirement that is suspended and reinstated within a week is indistinguishable from a requirement that was never introduced, except for the traveller who planned around it.

Arrival cards are now a separate obligation from visas, and iVisa's tracker lists at least 16 countries that required one in the past year, including South Korea, Vietnam, Thailand, Malaysia, Taiwan, Cambodia and Papua New Guinea. [source-2] Each is a short form, separate from any visa or ETA, that travellers must complete in advance. This is the part of the record that most reliably generates confusion, because the form is often described in the same breath as a visa requirement, and the consequences of getting it wrong are boarding denials rather than a fine.

4. The Direction of Travel, and the Two Systems That Broke It

Independent measurement confirms the direction. The Henley Passport Index 2026 Global Mobility Report found that both the United States and the United Kingdom recorded their steepest annual losses in visa-free access on record, with the United States losing seven destinations and the United Kingdom losing eight in a single year, and over two decades the United States has fallen from 4th to 10th place while the United Kingdom has fallen from 3rd to 7th. [source-2]

A separate source of pressure is a proposed rule posted by the US Department of Homeland Security in December 2025, still pending review at the White House budget office, which would require citizens of all 42 Visa Waiver Program countries, including the United Kingdom, Germany, Japan and Australia, to disclose five years of social media history, the email address used over the past ten years, and the phone numbers and home addresses of immediate family members. [source-2] The status of that proposal matters to the reading of the other numbers. It is pending, not enacted, and a count that folds pending proposals into a quarterly total produces a figure that reads as more severe than the enforcement reality.

Congressional Research Service data cited in the iVisa tracker puts the number of countries facing a full or partial US travel ban at 39, up from roughly 19 in mid-2025. [source-2] The direction is consistent across the sources. What differs is the denominator.

The practical conclusion for a traveller is narrower than the headlines suggest and more useful. The volume of change is not the binding constraint on a single trip. The binding constraints are that a rule you rely on may have changed since your last trip, that the system enforcing it may be down when you arrive, and that the tracker you use to check may not have covered the country you are entering. The second and third are the ones a headline count cannot express.

5. What to Do Instead of Reading the Count

The practical sequence is short and does not require trusting any single tracker. Check the primary source for the destination, not a tracker's characterisation of it, because the tracker's own coverage disclaimer is the honest part of its product. [source-2] For the European Union, that is the official Entry/Exit System information and the ETIAS official site; for Kazakhstan, the state news service carries the July 2027 electronic travel authorisation timetable. [source-4] [source-5]

Second, separate the three change types that are routinely merged, because their consequences differ. A new arrival card is a pre-departure form. A fee change is a cost change. A new system is a reliability risk, and it is the only one of the three that can strand a traveller at a gate with a valid document. The 60-hour UK ETA outage and the 122 passengers left in Milan are both system failures, and neither would have been visible in a rule count. [source-2]

Third, treat a tracker as a search index rather than a register. The iVisa tracker is updated weekly, browsable by country and document type, and states that every entry cites an official or otherwise verified source. [source-2] The EntryCheck report is a quarterly document that is stronger on breadth of coverage and weaker on disclosed method, and its own list mixes enacted rules with proposals and carries a mix of government and media sourcing. [source-1] Neither is wrong. Using the one that matches the question being asked is the entire skill.

The residual risk after that sequence is the one that arrives without a schedule attached. A rule that changes while a traveller is already in transit, or a system that fails at the gate on the day of departure, is not something a pre-trip check can prevent, and it is the failure mode the 2026 record documents most concretely. [source-2] [source-7] [source-8] [source-9] The instruments that address it are unglamorous and should be arranged before departure rather than reasoned about afterwards: an eSIM data plan active before the trip keeps a rebooking conversation possible while a flight is held on the ground, flight compensation is what converts a delay caused by someone else's rule change into a recoverable expense rather than a cost, and a fixed-price airport transfer booked in advance removes the assumption that a ground journey costs a variable amount on the day it happens. [source-12] [source-13] [source-14]

6. Conclusion: The Gap Is the Finding

Two trackers counted the same turbulent period and produced 121 and 97. Reconciling them is not possible from the published material, because one reports a quarter with an undisclosed method and the other reports a year with a disclosed coverage limit. [source-1] [source-2] The twenty-four change difference is a measurement artefact and should not be read as a disagreement about what happened.

What survives the reconciliation is the set of dated, attributed changes in the table above, the record of two major digital entry systems failing under real passenger load in 2026, and the recognition that the direction of travel for the most powerful passports is downward on every independent measure available. [source-1] [source-2] None of that required a total.

A reader who takes only one thing from this record should take the date of the change they care about and the name of the authority that made it, and should treat every aggregate number in this field, including the ones in this article, as a description of a tracker rather than a description of the world.