Know where you can go. Discover where you could.
A global mobility platform for PL Group. Not another passport ranking — the product your own tagline already promises.
PL Group describes itself, on its own homepage, as “Crafting Your Borderless Legacy.” When we went looking for a name, we found yours waiting.
That is not a coincidence worth glossing over. It means this product is not a new direction for PL Group. It is the missing artefact of a positioning you have already committed to in public — the thing a prospect can actually touch.
You also already published the argument. Passport Legacy’s own article on second-passport cost comparison makes the case that headline figures exclude fees and that published comparisons mislead. Your words: price is “the easiest variable to compare and the least predictive of outcome.”
Borderless turns that article into a product that updates itself.
Before writing a line of this proposal we took passportindex.org apart — the markup, the network waterfall, the tracking, the lead endpoints. Three things are worth your time.
First, ownership. Passport Index belongs to Arton Capital. The Henley Passport Index belongs to Henley & Partners. Both are your direct competitors, and both run their index as a lead engine. That is the entire point of them. In a category where every serious peer operates one, PL Group is the conspicuous absence.
Second, the plumbing is visible. The button that appears when a visitor adds a second passport is labelled “MAKE IT YOURS.” and points at a path literally named /leads/, tagged utm_source=pi_improve. The footer line “BECOME A GLOBAL CITIZEN.” resolves to Arton’s high-net-worth landing page. The newsletter builds Arton’s list. Their qualification step is an off-the-shelf Typeform titled, verbatim, “Passport Index > Arton Capital — Antigua & Barbuda.”
Third, and this is the opening: they monetise exactly five countries through it — Antigua, St Lucia, St Kitts, Dominica, Grenada. You carry twenty-one.
Arton already ships second-passport simulation at /improve.php. “Add a passport and watch your score rise” is table stakes, not an invention, and we will not pitch it to you as one. What their version does is return a bare number in a box — no map, no list of which countries were unlocked, no price, no timeline, no programme. It tells a prospect their score improved and then abandons them at the exact moment they are interested.
We measured passportindex.org live on 3 September 2026, in Chrome, on a warm cache. The load event fired at 17,983 milliseconds. That is not bandwidth — every file was already on the device. It is 1.86 megabytes of JavaScript blocking the main thread. While we were measuring, Chrome’s own developer protocol timed out twice waiting 45 seconds for the page to respond.
The quality problems go deeper than speed, and they matter because they are the credibility of a data product:
#. All 199 country pages exist — they are buried in hidden markup. Arton built the SEO and then hid it.This is the machine PL Group is being asked to compete with. It is beatable.
Rather than estimate the performance we could reach, we built it and measured it. One HTML file containing a 177-country inline SVG world map, every one of the 39,601 visa-matrix relationships, all styling and all interactivity.
The architectural insight is embarrassingly simple: countries do not move. Arton recomputes the world map in the browser, with jQuery, on every single visit. We compute it once at build time and ship the answer. The entire global visa dataset — all 39,601 country-to-country relationships — compresses to 2,398 bytes. That is smaller than one passport thumbnail on their site.
We tested MapLibre GL, the industry-standard mapping engine, and rejected it: 231 KB before a single tile loads, fifteen times the weight of our entire map, to solve a problem a passport index does not have.
Your markets are Lagos, Abuja, Cairo, Beirut, Manila and Kuala Lumpur — mobile networks where a fifteen-second page is an abandoned page. And Google throttles crawl budget on slow hosts, which means Arton is arithmetically unable to add the 45,000 pages this category supports. Their stack is their ceiling. That is why this window is open now.
Every one of the eleven passport indices in market models a single traveller. Your own headline metric is “2,000+ families.” Grenada is priced at $235,000 for a family of four. Nobody has ever shown a family its shared visa-free world — or named the member who shrinks it.
Every index counts where you may visit. None maps where you may live, work, study or start a company by right. A French passport is a different instrument once EU freedom of movement is on the map.
Arton’s robots.txt blocks GPTBot, ClaudeBot, CCBot, Google-Extended and four more, and declares ai-train=no. The most-cited passport tool on earth has switched itself off from every AI assistant. Borderless switches itself on.
199 passports × 227 destinations is 45,173 possible pages. Arton has built 398. Henley’s entire Passport Index is eight URLs in a 3,261-page sitemap. Nobody has built a single page for “can a Nigerian travel to Türkiye visa-free” — among the highest-volume questions in the market.
On that last point, the demand is measurable rather than assumed. Wikipedia’s 216 “Visa requirements for [nationality] citizens” articles took 7.37 million views in twelve months; its 189 “Visa policy of [country]” articles took another 3.87 million. That is 937,000 views a month, on one site, for exactly the two page types Borderless generates.
69% of that demand sits outside the top ten passports — in Nigeria, Egypt, Lebanon, Türkiye and the Philippines. Those are your office markets. Henley has not built a single page for any of them.
Your framing is the right one, and it becomes the spine of the build:
“What can I do with my citizenship today — and how can I improve my global access?”
That is not a ranking question. It is a delta question, and a delta is computable. Everything below serves it.
One field. Underneath it, My Global Access: 187 countries. Below that, Ask PL. No dashboard, no wall of forms — the profile builds itself as the user answers questions they wanted to ask anyway.
Passports and residencies and visas — the combination no index models. Travel access, live and work rights, business access, family mobility, and a single Global Optionality Score.
Passport, destination, date, purpose, length of stay. Answered with permitted stay, validity rule, onward-ticket and eTA requirements, transit rules, and a deep link to the government page we checked.
Not “+17 destinations.” The set difference: which seventeen, named, on a map — plus E-2 eligibility, family coverage, investment, timeline and physical-presence rules.
Eight questions, ranked matches, and the reason for each. Every recommendation also names the one criterion it fails, in red, unprompted.
Live, work, study, establish a company, obtain residence. EU freedom of movement, CARICOM, Mercosur, GCC, the Trans-Tasman arrangement, US treaty visas — encoded as rights, not visits.
The map shades by how many of the family can go — 0 of 4 through 4 of 4. Solid gold is the family’s actual shared world. Dependant age rules included.
Your own site already runs globe.gl, twice. We extend the visual language you built — then add the move that matters: tap “add a passport” and watch the world recolour.
Personalised to what the user holds, not generic winners and losers. This is also what earns the app its place in the App Store.
Every “no” becomes a programme moment. When Can I Travel? returns visa required, embassy appointment, three to four weeks, the next line is …and with Grenada, you could go tomorrow. That single sentence converts a utility into a sales instrument, and it is the whole commercial thesis.
Open with Palau. A Palauan passport ranks mediocre on every visa-free index on the internet. Under the Compact of Free Association, a citizen of Palau, the Marshall Islands or Micronesia may live, work and study in the United States indefinitely. Not visit — live. No index on earth shows that, because none of them model rights. It is the single clearest demonstration of why your product is different.
Gold means “yours,” never “ours.” A hard rule in the design system: #E3AB50 is reserved for the user’s own data — their number, their countries, their family. Every commercial surface renders quieter than the content around it. Nobody will consciously notice. Everybody will feel it.
You asked the most important question in the engagement, and your instinct on Timatic and Sherpa is right. But there is a conflict inside the obvious architecture that has to be designed around now rather than discovered later.
Sherpa ships a one-hour cache header and asks you to respect it, because the data refreshes hourly and anything staler risks being wrong. That is an accuracy constraint rather than a licensing one, but it has the same effect: if the app calls a live API on every query, you cannot ship fast static pages, you cannot win the 45,000-page search opportunity, and your costs scale with traffic. And if instead you extract the corpus to build your own standing index, you are relying on a permission neither supplier publishes — which you spotted yourself: API access does not give you the right to republish their dataset as your own index.
One data source cannot do both jobs. So we split them.
On top of that sit the mechanics you described, which are the right ones: a nightly sweep of roughly 250 government sources, an AI monitoring layer reading announcements and news, a diff engine that clusters changed cells into human-sized policy events, and a review queue where a named PL analyst clicks approve or reject. Nothing publishes without that click — and we enforce it as a database constraint, not a policy. Every record carries a last-verified date and a link to the government page it was checked against.
The open passport dataset on GitHub that every developer reaches for is scraped from passportindex.org — Arton Capital. Its own README names the source. It was archived in January 2025 and the replacement has been updated once, in February 2026. We are not building your flagship on a stale copy of your competitor’s website, and we would rather tell you that now than have it surface later.
Henley state in their own press material that their index is built on exclusive Timatic data from IATA, enhanced by their own research department; IATA has never commented publicly, so the commercial shape is unknown. Timatic is plainly not airline-only either — IATA’s own customer wall carries a visa-services group, two corporate travel managers, a cruise line, a ground handler and a national border force. You can buy a single Timatic seat online today. What nobody publishes a price for, anywhere, is the right to republish the data as your own index, and IATA closed its public API sandbox in June 2026. That is an enterprise negotiation, and our planning assumption is that it runs eight to sixteen weeks. That negotiation, not the engineering, is the critical path on this question, so it should open in week one through IATA’s Dubai office, on your doorstep.
Sherpa is the practical day-one path: a genuinely developer-friendly API, and — usefully — it monetises through commission on eVisa sales, so a revenue share is negotiable and the feed can partly pay for itself. It also ships an endpoint built for exactly your AI concierge.
Be aware: the data licence is an ongoing annual cost and is likely to be the largest running line in the product. We are not going to invent a number for you. The band to plan against for full commercial redistribution rights is meaningful, and we will bring you real quotes before you commit to anything.
We build once and ship three ways: the website, an installable home-screen app, and real iOS and Android binaries. One codebase. You are not paying for the product twice.
The rule that catches people is Apple’s Guideline 4.2, Minimum Functionality: Apple rejects apps that are “a repackaged website.” The store build has to earn its place with genuinely native behaviour — offline access to the visa matrix, real push notifications, Face ID over saved documents. Your Mobility Alerts feature is precisely what satisfies that, so we build it deliberately rather than discovering the problem at submission.
Three practical notes:
Store fees are trivial — Apple is USD 99 a year, Google a one-off USD 25. And we will put the app on your phone through TestFlight weeks before any public launch.
Ship the web product first and make it exceptional. The stores give you credibility standing next to Arton and a push channel to people who have already raised their hand — but it is the fast, indexable web experience that wins the search traffic, and search traffic is where the leads are. Spend in that order.
Borderless inherits PL Group’s identity exactly — these are the literal custom properties compiled into your own sites, not colours picked from a screenshot.
Fraunces for display, Manrope for body — your existing pairing, carried through. This document is set in them, so you are reading the product’s typography now.
Phase 1 delivers a written style guide covering the token set, the type scale, map and data-visualisation colour rules including the colour-blind-safe visa-status ramp, component patterns, and the gold-means-yours rule. It is a document your team and any future supplier can build against.
Arton’s own About page concedes that “passport designs are copyright of each respective government.” Their 199 cover photographs are served as 207 separate image requests. Reproducing a UK cover bearing the Royal Arms engages Crown copyright under s.163 of the Copyright, Designs and Patents Act 1988. We will draw all 199 covers as original vector illustrations. One decision removes four separate legal exposures and replaces 207 image requests with a single 96 KB sprite.
Straight answer first: your brief has grown considerably since the number was set, and we would rather show you the real shape of it than quietly deliver a fraction and call it done.
Fifteen features, an AI concierge, a legal rights matrix spanning thirty-plus treaty frameworks and native apps is a multi-phase programme. So the price below is not a quote for all of it. USD 4,690 R75,000 buys Phase 1 — a complete, shippable, revenue-capable product on its own — and everything after it is yours to take or leave.
llms.txt for AI assistantsThe freshness engine needs a human. Three tiers: Essential at USD 1,376 R22,000 (pipeline, monitoring, your analyst reviews), Managed at USD 2,814 R45,000 (our trained analyst does the reviews, sixteen hours a month, monthly accuracy report), Managed Plus at USD 4,253 R68,000. Roughly twelve to eighteen hours of review a month once it settles; closer to a day a week for the first three.
At today’s rate — USD/ZAR 15.99 on 3 September 2026 — R75,000 is USD 4,690, or AED 17,226. For context that we think matters more than our own view of it:
| Benchmark | Range | vs Phase 1 |
|---|---|---|
| Dubai agency — custom, non-templated website | AED 18,000 – 145,000 | Below their floor |
| Dubai agency — custom web application | AED 50,000 – 275,000 | Below their floor |
| South African market — complex web application | USD 6,254 – 28,143 R100,000 – 450,000 | Below the floor |
| Agent commission, one approved St Kitts file | ~USD 50,000 | 9% |
| Qualified RCBI lead via Google Ads | ~USD 5,000 each | One lead |
| One early-bird seat, IMI Connect Buenos Aires | EUR 2,499 | 1.6 tickets |
Phase 1 costs less than one qualified lead bought on Google, and roughly 9% of the commission on a single approved St Kitts application. It breaks even on less than half of one file.
There is no VAT on this invoice. Lumanaire is not a VAT vendor, so no VAT is charged to any client anywhere. USD 4,690 R75,000 is the total, and nothing is added to it. A Dubai agency would add 5%.
Payment: 50% to commence, 50% on delivery. Phase pricing beyond Phase 1 is indicative and will be fixed on scope sign-off. Data licence fees are contracted by PL Group directly and passed through at cost — we take no margin on them.
| Risk | Our position |
|---|---|
| Trademark on the name | Lower than assumed. Arton has no registered US trademark for “Passport Index” — two applications filed and abandoned, the second in April 2024. Their ® attaches to “Empowering Global Citizenship”. We will still clear the final name formally before launch. |
| Data licensing | The real constraint, and the reason for the two-layer design. Timatic runs 8–16 weeks to signature, so it opens in week one and the build does not wait on it. |
| Passport cover copyright | Resolved by drawing all 199 as original vectors. Also removes 206 of the 207 image requests. |
| Accuracy and liability | No operator guarantees this data, including IATA — every one of them disclaims. We make the disclaimer work for you: a last-verified date and the government source on every record, and every uncertain cell becomes a reason to speak to a PL adviser. |
| Scope | The brief has grown 3× since the price was set. Phasing is how we keep that honest rather than delivering a thinner version of everything. |
| Life after handover | The common failure in agency-built products. The admin console and the retainer exist specifically to prevent it. |
One further note on operations, because it reframes a cost as an asset. You asked for something “easy to operate,” and you were thinking about maintenance. Consider instead that you have seventy advisors across ten offices handling real border cases for real clients every day. That is a live visa-data sensing network Arton cannot replicate at any price. The admin console is what turns it into a data advantage.
Developer accounts opened in PL Group’s name — the longest-lead item. First contact with IATA’s Dubai office and with Sherpa. Style guide and design direction agreed.
The PL-owned dataset assembled and verified. The 199 passport covers drawn. Core architecture and the passport explorer built.
Portfolio Builder and the set-difference engine, wired to all 21 programmes. First review with you, on a working product.
Best Second Passport quiz, lead capture, CRM routing and office assignment.
Performance hardening against the contracted targets, SEO and structured data, launch at borderless.plgroup.app. Style guide delivered.
Sign-off on Phase 1 scope and the name. A nominated data owner in your team. Access to your brand assets. An introduction to whoever runs your CRM. And a decision on who opens the IATA conversation — it carries more weight from PL Group than from us.
Arton have had twelve years and a fifteen-second head start. Everything above is buildable, and the window is open now.