Daniel's been wrestling with the phone number portability problem from both ends. Last year he switched carriers in Israel, lost his old number in the process, and hit a brick wall trying to update things like Israel Post because the authentication system wanted to text the number he no longer had. Now he's setting up a Twilio system and discovered he can't port Israeli numbers onto it either. His questions are threefold. Why do these artificial restrictions exist? Why specifically can't Israeli numbers port to Twilio? And is there a middle ground — some provider that gives you a local number that actually works plus the SIP flexibility to configure your own routing?
The Israel Post detail is the part that sticks with me. It's not just an inconvenience — it's a deadlock. The system is designed around the assumption that your phone number is a permanent anchor, and when that anchor gets pulled up, there's no fallback. You're just locked out.
Permanently, in his case. He never resolved it.
Right. And that's the human cost of what looks on paper like a dry regulatory question. So let's start with what number portability actually is, because I think most people experience it as this consumer right that just... exists. You switch carriers, you keep your number. Done.
And then you try to port to something that isn't a carrier and suddenly the magic evaporates.
Number portability is the ability to take your phone number with you when you change service providers. It rolled out globally starting in the late nineties — Hong Kong was first in ninety-nine, the UK followed, the US got it in two thousand three. And the pitch was always about competition. If you can keep your number, you're more likely to switch to a cheaper or better carrier, which forces carriers to compete on price and service instead of just... holding your number hostage.
Which is what they'd been doing.
Which is exactly what they'd been doing. Before portability, switching carriers meant losing your number, and that meant updating everyone you knew, losing business contacts, reprinting anything with your number on it. The switching cost was enormous. Carriers loved it.
So portability was a consumer protection intervention. The regulator stepped in and said, no, the number belongs to the customer now, not the carrier.
That's the idea. But here's the thing that most people don't realize — portability is not a technical feature of the phone network. It's a regulatory construct. The phone system was built with the assumption that number ranges belonged to specific carriers. You could route a call to a number by looking at the prefix and knowing exactly which carrier owned that block. Portability breaks that assumption, and it requires a whole new layer of infrastructure to fix it.
The central database.
The central database. In the US it's called the NPAC — Number Portability Administration Center. Every ported number gets an entry that says, this number used to belong to Verizon but now it lives at T-Mobile, route calls accordingly. When you dial a number, the network checks the database, finds the current carrier, and sends the call there. Without that database, portability doesn't work. And building and maintaining that database is expensive and complicated, which is why carriers fought it for years.
So the technical complexity was real, but it was also a convenient shield.
The technical argument and the economic argument are welded together in these fights. Carriers genuinely had to rebuild their routing systems. They also didn't want to lose customers. Both things are true.
And that brings us to Israel. Domestic porting between carriers is mandatory — you can switch from Cellcom to Partner and keep your number. That works. The Ministry of Communications regulates it. So what breaks when you try to port to Twilio?
The classification breaks. The Ministry of Communications mandates portability between licensed carriers. Mobile carriers, fixed-line carriers — they all have licenses, they all participate in the porting system. Twilio is not a licensed carrier in Israel. It's classified as a service provider. It doesn't own network infrastructure in the country. It buys numbers from carriers and resells them, or it provides programmable telephony services that sit on top of the carrier network.
So the porting rules simply don't apply to Twilio because the rules were written for a world where only carriers have phone numbers.
That's the regulatory gap. It's not that Israel explicitly banned porting to Twilio. It's that the porting framework was designed for carrier-to-carrier transfers, and nobody wrote the rules for carrier-to-platform transfers. Twilio falls into a category that the regulation doesn't address.
Which is a kind of ban by omission.
Functionally, yes. Twilio's own porting documentation lists Israel as a country where porting is not supported due to local regulatory restrictions. The documentation is blunt about it — they just can't do it. And this isn't unique to Israel. There's a global pattern here. Many countries exclude VoIP providers from porting regimes because those providers aren't licensed carriers in the traditional sense. The regulatory framework was built for a world of physical infrastructure and licensed operators, and it hasn't caught up to software-based telephony.
So the restriction isn't technical. It's not that Twilio's systems can't handle a ported number. It's that the regulatory plumbing doesn't connect Twilio to the porting database.
Right. Technically, there's no reason a VoIP provider couldn't participate in number portability. The NPAC-style database just needs to route the number to the right destination, and that destination could be a SIP endpoint as easily as a carrier switch. The issue is that the regulator doesn't recognize Twilio as a participant in the porting system, so there's no mechanism for Twilio to receive a ported number even if they wanted to.
And the carriers have no incentive to push for this. They're the ones who benefit from the gap.
Of course. Every number that can't port to a VoIP provider is a number that stays with a carrier. The carriers don't want to open the porting system to competitors who aren't even paying for spectrum licenses and network infrastructure. From their perspective, Twilio is getting a free ride on a network they didn't build, and porting would make it easier for customers to leave.
Let's sit with the technical mechanism for a second, because I think it explains why the regulatory gap is so stubborn. Walk me through what happens when a number ports in a system that works.
Okay. Imagine you've got a number, oh five four dash something, that originally belonged to Partner. When someone dials that number, the network looks at the prefix and says, Partner owns this block, route it to Partner's switch. That's the old model. After portability, there's a database lookup first. The network says, let me check — nope, this number moved to Cellcom last month, route it there instead. The database is the source of truth, not the prefix.
And the database is maintained by...?
In Israel, it's administered under the Ministry of Communications, with the carriers participating in a shared system. When you port your number, the losing carrier and the winning carrier coordinate through that system to update the routing entry. The whole thing is designed for two licensed carriers to talk to each other.
So if Twilio wanted to receive a port, they'd need to be a recognized participant in that system. They'd need credentials, a routing entry, a point of interconnection.
And they'd need the regulator to say, yes, you count. You're allowed to play. That's the step that hasn't happened. And it's not just about filling out a form. The regulator would need to answer a bunch of questions they haven't had to answer before. What obligations does a VoIP provider have for emergency calling? For lawful intercept? For number management? These are real regulatory questions, and nobody's in a hurry to answer them just so a few tech-savvy users can route their calls through Twilio.
So the catch-twenty-two for someone like Daniel is: you can't port your Israeli number to Twilio, so if you want to keep your number you're stuck with a carrier. But if you want the programmability of Twilio — the SIP flexibility, the custom routing, the API access — you need a new number, which defeats the whole point of portability.
And the Israel Post situation is the nightmare version of that tradeoff. He lost his old number during the carrier switch, couldn't port it, and then couldn't update his records because the authentication system was tied to the number he'd lost. It's a cascade failure. Each system assumes the previous system worked perfectly, and when one link breaks, the whole chain falls apart.
The authentication deadlock is worth pausing on, because it's not just a portability problem. It's a design problem. The Israel Post system apparently had no alternative authentication path — no email fallback, no identity verification in person, no recovery code. Just SMS or nothing.
And that's not unique to Israel Post. How many services have you signed up for where the only second factor is your phone number? Banks, government services, payment platforms. The phone number has become this accidental identity layer, and nobody designed it to be that. It just... happened.
Because the number was stable. For decades, your phone number was the most durable piece of personal data you had. Addresses change, email addresses come and go, but your phone number? That thing was welded to you.
And portability was supposed to make it even more durable. The promise was, you can keep this number forever, across any carrier. But the promise only extends to the edge of the carrier system. Step outside that system — to a VoIP provider, to a platform like Twilio — and the durability evaporates.
So let's talk about the middle ground. Daniel's third question. Are there providers in Israel that give you a local number that actually works for things like SMS authentication, plus the SIP flexibility to do your own routing?
There are, but they come with tradeoffs. The key players are providers like Bezeq International and zero one two Telecom. These are licensed carriers — or they operate through licensed carrier infrastructure — so they can issue local Israeli numbers that work for inbound and outbound calling, and for SMS. And they offer SIP trunking, which means you can connect your own PBX or your own Twilio-style routing setup to their network.
So you get a real Israeli number, and you get SIP access. That sounds like exactly what Daniel's asking for.
It's close. The catch is that you're getting a new number, not porting your existing one. These providers can issue you a fresh local number and give you SIP credentials to register your softphone or your PBX. But if you want to bring your old number from Cellcom or Partner, you're back in the porting problem. Some of these providers do support porting between carriers — you can port from Partner to Bezeq International, for example, because they're both in the licensed carrier system. But the port is to their carrier service, not to a pure VoIP platform like Twilio.
So the middle ground exists, but it's still carrier-tethered. You're not escaping the carrier system; you're just picking a carrier that also offers SIP.
The SIP offering is often designed for business customers. Bezeq International's SIP trunking is marketed as a business phone solution, not a consumer product. You might be able to get it as an individual, but you're paying business rates and dealing with business support channels. It's not the same as signing up for Twilio and getting API access in five minutes.
What about the mobile carriers themselves? Partner, Cellcom — do any of them offer SIP access on a consumer or small business plan?
Some do, but it's inconsistent. Partner has offered SIP trunking for business customers. Cellcom has had business VoIP products. The landscape shifts — these products come and go depending on market demand. But the common thread is that they're not designed for the use case Daniel's describing. They're designed for a company that wants to replace its office PBX, not for an individual who wants programmable call routing with a local number.
The pricing reflects that.
The pricing absolutely reflects that. You're paying for a business service, with per-channel costs and minimum commitments. It's not the pay-as-you-go model that makes Twilio attractive for personal projects.
The realistic options for someone in Daniel's position are: one, accept the tradeoff and get a new number from a SIP-friendly provider, then update all your contacts. Painful but workable. Two, keep your carrier number for inbound and SMS, and use a separate SIP number for outbound calling with custom routing — a split setup.
The split setup is actually what I'd recommend for most people in this situation. Keep your carrier number for the things that need it — two-factor authentication, services that have your number on file, anything where changing the number creates a cascade of problems. Then get a separate SIP number for the programmable stuff. Forward calls from the carrier number to the SIP number if you want a single point of contact, or just give out the SIP number for new contacts and let the old number fade over time.
That's a migration strategy, not a solution. It works, but it's a workaround.
It's absolutely a workaround. And it's a workaround that shouldn't be necessary if the regulatory framework were updated to reflect how telephony actually works in twenty twenty-six.
Let's zoom out for a second, because this Israeli situation is really a case study in a global pattern. The regulatory definition of a carrier was written for a world of physical switches and licensed spectrum. VoIP providers don't fit that definition, so they're excluded from portability regimes by default. It's not malice — it's a category error that nobody's bothered to fix.
The category error has real consequences. It fragments the telephony landscape. You've got the carrier world, where numbers are portable between carriers but only within the licensed system. And you've got the VoIP world, where numbers are programmable and flexible but can't be ported from the carrier world. The two systems don't talk to each other, and the customer is stuck in the middle.
The promise of number portability was that your number is yours, permanently, across any provider. The reality is that your number is yours only as long as you stay within the licensed carrier system. Step outside and you lose it.
The knock-on effect is that this tilts the playing field toward incumbents. If you're a startup trying to build a new kind of phone service, you can't offer number porting because you're not a carrier. So customers who want to try your service have to give up their number, which almost nobody will do. The incumbents get a built-in switching cost that regulation was supposed to eliminate.
It's a moat. An accidental regulatory moat, but a moat nonetheless.
Moats are hard to drain once they're in place, because the people inside the moat have no incentive to help you drain it, and the regulator has other priorities.
What would it actually take to fix this? If the Ministry of Communications wanted to allow porting to VoIP providers, what would they need to do?
They'd need to create a new category of license — something like a virtual carrier or a service-based operator — that comes with a defined set of obligations. Emergency calling access, lawful intercept capability, number management rules, interconnection requirements. Once that category exists, qualified VoIP providers could apply for it, and once they're licensed, they could participate in the porting system.
It's not technically hard. It's a paperwork problem.
It's a policy problem. The technical infrastructure exists — the porting database can route to a SIP endpoint as easily as to a carrier switch. The challenge is defining the regulatory framework and getting the stakeholders to agree on it. And the carriers will fight it, because every number that ports to a VoIP provider is revenue they lose.
Has any country done this well? Created a VoIP-friendly portability framework?
The US is the closest example. The FCC classifies interconnected VoIP providers as telecommunications carriers for certain purposes, which means they can participate in number portability. Twilio supports porting in the US precisely because the regulatory framework allows it. It's not perfect — there are still restrictions and the process can be slow — but the basic mechanism exists.
The US said, if you connect to the public phone network, you're a carrier for portability purposes, even if you don't own physical infrastructure.
Right. It's a functional definition rather than an infrastructure-based one. And that's the shift that most countries haven't made. They're still using the old definition, where a carrier is someone who owns switches and spectrum. If you're just writing software that talks to the network, you don't count.
Israel is firmly in the old-definition camp.
Firmly. The Ministry of Communications has shown no interest in creating a virtual carrier category. The focus has been on managing competition among the existing licensed carriers, not on opening the market to new kinds of providers.
Which brings us to the practical question Daniel's really asking. Given that the regulatory situation isn't changing anytime soon, what's the best move for someone who wants a local Israeli number with SIP flexibility?
I'd say the best move is to accept the split. Get a new local number from a SIP-friendly provider — Bezeq International or zero one two Telecom if you want carrier-grade reliability, or explore whether any of the mobile carriers' business SIP products are accessible to you. Set up your Twilio-style routing on that number. Then keep your old carrier number active on the cheapest possible plan, forward it if you can, and gradually migrate your contacts and services to the new number.
It's not elegant.
It's not elegant at all. But it's the reality of a regulatory framework that hasn't caught up to the technology. You can have continuity or you can have programmability. You can't have both in Israel right now.
The one thing I'd add is that if you're going to do the split, do it deliberately. Don't let the old number linger as a zombie — make a plan to migrate everything within six months, then cancel it. Otherwise you're paying for two numbers forever out of inertia.
Document which services have your old number. That's the Israel Post lesson. Before you start the migration, go through your accounts and update everything you can think of. Banks, government services, two-factor authentication, WhatsApp, everything. The ones you forget are the ones that will lock you out later.
A migration checklist is the unglamorous but essential tool here.
The least exciting document you'll ever make, and the one you'll be most grateful for when something breaks.
Before we wrap up the main discussion, I want to name the misconception that I think drives a lot of frustration here. People assume number portability is a technical feature of the phone network — that your number is inherently portable, and carriers just need to flip a switch. The reality is that portability is a regulatory achievement, hard-won and incomplete. Every port that works is a small miracle of policy coordination. Every port that doesn't work is a reminder that the system was designed for a different era.
The specific misconception about Twilio and VoIP is that the restriction is technical. People assume there's some incompatibility between VoIP and ported numbers. There isn't. It's purely a classification issue. If the regulator said tomorrow that Twilio counts as a carrier for portability purposes, the technical side would be solved in weeks.
The bottleneck is policy, not engineering.
As it so often is.
Hilbert's been listening to all of this with what I assume is a growing list of objections.
Hilbert: Not objections exactly. More like... a different framing. You've both been describing these restrictions as a regulatory failure — a gap that needs closing, a framework that hasn't caught up. And I think that's half the story. The other half is that some of these restrictions exist for reasons that are actually worth taking seriously.
Go on.
Hilbert: If anyone can port a number to any VoIP provider, how does a nine one one operator know where the caller is? The carrier system ties numbers to physical addresses and cell tower locations. VoIP breaks that link. A number that looks like it's in Tel Aviv could be routing through a server in Frankfurt with the caller sitting in a coffee shop in Tokyo. That's not a theoretical problem — it's a real public safety issue that regulators are worried about.
That's fair. But it's also solvable. VoIP providers can register physical addresses for emergency calling. The US requires it.
Hilbert: They can, but enforcement is patchy and compliance is inconsistent. And emergency services isn't the only concern. Lawful intercept is another one. Carriers are required to provide wiretap capabilities to law enforcement with a warrant. VoIP providers operating across borders make that much harder to enforce. If you're the Ministry of Communications, and you're weighing consumer flexibility against network integrity and public safety, it's not obvious that opening the floodgates is the right call.
I don't think anyone's arguing for opening the floodgates with no rules. The argument is for creating a regulatory framework that allows qualified VoIP providers to participate under defined obligations. Emergency calling, lawful intercept, number management — all of that can be required as conditions of the license.
Hilbert: In theory, yes. In practice, you're asking a small regulatory agency to monitor and enforce compliance across a much more complex landscape. Carriers are big, stable, and few in number. VoIP providers are small, numerous, and some of them are just a guy with a server. The regulatory capacity problem is real.
That's a stronger version of the argument. It's not that the restrictions are good in principle — it's that the regulator doesn't have the bandwidth to manage a more open system safely.
Hilbert: That's a legitimate reason for caution. You can call it inertia, but it's also prudence. The current system works for most people most of the time. Opening it up creates new risks that someone has to manage, and the Ministry of Communications isn't exactly overstaffed.
I'll concede that the regulatory capacity argument has weight. But I'll push back on the idea that the current system works for most people. It works for people who never leave the carrier system. For anyone who wants to do anything interesting with telephony — build custom routing, integrate with software, escape the carrier pricing model — the system is a wall. And the number of people hitting that wall is growing.
Hilbert: It's growing, but it's still a small fraction of the population. The regulator's job is to protect the broad public interest, not to optimize for the edge cases.
I think that's where we land, then. The restrictions aren't purely about carrier greed, and they're not purely about public safety. They're a messy compromise between competing interests, and the compromise currently favors stability over flexibility. Whether that balance is right depends on whether you think the future of telephony looks more like the carrier model or more like the software model.
I think the future looks more like software. Which means the pressure on regulators will grow, even if change is slow.
Where does that leave us? The number you keep is a relic of a carrier-centric world. It's portable within that world, but it's tethered to it. The future might be numbers that are just data — portable by design, not by regulatory mandate. But we're not there yet, and in the meantime, the practical advice is to plan around the limitations rather than waiting for them to disappear.
The misconception to leave with: number portability feels like a technical feature of the phone network, but it's actually a regulatory construct with sharp edges. It works between carriers because regulators forced it to. It doesn't work between carriers and VoIP providers because nobody's forced that yet. The limitation isn't in the technology — it's in the policy.
Thanks to Hilbert Flumingtop for producing and for making us sit with the public safety angle we were conveniently skating past.
This has been My Weird Prompts. If you've got a question like Daniel's — something that seems like it should be simple but turns out to be a regulatory rabbit hole — email the show at show at my weird prompts dot com.
We'll be back soon.
