Italy eSIM Providers Compared: Three Business Models, Not Three Products

Three brands, three pricing models, one set of towers. Airalo, Saily and Holafly all sell data on the same three Italian networks. What differs is how they charge you, and that difference decides which one is cheaper for your trip. Airalo prices by gigabyte, Holafly prices by day with a speed policy behind the word unlimited, and Saily sits between them with day-sized bundles at low entry prices. For a three-day break the daily model often wins; for two weeks the per-gigabyte model usually does; for a month neither is the right shape.

This is not a ranking. If you want the options ordered by trip type, the ten-plan comparison does that. What follows is the part that ranking cannot cover: why the three price differently, and what that means for your specific itinerary.

Three Italy eSIM providers, three different bets

The three brands are not variations on one idea. They made different bets about how travellers buy data, and the pricing reflects the bet.

Airalo built the largest catalogue in the category and monetised it through breadth. Its model is per-gigabyte: you buy a quantity and a validity window, and the company’s margin comes from the gap between wholesale capacity cost and retail pricing. Breadth is the moat — the brand is available in more countries than most competitors, which matters more to a multi-country traveller than to someone going to Rome for four days.

Holafly made the opposite bet. Instead of selling gigabytes, it sells days with no quantity attached, and the customer stops counting. That is a genuinely different product experience, and it is priced accordingly: the daily rate is higher than a metered plan’s effective daily rate for light use, and lower for heavy use. The bet is that travellers will pay a premium for not thinking about data.

Saily entered later, from a company whose main business is consumer privacy software, and positioned on price and simplicity. The plans are day-sized with fixed data, the app is clean, and the checkout has few upsells. The bet is that a large share of travellers want a small, cheap, unambiguous purchase rather than an optimised one.

None of the three operates a network in Italy. All three buy capacity from TIM, Vodafone or Wind Tre, or from an aggregator that does. So when a review says one has “better coverage in Italy”, what it usually means is that one has a partner arrangement that allows switching between carriers and another does not — a real difference, but a commercial one rather than a technical one.

The pricing models, and why they diverge

Understanding why the models differ explains why your bill does.

Per-gigabyte pricing is a bet on light users. A brand selling GB sets a rate that must cover its wholesale cost plus margin. Light users subsidise the model, heavy users cost it money. To protect against the latter, per-GB plans impose validity windows: 7 days, 15 days, 30 days. The window is not a customer convenience, it is a risk control.

Per-day pricing is a bet on average consumption. A brand selling days has to model how much the average customer consumes per day and price above it. Because consumption is heavily skewed — a few users consume enormous volumes — the price has to cover the tail. That is why a daily unlimited plan can look expensive next to a small metered plan and cheap next to a large one.

Day-sized bundles are a hedge. Fixed data per day combines both: the customer knows the daily number, and the brand caps its exposure. It is the easiest product to understand and the least optimised for any specific user.

The practical consequence for a buyer is that the models do not converge. No amount of shopping will make a per-day unlimited plan cheap for a light user, because the price structurally includes the heavy users. And no amount of shopping will make a per-GB plan cheap for a user who streams daily, because the model prices that behaviour out.

The same ten days, three different bills

Here is the comparison that matters, using a trip description rather than a product list. Ten days in Italy, one traveller, moderate use: maps, messaging, photos, an hour of video most evenings, one video call every other day.

For this specific traveller — moderate use, ten days — the per-GB and trip-length models land in the same place and the daily unlimited model costs more. Reverse the profile to a heavy user streaming three hours a night, and the ordering flips: the per-day unlimited model becomes the cheaper one, because the per-GB plan cannot supply enough data at any reasonable price.

The lesson is not that one model is better. It is that your consumption profile, not the brand, decides the winner — and most people guess their profile wrong, usually by overestimating how much they will stream and underestimating how much photo backup costs.

What each model does to a light user

A light user is the majority: maps, messaging, the occasional photo, no streaming. Perhaps 300–500MB a day.

Per-GB. Excellent. A 3GB plan over seven days covers it with margin, and the entry prices in this model are the lowest available. The risk is buying too small a tier and topping up, which costs more than buying the right tier once.

Per-day unlimited. Poor value. You are paying for a tail you will never reach, and the daily rate reflects heavy users you are not one of. The only justification is certainty, which for a light user is worth very little because running out is unlikely.

Day-sized bundle. Neutral to good. The daily allowance is more than a light user needs, and the reset means unused data is not carried forward — a minor inefficiency that rarely matters at this consumption level.

The verdict for light users is unambiguous: buy metered, buy the smallest correct tier, and ignore unlimited. This is the single most common overspend in the category.

What each model does to a heavy user

Heavy use means streaming, tethering, or daily uploads — 2GB a day and up.

Per-GB. Expensive, and eventually unavailable in the tier sizes offered. A user consuming 20GB over ten days needs a large tier, and the per-GB rate at that volume is where the model is least competitive.

Per-day unlimited. The right shape. The threshold may still bite, but the product is designed for exactly this user, and the price per day is defensible once consumption passes a couple of gigabytes daily.

Day-sized bundle. Workable only if the daily allowance is large. A 1GB daily cap is a poor fit for someone consuming 2GB; a 3GB daily cap is fine.

For heavy users the argument for unlimited is legitimate rather than psychological, and the deciding detail becomes the threshold — how much high-speed data sits behind the word. That mechanism is covered in the unlimited data breakdown, and it is worth reading before choosing between two unlimited plans.

The network question: all three resell the same towers

A claim worth dismantling, because it appears in most comparisons between Italy eSIM providers: that one has better coverage than another.

There are three national networks in Italy, plus two newer entrants with narrower footprints. An eSIM brand does not build towers; it buys access. So the coverage question reduces to two much smaller questions: which Italian network does this brand’s plan attach to, and can it switch between them?

Single-carrier attachment means your profile is tied to one partner for the duration. If that partner is congested or weak where you are standing, you wait or move.

Multi-carrier attachment means the profile can re-attach to whichever partner has the stronger signal. This is a genuine, noticeable advantage in three situations: crowded city centres in the evening, high-speed trains crossing coverage boundaries, and thick-walled historic buildings where one band penetrates and another does not.

The variation between brands on this point is real but rarely advertised, because it is hard to express as a marketing claim. The way to find out is to look for the phrase “connects to multiple local networks” or an equivalent in the plan description. If it is absent, assume single-carrier.

What does not vary is the underlying quality of Italian mobile networks, which is good by European standards. Rome, Milan, Florence and Venice are covered to saturation by all three. The gaps are geographic — the Apennines, inland Sicily, Sardinia’s interior — and they affect every provider equally.

Support, and the difference between scripted and useful

All three run support through chat, and all three start with the same four questions: restart, toggle airplane mode, check the roaming switch, check the APN. That sequence resolves most cases because most cases are those four settings.

The differences appear in three places.

Hours. A brand with no coverage during Italian business hours leaves you waiting overnight on a short trip. Check whether support is genuinely 24/7 or nominally so.

Escalation. When the script fails, the case has to go to the carrier partner. Some brands do this quickly and keep you informed; others go quiet. Reviews that mention a specific unresolved problem are more informative than star averages.

Language. English support is standard at all three. The quality varies, and it varies by shift rather than by brand.

One structural point that reviews rarely explain: the Italian carrier cannot help you. You are not their customer, you have no account with them, and walking into a TIM shop with an eSIM problem usually ends with a polite refusal. Your only route is the brand you paid.

Refunds: the fine print differs more than the price

Refund behaviour is where providers distinguish themselves, and it is almost never compared.

The common baseline. An unactivated plan is normally refundable. An activated plan with data consumed is normally not, on the reasoning that the service was delivered as specified. A plan that never connected at all is usually refundable after support confirms the failure.

Where they diverge. Speed of resolution, whether a partial refund exists for a plan that worked for two days out of ten, and whether the decision is automated or requires a human. These differences matter more than a dollar of plan price when something goes wrong, and they are invisible until then.

What to do before you need it. Read the refund clause on the plan page, not the general terms. Note whether it distinguishes between “did not work” and “worked badly” — most do, and the second category is usually excluded. And keep the support transcript, because a dispute that reaches your card issuer will be decided on documentation.

App experience, judged by the first ten minutes

App quality is usually reviewed as a design question. It is more useful to judge it as a workflow question: what happens in the first ten minutes after purchase.

Installation path. Every brand sends a QR code, and most also support in-app installation. In-app installation is faster and less error-prone, because it removes the step where people scan a code on the same phone they are installing it on — a step that requires a second screen and confuses a surprising number of users.

Clarity of remaining data. A metered plan needs a clear remaining-balance display, because the whole anxiety of the model is running out. A per-day unlimited plan needs a clear validity countdown instead, because its anxiety is different. Apps that show the wrong metric for their own model are common.

Top-up flow. For metered plans this is the most important screen in the app, and it is frequently buried. A top-up that takes five taps is fine; one that takes a new purchase flow is a sign the brand’s model does not really expect you to top up.

Language and currency. Both are usually handled. The failure mode is a plan priced in one currency and charged in another, with a conversion spread that quietly adds a few percent.

What none of the three does well

An honest comparison between Italy eSIM providers has to include the boundary of the category, and there are four things no international brand does well in Italy.

A real Italian phone number. None of them provides one, because a +39 number requires a carrier contract and identity verification. If you need one, the route is a local carrier, and the visitor’s walkthrough covers what that involves.

Inbound SMS on an Italian number. Related and distinct: some plans include outgoing minutes, which is a different capability from receiving a verification code on a local number. What a +39 number actually requires separates the two.

Long-stay economics. Beyond about a month, all three are more expensive per gigabyte than a local bundle. The category is built for trips, not for residence.

In-person problem solving. There is no counter to walk into. When something goes wrong at 22:00 on a Sunday, the only channel is chat, and a refund is the only remedy on the table.

The wholesale layer that explains most of the differences

There is a part of this market that never appears on a plan page, and it explains why two brands selling “Italy data” can behave differently with the same underlying towers.

The chain has three levels. At the top are the mobile network operators — TIM, Vodafone Italia, Wind Tre — who own spectrum and radio equipment. In the middle are aggregators and mobile virtual network enablers, who buy wholesale capacity across many countries and resell it in standardised packages. At the bottom are the eSIM brands you actually buy from.

Most brands do not hold direct contracts with every Italian operator. They hold a contract with an aggregator, and the aggregator holds the carrier agreements. Three consequences follow, and all three are visible to a customer.

Your experience can differ from the brand’s reputation. A brand praised for reliability in Japan may attach to a different partner in Italy, with a different switching capability. This is why country-specific reviews matter more than brand-level ones.

Multi-network switching depends on the aggregator’s terms. If the agreement allows attaching to any of the three Italian networks, you get resilience. If it names one, you do not — regardless of what the brand’s global marketing says.

Support escalation has one more hop. When your problem goes beyond settings, the brand raises it with the aggregator, who raises it with the carrier. Each hop costs hours. This is the structural reason first-line chat support often cannot resolve a genuine network fault.

None of this is a scandal; it is how the wholesale market works. But it changes what a useful question looks like. Instead of asking “which brand is best in Italy”, ask “which Italian networks can this plan attach to, and can it switch”. That question has an answer, and it predicts your experience better than any brand ranking.

Multi-country plans: where the three diverge most

For a single-country trip the three models differ mainly on price. For a multi-country trip they diverge on structure, and the differences become consequential.

Coverage counted in countries is a weak metric. A plan covering 40 countries and one covering 120 are not meaningfully different for a traveller visiting three. What matters is whether the specific countries on your route are covered at full allowance, reduced allowance, or excluded.

Per-day unlimited models travel better. A daily plan that works across a region has an obvious advantage for a multi-country itinerary: you do not re-buy at each border, and the daily meter keeps running regardless of which country you are in. The catch is that a fair-use policy may still apply per country rather than per trip, so a month split between two countries can hit two thresholds.

Per-GB models need a regional tier. Buying country plans for four countries means four purchases and four installations. Regional per-GB plans exist and are usually better value, but the validity window becomes the constraint: a 15-day regional plan does not help a 30-day trip.

Switzerland remains the exception. It is outside the EU roaming framework, so an “Europe” plan may include it at a different rate, exclude it, or include it fully. A large share of Italian itineraries touch Switzerland, and this is the most common expensive surprise in the category. The border mechanics are set out in what an Italy eSIM does and does not cover.

The practical rule: count countries on your route, not countries on the plan. A 40-country plan that covers your four is better than a 120-country plan that treats one of them as an exclusion.

How to test a provider in the first hour

Whatever you buy from an Italy eSIM provider, the first hour after landing tells you whether it works. Five checks, ten minutes, and you know what you have.

1. Confirm which network you are on. The phone’s settings screen shows the carrier name for the active line. If it shows the brand name rather than an Italian operator, you are on a roaming profile — normal, but worth knowing.
2. Run a speed test. Record the number. You now have a baseline, and if the plan later slows you will know whether it is the threshold or the location.
3. Toggle airplane mode and watch it re-attach. A profile that re-attaches in under a minute is behaving correctly. One that takes five minutes or needs a restart has a switching problem you should raise early.
4. Test the thing you actually need. Send a message, load a map, make a video call. Testing a web page proves very little; testing your actual use case proves everything.
5. Screenshot the plan status page. Remaining data, validity end date, and the speed test result. If you later need a refund or a support escalation, documentation decided weeks earlier is worth more than a description.

Two things not to conclude from a first-hour test. A slow speed test in an airport terminal proves nothing, because terminal buildings are congested and shielded. And a fast one does not guarantee the same speed in a stone building in a historic centre. Test again at your accommodation, which is where you will actually use the phone.

Three claims that should make you suspicious

Marketing language in this category is consistent enough that specific phrases can be read as signals.

“Coverage in 200+ countries.” This describes where the brand can sell a plan, not where the experience is good. A country can be covered by a single partner network with no switching and still count. The number is a catalogue size, not a quality measure.

“5G ready” or “5G coverage.” The plan permits 5G attachment where the network provides it. It says nothing about your speed, which depends on band, congestion and building. Every plan in this comparison is “5G ready” in Italy, because the networks are — which makes the claim worthless as a differentiator.

“Unlimited.” Always a speed policy. The two numbers that matter, threshold and post-threshold speed, are usually in the terms rather than on the sales page. That mechanism gets its own treatment in the post on throttle mechanics.

One more that is not a lie but is a distraction: star ratings. A 4.6 average across thousands of reviews mostly measures how often the product works at all. What you want to know is what happens in the failure case, and that lives in the one-star reviews — specifically, whether the person got a refund and how long it took.

What a per-day price is actually pricing

Per-day plans can range widely — from about two dollars a day to five or more for what looks like the same promise. The gap is not arbitrary, and reading it correctly is the difference between a good and a bad unlimited purchase.

A low daily rate prices a low threshold. The cheapest per-day unlimited plans tend to carry small high-speed allowances, sometimes under a gigabyte a day, with a slow fallback. They are structurally similar to a metered plan with a daily reset.

A high daily rate prices resilience. More high-speed data before the threshold, faster fallback, sometimes multi-network attachment, sometimes better support. You are buying margin, not gigabytes.

The middle range is where the marketing does the most work. Same word, different thresholds, similar price. This is where reading the terms actually pays.

The useful calculation is to convert the daily rate into a data equivalent. A ten-day plan at $5 a day is $50. A metered 20GB plan might be $14. The difference, $36, is what you are paying for “no counting” — and it is only rational if you would genuinely exceed 20GB in ten days, which is roughly 2GB a day. Most people do not, and the ones who do usually know it because they stream.

Two people travelling: the arithmetic changes

The provider question changes shape when there are two of you, and the answer is not simply “buy two”.

The split-up scenario is the one people underestimate. A couple that separates for two hours in a city — one to a gallery, one to a market — is a couple where one person has no connection, no map and no way to arrange a meeting point without WiFi. On a ten-day trip that happens most days.

The practical compromise that most experienced couples land on: two cheap metered plans rather than one premium unlimited plan. Two 5GB plans typically cost less than one unlimited plan with hotspot, give both people independence, and cover moderate use comfortably. The hotspot route is better for families or for a group that stays together.

The one number that predicts your bill

If you want a single input that determines which model is cheaper for you, it is gigabytes per day. Not total gigabytes, not trip length — the daily rate.

Estimate it honestly, using the activity costs rather than a guess: navigation is negligible, messaging is negligible, an hour of social video is roughly 300–600MB, an hour of streaming video is roughly 700MB–1GB, half an hour of video calling is roughly 200MB, and photo backup is whatever your library did that day.

Then read the answer off this scale.

Two adjustments to that scale, both learned the hard way.

Your first day is always the heaviest. Photo backup, map downloads and app updates tend to fire on arrival, which is why a plan that looks adequate on day two can be gone by day one evening. Turn off mobile photo backup before you land and the daily rate drops by roughly a third.

Hotel WiFi changes the number completely. A traveller with reliable accommodation WiFi effectively consumes in the evening at zero cost, which can halve the daily rate. A traveller whose accommodation WiFi is unusable ends up tethering a laptop, which triples it. Neither is knowable before booking, which is an argument for a model that tolerates being wrong — a small plan plus a top-up, rather than a large plan bought on an estimate.

Which one for which trip

The table is deliberately short, because the decision usually comes down to two variables: how many days, and whether you will stream. Everything else is a refinement.

Questions that come up when comparing Italian eSIM providers

Is Airalo better than Holafly for Italy? Not in general — they are different models. Airalo is cheaper for light and moderate use; Holafly is cheaper for heavy daily consumption and removes the running total. The trip decides.

Do Saily and Airalo use the same networks in Italy? Both resell Italian carrier capacity, and the specific partner arrangements are not always disclosed. Assume the same underlying three networks and compare on price, plan shape and support instead.

Which has the best coverage in rural Italy? None has an advantage, because none owns a network. What can differ is multi-carrier attachment, which helps in weak-signal areas. Look for that in the plan description.

Are the cheaper brands worse? Usually not on network quality. The cheap plans tend to be smaller data tiers at shorter validity, which is a product decision. What you sometimes give up is support responsiveness and refund flexibility.

Can I switch providers mid-trip? Yes, if your phone holds two profiles. Install the second and set it as the data line. It is a reasonable strategy for a long trip, and cheaper than buying a large plan up front that you may not need.

Do any of them include calls to Italian landlines? Some plans include a small allocation of minutes, usually with a non-Italian number attached. That is useful for calling a hotel and useless for receiving a verification code. Read the exact wording.

Which is best for a family? Whichever sells unlimited with hotspot included, plus a power bank. The provider matters less than the hotspot terms.

Is it worth paying more for a well-known brand? Only if support responsiveness matters more to you than a few dollars. The network underneath is the same, and the app differences are smaller than the price differences.

Is a regional plan better than a country plan for Italy? Only if the itinerary leaves Italy. A country plan is cheaper for a single-country trip, and buying a regional plan for Italy alone means paying for coverage you will not use. The moment Switzerland, France or Greece appears on the route, the calculation reverses.

Which provider is fastest in Italy? None reliably, because speed is decided by the cell you are attached to, the band, and how many people are on it. Two users on the same plan can differ by a factor of ten depending on which side of a wall they are standing. What can differ is multi-carrier attachment, which helps when one network is congested.

Where this comes from. Italian network structure — TIM, Vodafone Italia and Wind Tre as the national networks, with Iliad and Fastweb as newer entrants — reflects the market as it stands. Plan structures rather than promotional prices are the basis of comparison, because promotional prices change weekly and structures do not. Refund and support descriptions follow standard practice across the category; the specific terms are in each provider’s own documents. The eSIM standard is maintained by the GSMA.

The comparison, compressed

Three brands, three bets. Per-gigabyte pricing rewards light use and punishes heavy use. Per-day unlimited pricing does the reverse, and hides a threshold. Day-sized bundles sit in the middle and are the easiest to understand.

Pick by consumption profile first, then by price, then by support. If you are light, buy metered and buy small. If you stream daily, buy unlimited and read the threshold. If you need a number or a month, none of the three is the right category — and that is the most useful conclusion a provider comparison can offer.

One last reframe, because it is where most buying decisions go wrong. The three brands are not competing to give you the best product; they are competing to sell you the most profitable package for your profile. That is not sinister, but it means the question “which is best” has no answer, while “which model fits ten days of moderate use by one person” has a precise one. Ask the second question and the market becomes simple.

Italy plans by duration and data size include unlimited and metered tiers at every trip length, with multi-network attachment across TIM, Vodafone and Wind Tre. For the ordered shortlist rather than the model comparison, see the ten-plan ranking.