Scaling eSIM Operations: The Arithmetic of Growth

Line chart titled Scaling eSIM Operations, showing global travel eSIM retail spend rising from 3.3 billion dollars in 2025 to around 5 billion in 2026 and close to 10 billion by 2028.

The operational problem in a travel eSIM business is not complexity, it is arithmetic. Every task that requires a person costs roughly the same whether the order was $10 or $10,000, and at an average order value around $28 there is very little room for anything manual.

This covers what breaks at each stage of growth, which tasks have to be automated and in what order, what to document before the second person joins, and the single metric that determines how large the business can get.

What actually breaks as you grow

  • Support volume, because it scales with orders unless the causes are removed.
  • Reconciliation, because every profile is a charge you need to match to a payment.
  • Refunds, which stop being individually reviewable somewhere in the hundreds.
  • Rate management, as your destination list and volume tiers multiply.
  • Your own knowledge, once more than one person answers customers.

What breaks, and when

StageWhat worksWhat starts to breakWhat to build
First salesDoing everything manuallyNothing yetNothing; learn from each order
Tens per monthManual support, personal repliesRepeating the same answersA help centre and canned responses
Hundreds per monthOne person can still see everythingReconciliation and refund consistencyWritten policies, automated retention, order search
Thousands per monthNothing manual survivesSupport headcount, rate managementDeflection, self-service reissue, dashboards
Multiple marketsSingle-language operationTime zones, languages, local expectationsCoverage that follows customers, not your office

The transition that catches people out is between hundreds and thousands of orders a month. Up to a few hundred, one attentive person can hold the whole business in their head, remember individual customers and make sensible one-off decisions. Past that, the same approach produces inconsistency, slow responses and a founder who cannot leave the business for a week.

Global travel eSIM retail spend

$10B$7.5B$5B$2.5B0 $3.3B~$5B~$10B 202520262028 forecast

Source: Kaleido Intelligence. Growing with the category means your operational load grows with it unless the per-order costs come down.

1metric that decides whether you can scale: contacts per hundred orders
$28average order value, which cannot absorb manual handling
~$10Bforecast travel eSIM retail spend by 2028

Source: Kaleido Intelligence, 2026.

Contacts per hundred orders is the number that decides your ceiling. If it stays constant, every additional thousand orders brings a proportional increase in support cost, and at a $28 average order value that arithmetic runs out well before you would like it to. Growth in this category is mostly a matter of driving that number down rather than of finding more customers.

The tasks that must stop being manual

TaskManual cost at scaleWhat removes it
Resending an activation codeA ticket per occurrenceCodes retrievable in the account area
Reissuing a redeemed profileAn escalation to your providerSelf-service reissue in your own tooling
Answering "how do I install this"Repeated identical repliesPlatform-detected instructions at delivery
Checking remaining dataA lookup per requestUsage visible to the customer in-product
Matching orders to provider billingHours at month endAutomated daily reconciliation
Deciding individual refundsInconsistency and escalationsA written rule applied by default

Each row is a fixed cost per order. At a $28 average order value, any of them left manual will cap how large the business can get.

Order matters here. Code resends and install questions are the highest-volume items and the easiest to remove, which makes them the right starting point. Self-service reissue is next because it eliminates an entire escalation path to your provider. Reconciliation is less visible but becomes the most painful if deferred.

Reconcile before you need to. Every profile you issue is a charge against your wholesale account, and mismatches between your orders, your payments and your provider\'s billing accumulate quietly. Traced the next day, a discrepancy is a five-minute question. Traced six months later, it is an afternoon of forensics for an amount smaller than the time it took.

Documenting before you hire

What to write down before hiring

  • The refund decision rule, with the common cases
  • Diagnosis steps for the five recurring failures
  • When to reissue versus when to refund
  • What to escalate to the provider, and how
  • Which data an agent may access and why
  • The response you send for each common case

Signs you are scaling badly

  • Support headcount rising in line with orders
  • Month-end reconciliation taking days
  • Different agents giving different refund answers
  • Nobody able to answer outside one time zone
  • Rate cards maintained in a spreadsheet nobody trusts
  • The founder still handling escalations

The refund rule is the document that matters most. Two agents making reasonable but different decisions about the same situation is how a business acquires a reputation for being arbitrary, and customers compare notes publicly. Writing the rule down is a smaller job than it sounds and prevents a category of problem that is difficult to unwind.

The right-hand column is worth reading as a checklist. Support headcount rising in step with orders is the clearest signal that you are scaling the cost of a design problem rather than solving it.

A sequence for growing

  1. Measure contacts per hundred orders

    This single number tells you whether you can grow. If it stays flat while orders rise, your support cost rises linearly and eventually exceeds your margin. Driving it down is the whole of operational scaling.

  2. Automate the six recurring tasks

    Code resends, reissues, install instructions, usage checks, reconciliation and refund decisions. Each is a fixed per-order cost, and at this order value any one left manual will cap your growth.

  3. Reconcile daily, not monthly

    Every profile issued is a charge against your wholesale account. Daily matching catches discrepancies while they are traceable; monthly matching turns them into an investigation nobody has time for.

  4. Write the playbook before the second hire

    Diagnosis steps, refund rules and escalation paths need to exist as documents rather than in your head. The cost of not doing this is inconsistency that customers notice and compare.

  5. Follow your customers into their time zones

    Support demand concentrates when travellers land, which has nothing to do with your working hours. Self-service and asynchronous channels carry most of it; coverage should follow your destination mix as it grows.

  6. Review rates and quality on a schedule

    As destinations multiply, rate cards and carrier changes stop being memorable. A quarterly review of rates, volumes and refund clusters by destination keeps both margin and quality from drifting.

Step one is the discipline that makes the rest legible. Most operators track orders, revenue and perhaps refund rate, and never look at contacts per order, which means they cannot tell whether an improvement worked. Once you are watching that number, every automation and every instruction improvement has a visible effect, and the ones that do not work become obvious quickly.

Frequently asked questions

Support volume, because it scales with orders unless the underlying causes are removed. At an average order value around $28 there is very little room for any task that requires a person. The metric that captures this is contacts per hundred orders, and driving it down is essentially what operational scaling means in this category.
Activation code resends and install instructions, because they are the highest-volume contacts and the easiest to remove. Then self-service profile reissue, which eliminates an entire escalation path to your provider. Then daily reconciliation, which is less visible but becomes the most painful item if deferred.
Usually between hundreds and thousands of orders a month. Below that, one attentive person can hold the business in their head and make sensible individual decisions. Above it, the same approach produces inconsistency, slow responses and a founder who cannot step away, because everything depends on their judgement.
Because every profile issued is a charge against your wholesale account, and mismatches between orders, payments and provider billing accumulate quietly. A discrepancy traced the next day is a quick question; traced six months later it is an afternoon of forensics, usually for an amount smaller than the time spent finding it.
The refund decision rule with its common cases, diagnosis steps for the recurring failures, when to reissue versus refund, what to escalate to the provider and how, which customer data an agent may access, and the standard response for each common case. Inconsistency is what customers notice and compare.
Support demand concentrates when travellers land, which is unrelated to your office hours. Self-service and asynchronous channels carry most of it, so a well-built help centre with platform-specific instructions does more than extended staffing. As your destination mix broadens, coverage should follow customers rather than headquarters.
No, and if it is, that is the clearest signal you are scaling the cost of a design problem. Most eSIM support contacts are preventable in the purchase flow and the instructions. Adding people to absorb them makes the problem more expensive rather than smaller.
On a schedule rather than from memory. A quarterly review of rates, volumes and refund clusters by destination keeps margin from drifting as your country list grows and volume tiers change. Rate cards maintained informally in a spreadsheet nobody trusts is a common failure at this stage.
There is no reliable published benchmark, and any figure quoted without a definition should be treated cautiously. What matters is your own trend: measure it, remove the top causes, and check that the number falls. A flat contact rate while orders grow is the problem, whatever the absolute level.
After you have removed the preventable contacts, not before. If contacts per order are already low and volume is genuinely rising, headcount is a reasonable response. Hiring to absorb preventable tickets converts a fixable design problem into a permanent cost line.

Automate the tasks that cap your growth

eSIM Island provides API access, profile diagnostics, self-service reissue and real-time usage, so code resends, reissues and usage checks stop being tickets. Tell us about your volumes and we will set up access.

Book a Free Demo

Or explore the Reseller Program, API Integration and Business Roaming.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>