Anyone quoting you a single number for outsourced customer service before asking about your volume, hours and contact types is guessing. The same workload can cost very different amounts depending on where it is delivered, who staffs it, when it is covered and how skilled the agents need to be. Two proposals that look alike on the first page can be built on completely different assumptions.
This article does not give you a figure, because an honest figure does not exist until scope is defined. What it does is explain what drives the cost, how the pricing models you will be quoted actually work, what tends to sit outside the headline rate, and how to compare proposals so that the cheapest-looking one does not turn out to be the most expensive.
Why nobody can quote without scope
Customer service cost is a function of staffed time, and staffed time is a function of how many contacts arrive, how long each takes, when they arrive and what service level you want to hold. Change any one of those and the headcount changes. A provider who quotes before understanding them is either padding the number to cover the unknown or planning to renegotiate once the real workload appears.
This is why our own pricing is structured around engagement models rather than a price list, and why final pricing depends on role complexity, hours, coverage, tools and management requirements. It is also why the first step in any engagement with us is a call about needs, pain points, timeline and the approximate hours or agents required. The scope conversation is the quote.
The drivers that set the price
Delivery location has the largest effect. The same shift costs different amounts onshore, nearshore and offshore, and if one quote sits far below every other, delivery location is almost always the explanation. Location also changes what you get around the seat: time zone overlap with your own team, language mix and how easily your managers can work with theirs.
Dedicated or shared agents is the second driver. Dedicated agents work only your queue and build real product knowledge. Shared or pooled agents cover several clients and cost less per hour of coverage because idle time is spread across clients. Dedicated is worth the premium when product knowledge is deep or brand voice is distinctive. Pooled works well for high-volume, well-documented contact types where the answer is the same regardless of who gives it.
Coverage hours come next. Business hours in one time zone is the cheapest configuration. Extending to evenings, weekends and overnight multiplies the headcount required to hold the same service level, often by more than people expect, because a queue still needs a minimum staffing floor at three in the morning even when volume is low. Our article on what 24/7 support really takes covers that arithmetic.
Skill level is the fourth. Order status and password resets sit at one level. Technical diagnosis, regulated work, licensed activity or anything requiring specialist vocabulary sits at another. Tiering matters: paying a specialist rate for the whole queue when most of it is routine is the most common way to overspend. Sort your contacts by the skill each one genuinely needs before you ask for a quote.
Management and tooling make up the fifth. Supervision, quality review, reporting, workforce scheduling and the software agents work in all cost something. Whether they appear inside the rate or as separate lines is a matter of how each provider structures its proposal, not whether they exist.
Tier your contacts before you buy. Paying a specialist rate for password resets is the most common way to overspend on support.
The pricing models you will be quoted
Per hour or per full-time equivalent is the most common structure and the easiest to compare across providers. You pay for staffed time regardless of how many contacts arrive. It rewards accurate forecasting on your side, because you pay for quiet hours as well as busy ones, and it gives you a predictable monthly bill.
Per seat or per month is similar to per full-time equivalent but usually bundled with tooling, supervision and reporting. Read what is included carefully. Supervision and quality assurance are sometimes priced separately, which makes a headline rate look better than it is. A part-time or full-time dedicated professional on a monthly basis is the model we call dedicated support, and it suits recurring work that needs consistent ownership.
Per ticket or per contact means you pay for what is handled. It is attractive when volume is unpredictable and it shifts forecasting risk to the provider. Watch how a contact is defined. Whether a follow-up email on the same issue counts again, whether an abandoned call counts, and whether a transferred contact counts twice all change the bill substantially.
Per minute is common for inbound voice and answering services. It suits low, spiky volume where a dedicated agent would sit idle. It becomes expensive at scale, and it rewards short calls in ways that occasionally conflict with resolving the problem, so pair it with a resolution measure if you use it.
Project or task-based pricing covers a defined backlog or short-term workload with clear scope, deliverables and an agreed timeline. It suits cleanup work, overflow and one-off migrations rather than an ongoing queue. A managed team model combines several roles and skills under managed coverage with quality and capacity oversight, and is custom-scoped because no two combinations look alike.
What is usually billed separately
The difference between two proposals is often in what sits outside the rate. Common separate lines include telephony and per-minute carrier charges, software licences for the tools agents use, initial training time before agents take live contacts, ongoing quality assurance, dedicated supervision above a certain team size, reporting beyond a standard pack, language premiums for bilingual agents, and after-hours or holiday differentials.
None of these are unreasonable. The problem is only when they are discovered after signature. Ask every provider to list what is included and what is billed separately, in writing, before you compare rates. A provider that answers that question clearly is showing you how they will behave for the rest of the relationship.
The cost nobody puts in the proposal
Your own management time. A programme with clear documentation, defined escalation rules and a regular reporting rhythm takes very little ongoing attention. One without them consumes more of your week than the support did before you outsourced it. That cost never appears on an invoice, and it is the one most likely to decide whether the engagement feels like a saving.
Budget real hours for the launch period regardless of provider. Writing the process down, reviewing early contacts and correcting drift is the work that determines whether the engagement succeeds, and it sits on your side of the line. Our guide to the first ninety days sets out what that time gets spent on.
How to compare quotes fairly
Give every shortlisted provider the same scope: same volume, same hours, same contact types, same channels, same service level. Ask each to quote it in their own model. Then convert every quote to a cost per resolved contact using the inclusions each one has confirmed in writing. A per-minute quote, a per-seat quote and a per-contact quote cannot be compared until they are expressed in the same unit.
Ask what happens when volume rises sharply for a month and what happens when it falls. Ask about minimum commitments, notice periods and what a change of scope costs. Ask what a bad month looks like and what the provider did the last time one happened. The answers tell you more about the real cost than the rate does.
- Identical scope to every provider, quoted in their own model
- Written list of inclusions and separate charges
- Everything converted to cost per resolved contact
- Rules for volume up, volume down and scope changes
- Minimum term, notice period and exit arrangements
When the cheapest option costs the most
A low rate with pooled agents, no dedicated supervision and reporting priced as an extra can produce more repeat contacts, more escalations to your own staff and more of your management time than a higher rate with the right structure. Cost per contact handled is not cost per problem solved. The real savings usually sit in fewer contacts, faster resolution and less rework, rather than in a lower hourly figure.
Equally, paying for dedicated, onshore, round-the-clock coverage for a queue that is mostly routine and mostly daytime is money spent on a benefit you do not use. The right answer is the structure that matches the work, and that is only visible once the work is described properly.
Getting a number for your situation
Cost only becomes meaningful once scope is defined, which is why any honest answer starts with your volume, hours, channels and contact mix. You can move between project-based, dedicated and managed-team engagement models as workload, ownership and coverage needs change, so the first choice is not a permanent one. Our customer support outsourcing page covers the engagement models in more detail, and a discovery call is where the actual number gets built.