Outsourced Appointment Setting vs In-House: The Cost Math

The comparison that kills most outsourced appointment setting deals happens in a hallway, and it goes like this: "seven grand a month? We could hire someone for that." The sentence is doing something dishonest without meaning to. A vendor retainer is a price, visible and complete on an invoice. An in-house SDR is a cost you have to construct, and the constructed number is reliably larger than the salary that started the conversation.

This piece is the arithmetic. It is a method rather than a set of figures, because the figures that matter are yours: your market's salary bands, your finance team's loaded-cost multiplier, your ramp history. Inventing those numbers here would make the comparison look precise and be worthless. What follows is the shape of the calculation, the inputs you have to supply, and the parts of the decision that the arithmetic does not settle.

The build side is seven numbers, not one

Salary is the input everyone starts with and the smallest part of the answer.

What you must supply before the comparison means anything

• ✓Yes: Base salary for the role in your market, at the seniority you would actually hire
• ✓Yes: Variable compensation, and whether it pays on meetings booked or on closed revenue
• ✓Yes: Employer taxes, benefits and insurance, expressed as your finance team's loaded-cost multiplier
• ✓Yes: Per-seat tooling: data, sequencer, dialer, verification, enrichment, CRM seat
• ✓Yes: Sending infrastructure: domains, mailboxes, warmup and deliverability monitoring
• ✓Yes: Manager time, costed at the manager's loaded rate, not treated as free
• ✓Yes: Ramp: months of full cost before steady-state output, from your own onboarding history
• ✓Yes: Recruiting cost and expected tenure, which together set your annual replacement charge

Three of those are the ones that get left out, and they are the ones that move the answer.

Manager time is a real cost and it is usually invisible. One SDR does not manage themselves. Somebody writes the criteria, reviews the copy, sits in on call reviews, handles the escalations, and rebuilds the target list when the first one underperforms. If that person is a sales leader, price their hours at their loaded rate and add it. A few hours a week at a director's cost is not a rounding error against an SDR salary.

Ramp is a cost, not a delay. A new hire draws full salary from day one and produces steady-state output some months later. The gap is money spent for partial output, and it belongs in the first year's total. You already know roughly how long your ramp is, because you have hired before. Use that number rather than a published average, because ramp length is mostly a function of how complicated your product is to explain.

Attrition converts into an annual charge. SDR tenure is short in most organisations. Divide recruiting and onboarding cost by expected tenure in years and add the result to every year of the model. If the honest expectation is that you rehire this seat inside two years, half of that cost lands annually. The second-order cost is worse than the recruiting fee: a departure resets the ramp, so the seat spends part of every replacement cycle back at partial output.

One structural point about the build side is easy to miss. Most of these costs do not halve when you hire one person instead of two. Tooling has per-seat components and fixed components, sending infrastructure is a fixed build regardless of how many people use it, and management overhead per head is highest at one head. A single-SDR experiment therefore carries the worst cost per meeting the model will ever produce, which matters if the plan is to test outbound cheaply before committing. Testing outbound with one hire is not a cheap test.

1. Step 1Build the loaded annual cost

   Salary plus variable, multiplied by your loaded-cost factor, plus tooling, infrastructure, manager time and the annual replacement charge.

2. Step 2Apply the ramp discount

   Year one carries full cost against partial output. Model year one and steady state separately.

3. Step 3Annualise the vendor quote

   Check the billing unit and include any minimum commitment or pilot period in the first-year figure.

4. Step 4Divide both by attended qualified meetings

   Cost per attended meeting is the only unit on which the two sides are commensurable.

Step four is where most comparisons quietly fail, because the denominator on the build side is a forecast and the denominator on the buy side is sometimes a commitment. Those are different kinds of number and the model should say so rather than averaging them into a single confident figure.

The buy side has published anchors, and only two of them

Vendor pricing in this category is mostly gated behind a discovery call, which makes anchoring harder than it should be. Two of the better-documented vendors do publish a starting figure.

$60,000Belkins starter, annualised

Published from $5,000 per month, stating 100 guaranteed appointments a year.

$90,350SalesRoads fractional SDR, annualised

Published from $6,950 per four weeks, with no outcome guarantee stated.

13Billing periods in a year on a four-week cycle

A four-week price is not a monthly price. Thirteen periods, not twelve.

Two things about those anchors. They are floors, describing entry configurations rather than a quote for your market. And the four-week billing unit is a genuine trap in a spreadsheet: $6,950 per four weeks reads as cheaper than a $7,400 monthly quote and is not.

The other vendors worth shortlisting mostly publish nothing. SalesHive states one flat monthly fee covering the SDR team, a strategist, the platform, data and tools, varying by team location, channel mix and daily touch volume, with no setup fees and cancellation on written notice. Martal charges a flat monthly fee at its entry tier and a flat fee plus sales commission above it, starting with a three or four month pilot depending on tier. Neither publishes a number. For what each vendor actually puts on the page, we went through them in appointment setting companies.

Two structural items belong in the buy-side total that the headline figure does not carry. Minimum commitments turn a monthly price into a first-year floor: a three-month pilot is a real amount of money you cannot walk away from, and it should appear in the model as committed spend rather than as a monthly rate. And where a vendor publishes no committed output, the cost per meeting is unknown at signing, which is information rather than a gap to fill with an estimate.

What genuinely favours outsourcing

Strip out the marketing and four advantages survive.

Sending infrastructure already exists and is already warm. This is the strongest argument and the one buyers underrate most. Outbound sends from separate domains with their own mailboxes, authentication and reputation, and new mailboxes have to be warmed over weeks before they carry volume. An in-house build pays that cost in calendar time before the first real send, and pays it again every time the sending pool expands. A vendor with running infrastructure starts sending on day one. The maintenance side, which is continuous rather than one-off, is covered in the cold email deliverability guide.

Deliverability expertise is expensive to learn on your own domain. Authentication, blocklist recovery, reputation monitoring and inbox placement are specialised and unglamorous. The learning curve exists either way, and the difference is whose domain absorbs the mistakes.

There is no ramp. The vendor's people already know how to do the work. There is still a setup period while criteria are agreed and infrastructure is prepared for your specific campaign, and that period is shorter than onboarding a hire.

Capacity is variable and replacement risk sits elsewhere. Some vendors state cancellation on notice, which converts a fixed annual commitment into something closer to a variable cost. And when the person doing the work leaves the vendor, that is the vendor's problem to solve rather than a hole in your quarter.

What genuinely favours in-house

The case for hiring is real and it is mostly not about cost.

Product knowledge compounds. An SDR who has sat next to your closers for a year handles objections a vendor's setter will hand back to you as a question. In a technical or heavily regulated category, that gap is large.

You control the message. Copy changes happen the same afternoon, positioning experiments do not need a briefing cycle, and what the market says back reaches your team unfiltered.

The person stays. A hire is an asset that appreciates and often becomes a closer. A vendor relationship ends and takes the accumulated knowledge with it, and sometimes the sending domains too, which is worth settling in the contract rather than discovering at the end.

No definitional dispute. There is no argument about what counts as a qualified meeting when nobody is billing you per meeting. That whole category of friction disappears, and it is a genuine operational cost of the vendor model.

Favours outsourcingMostly speed and infrastructure

• Sending domains already warm and monitored
• Deliverability expertise already paid for
• No hiring cycle and no ramp period
• Some vendors cancel on notice, so capacity is variable
• Attrition is the vendor's problem

Favours in-houseMostly knowledge and control

• Product knowledge deepens over time
• Same-day changes to messaging and targeting
• Market feedback reaches your team unfiltered
• The person can grow into a closing role
• No dispute about what counts as a qualified meeting

Reading the result honestly

Two failure modes are worth naming before you run the model.

The first is comparing a vendor's committed output against your own optimistic forecast. If a vendor guarantees a count and you assume a hire will match it, you have compared a contract to a hope. Model the hire against what your existing team actually produces if you have one, and against a deliberately conservative figure if you do not.

A related version of the same error is comparing the vendor's price against the hire's salary while comparing the vendor's output against the hire's best month. Pick one basis and hold it on both sides. If the vendor figure is a floor for an entry configuration, the hire figure should be a competent hire rather than an exceptional one.

The second is treating the answer as permanent. The build-versus-buy calculation is a snapshot of your current constraints. A team with no sending infrastructure and no outbound experience gets a different answer from the same spreadsheet than the same team eighteen months later, which is why a lot of companies buy first and build once they know what good looks like. The role-level version of this comparison, focused on the SDR seat rather than on meetings, is in outsourced SDR versus in-house, and the tooling side of the build is in SDR cost versus an automation stack.

If you decide to buy, the terms that determine what you actually receive are not the price. They are the written meeting definition, the counting source, the dispute window and the no-show policy, none of which appear on a pricing page. Scoring vendors on those is a separate exercise, set out in how to compare B2B appointment setting companies.

The short version

An in-house SDR costs salary times a loaded multiplier, plus variable compensation, tooling, sending infrastructure, manager time at a manager's rate, a ramp period of full cost against partial output, and an annual replacement charge derived from expected tenure. Build that number from your own inputs, annualise the vendor quote with the billing unit checked and any pilot period included, then divide both by attended qualified meetings. Outsourcing wins on infrastructure, deliverability and speed to first send. Hiring wins on product knowledge, message control and permanence.

RevenueFlow is paid on attended meetings against criteria agreed in writing before launch, which puts a real denominator on the buy side of that comparison. You can see what a campaign would look like for your market.

Vendor pricing and terms verified against the vendors' own pages in August 2026. All are subject to change; confirm current terms directly before contracting.

Sources: Belkins appointment setting, SalesRoads appointment setting services, SalesHive pricing, Martal pricing