Hire or outsource: how to decide on cost alone
Build the fully loaded cost of an employee and a vendor on the same basis, find the workload where hiring wins, and avoid the salary comparison that misleads everyone.
A comparison almost nobody makes correctly
The decision usually arrives as a single sentence. A designer costs this much a month and the agency costs about the same so we may as well have our own person. The sentence is comfortable and the two numbers in it are not comparable. One is a salary and the other is a total cost of delivery.
Cost is not the only thing that should decide this. Control and speed and institutional knowledge matter too. But the cost question can be answered properly and most businesses never answer it.
What each side costs beyond the headline
Salary excludes most of the cost of an employee. On top of it sit the employer's statutory contributions to provident fund and to state insurance where applicable and the annual provision for gratuity. Then come the real items. A laptop and a seat and the software licences that person needs. A recruitment fee which is commonly a share of the first year's pay. An annual increment you will be asked for whether or not revenue grew.
Two further costs rarely appear in anybody's sheet. The first is ramp. A new hire produces little in month one and partial output in months two and three so the first quarter of salary buys a fraction of a quarter of work. The second is management time and it usually comes from the founder whose hour is the most expensive in the company.
A vendor fee excludes your own briefing time and excludes anything the scope does not name. It carries GST at 18% which a registered business ordinarily recovers as input credit so the cash cost and the accounting cost differ. Payments to a vendor also attract tax deducted at source which is a deduction against their liability rather than a cost to you. The shape of both depends on your registration and the service. A qualified chartered accountant should confirm how they apply to you.
When hiring is the wrong answer
Do not hire for work you cannot fill. A specialist needs twenty productive days a month to justify a full salary. If the real workload is six days you are paying full cost for a third of a role and the rest gets filled with tasks the person was not hired for.
Do not hire a craft you cannot judge. An unmanaged junior in a discipline nobody internally understands produces work of unknown quality and the bill arrives as a lost year rather than a visible cost. Equally do not outsource the thing that makes you different. If your edge is how you serve customers then that work belongs inside no matter what the sheet says. And do not outsource if you have nobody to own the brief. A vendor with no counterpart on your side produces activity rather than outcomes and both sides end up frustrated.
How to build the two numbers
Put twelve months in a sheet for each option. For the employee add salary and statutory employer contributions and gratuity provision and equipment and software and a recruitment fee amortised across the year. Reduce the first quarter's output by a realistic ramp factor rather than reducing the cost. Then add the management hours it will take and price them at your own rate.
For the vendor add the fee across twelve months and the GST and your own briefing and review hours. Now divide each total by a unit of work that means something in your business. Campaigns launched. Articles published. Designs delivered. The result is a loaded cost per unit of output and it is the only comparable number in the exercise.
Then find the crossover. Raise the monthly workload until the employee's cost per unit drops below the vendor's. That volume is your decision rule. Below it you buy and above it you hire. Writing the rule down stops the argument being refought on instinct every quarter.
What to track afterwards
Loaded cost per unit of output is the number that matters and it needs recalculating yearly because salaries move and scopes drift. Track role utilisation next to it. If the internal specialist is busy sixty per cent of the time you bought forty per cent of a salary for nothing.
Monthly salary against monthly retainer is the flattering comparison and it is the one almost every business makes. Headcount flatters in the same way. A team of four looks like capability while telling you nothing about what it produces per rupee.
The common mistake
The error is the comparison in the first paragraph. Salary gets set against vendor fee and the employee wins by a margin that evaporates once contributions and equipment and ramp and management time are added. Businesses hire against a number wrong by a wide margin and discover it in month eight when the budget will not close.
The costlier variation is hiring one generalist to replace a vendor covering several disciplines. The sheet shows a saving because one salary is less than the fee. What actually happens is that four specialisms are now done at the level of one person's weakest skill and nobody can say which of the four is failing. That is an expensive year to unwind and the salary saved never covered it.
This guide explains how something works. It is not legal, tax or financial advice, and rules change. Confirm anything regulatory with a qualified professional before you act on it.