Let us be straightforward about the conflict of interest: we are a party that does this work for money, writing about whether you should hire a party that does this work for money. Here is the version we actually give people on the phone.
Where your own team is better
- They know why things are the way they are. That odd module exists for a reason someone remembers. An outsider spends billable hours discovering what your team already knows.
- Small recurring work. Keeping images optimised and watching what new extensions do is continuous, and continuous work belongs with people who are already there.
- Anything touching your business logic. Nobody external understands your pricing rules better than the person who built them.
Where a specialist is cheaper
- Diagnosis. Recognising a pattern you have seen twenty times takes an hour. Finding it for the first time takes a week. This is the largest single difference and it is why we sell an audit rather than a retainer.
- Work you do once. Setting up Varnish properly is a skill you use once every few years. Learning it for a single occasion is expensive learning.
- A second opinion when the team is stuck. Not because they are less capable, but because familiarity makes some things invisible.
When outsourcing goes wrong
Two ways, and both are avoidable. The first is buying hours instead of an outcome: an open-ended retainer with no measurement produces activity rather than results. The second is a report that lands on a desk and nobody has the capacity to act on it, which is money spent on knowing rather than on improving.
Ask for a fixed scope with a measurement attached, and be honest about whether anyone on your side has time to implement. If the answer is no, that changes what you should be buying.
The uncomfortable one
If your store is slow because of years of deferred maintenance, no external party fixes that permanently. They fix it once. Six months later it drifts back, because the thing that caused it was capacity, and that has not changed.
In that situation the useful spend is not a one-off project. It is either a maintenance arrangement or an honest internal conversation about capacity. We would rather say that now than deliver a project that quietly undoes itself.