Article · · By Innam Dustgir

What Not to Outsource as a Founder

What not to outsource as a founder: two dark slate stone blocks, one anchored on a brass base engraved KEEP IN-HOUSE and one set apart on a brass rail engraved OUTSOURCE, in warm window light. By Innam Dustgir.

Outsourcing is one of the most useful tools a founder has. It is also one of the easiest ways to make a problem bigger while feeling like you solved it. Knowing what not to outsource matters more than knowing what to hand over, because the wrong handoff rarely fails on day one. It fails slowly, when you realise nobody inside the business understands the thing you depend on.

I am not against outsourcing. Most early companies should use it. But over 12 years of building businesses I have learned to ask one question before signing anything: am I buying capacity, or am I trying to get rid of a problem I don't want to look at?

Outsourcing moves work, it doesn't fix it

When a function inside a company is slow, messy or full of approvals nobody can explain, handing it to an outside team feels like relief. Someone else will deal with it.

The trouble is that the mess usually travels with the work. The vendor inherits unclear ownership, half-documented processes and decisions that were never really made. Now you have the same problem, plus a contract, plus a team that is paid to follow instructions rather than question them.

So before outsourcing anything, I want to know why it is broken internally. If the answer is "we never agreed what good looks like," no vendor can fix that for me. The same is true for automation: if a workflow is unclear, fixing the process before automating it saves you from paying to run a bad process faster. Outsourcing works the same way. Clean it up, then decide who should run it.

What not to outsource: the work that makes you different

My simplest rule is this. Outsource the work that keeps the business running. Keep the work that makes customers choose you.

For most founders, the things I would not outsource are:

  • Strategy and the big decisions. Pricing, which markets to enter, what to stop doing. Advisors can help you think. The decision stays with you.
  • Your core product knowledge. If your edge is how your product works or how your service is delivered, someone on your payroll must understand it deeply.
  • Key customer relationships. A vendor can support customers. The relationship with the customers who matter most should sit inside the company.
  • Hiring decisions and culture. You can get help finding people. Who joins, and how the team behaves, is yours.
  • Ownership of your data, accounts and systems. Domains, admin access, customer data, the logic behind your automations. If a vendor holds the keys, you don't really own the business.

Everything else is a fair candidate. Bookkeeping, payroll, specialist design, some development, routine support, logistics. These matter, but being excellent at them is usually not why people buy from you.

The tricky part is that founders often confuse "I have always done this" with "this is core." A task can feel important because you built it yourself. That doesn't make it a competitive advantage.

Keep the knowledge, even when you buy the labour

This is the mistake I see most. A company outsources a technical function, and within a year or two nobody left inside can judge whether the vendor is doing good work. They can't tell if a quote is fair, if a delay is reasonable, or if a shortcut will cause trouble later.

At that point you are not managing a supplier. You are depending on one.

You don't need a full internal team for every outsourced area. But for anything important, I want at least one person inside who understands it well enough to ask hard questions, review the output and take it back in-house if needed. That person is your insurance policy. They also become the start of a talent pipeline, so the skill doesn't disappear from the business entirely.

AI makes this more important, not less. If an outside partner builds the automations and AI tools your team runs on, make sure the logic, the prompts, the data and the access stay with you. Otherwise you have outsourced a piece of how your company thinks.

Pay for outcomes, not activity

A lot of outsourcing contracts reward the wrong thing. They pay for hours, tickets closed or people assigned. None of that tells you whether the business got what it needed.

Before work goes to an outside team, I want a small number of outcomes agreed first. How fast should a customer issue be resolved? What error rate is acceptable? What should be live by when? Then the payment and the reviews connect to those results. The same thinking applies to choosing KPIs that drive real decisions: measure what changes behaviour, not what is easy to count.

This also makes the handoff cleaner. A vendor who knows exactly what success looks like can make good decisions without asking you about everything. Much like delegating effectively to your own team, the quality of the handoff decides the quality of the result.

When should you outsource? A simple test

Before outsourcing any function, I would ask four questions:

  1. Does this influence why customers choose us? If yes, keep it close.
  2. Is the process clear enough to hand over? If not, fix it first.
  3. Will someone inside still understand it? If nobody will, plan for that before you sign.
  4. Can we describe success as an outcome? If you can only describe activity, the contract will reward activity.

If a function passes all four, outsourcing is often the smart call. It frees your team for the work that actually moves the business.

One more signal worth watching. When your own team starts quietly finding workarounds, buying their own tools or going around an outsourced function, don't treat it as a discipline problem. It usually means the setup isn't serving them. That friction is information.

Harvard Business Review recently published a thoughtful piece on why outsourcing can't fix a broken organization, and it pushed me to write down how I think about this. My conclusion is simple. Outsource tasks freely. Never outsource understanding.

Frequently asked questions

What should a founder never outsource?

Strategy, key decisions, core product knowledge, your most important customer relationships, hiring decisions and ownership of your data and systems. These shape why customers choose you and who controls the business.

When should you outsource?

Outsource when the work keeps the business running but isn't why customers buy from you, the process is clear enough to hand over, and you can define success as an outcome. Keep at least one person inside who understands the work.

Can outsourcing fix a broken process?

No. Outsourcing usually moves the problem to someone else and adds a contract on top. Fix the unclear ownership or process first, then decide who should run it.

Is outsourcing good for startups?

Often, yes. It gives startups access to specialist skills without full-time cost. The risk is losing internal knowledge, so outsource tasks while keeping the understanding inside the company.

Originally published on LinkedIn