Where you already outsource is where AI lands first
There is a ready-made opportunity map inside your company, and it is not in the technology plan. It is in the contracts you already sign every month.
Read your portfolio of outsourced contracts as a priority map: identify the intelligence share of each contract, recognise which budget line already exists, and choose the first contract to review.
Sign up once to unlock this course
Course 01 is open. For courses 02 to 05 we ask for six fields. It is a single signup: done once, valid for the whole track. It is not a free diagnosis and it does not trigger automatic sales contact.
The question of where to start is already written down
When leadership asks where to apply AI first, the search turns inward: which internal processes hurt most, which team complains most, where the spreadsheets pile up. It is an expensive search, because everything in-house looks like a candidate and nothing has a declared price.
There is a shorter path. Every company already outsources intelligence work and already pays for it with an invoice, a contract and a cost centre. Legal counsel reviewing standard contracts. A BPO checking documents. An agency producing variations of a creative. A contact centre answering the same thirty questions. A consultancy assembling a monthly report from data you supplied.
These contracts have three properties no internal process has: a known price, a written scope and a measured volume.
An outsourced contract is opportunity with a price tag
An internal AI project competes for new budget. A review of an outsourced contract competes for budget that already left the account last month. That difference changes who has to approve and how long it takes.
The contract also hands you, for free, what an internal project takes weeks to produce: the description of the work. It is in the scope, the SLA, the price-per-volume table. Someone was already obliged to write down what must be done, by when and against which acceptance criteria.
Monthly volume, price per unit, agreed turnaround, acceptable error threshold, what requires approval on your side, and how much of the work is explicit rule versus the judgement of whoever performs it.
Intelligence share: how much of the contract is explicit rule
No contract is entirely automatable, and treating it that way is how credibility is lost in the first meeting. What exists is a share. In a document-checking contract, the intelligence share can approach the whole thing, because the criteria sit in public regulation. In counsel that negotiates a sensitive clause, the share is small, because the value lies in judgement under pressure.
The estimate does not need to be precise, it needs to be defensible. Take the monthly volume, separate the cases that follow the standard path from those requiring an exception, and you have the share. A contract with 80% standard cases and 20% exceptions does not become an 80% saving: it becomes an operation where the exception finally gets real senior attention.
Starting inside usually stalls, and the reason is political
Automating an internal process means touching a team that has a manager, a target and a history. The gain shows up as people with free time, which nobody knows how to turn into a number, and the cost shows up as resistance, which everyone feels immediately.
Reviewing an outsourced contract flips the sign. The gain shows up as a cost line going down, something finance already knows how to measure. And there is no internal team defending territory, because the work sits outside.
This is not an argument to outsource more, nor to cut suppliers for sport. It is the observation that the first operation has to be the one whose result is visible without argument, because that is the one that funds the second.
When the contract is not the door
A low-volume contract does not justify the engineering. A contract whose real value is legal liability assumed by a third party should not come back in-house without that conversation. A contract where the supplier holds data you do not have requires negotiation before any design work.
There is also the case where the supplier already uses AI and passes the gain through in the price. There the opportunity is not to replace: it is to renegotiate with information, knowing which share of the service stopped being human work.
Outsourced contract inventory
List up to five service contracts your company holds today. For each, record the approximate monthly cost and the share that follows explicit rules. The ordering falls out on its own.
| Contract or supplier | Monthly cost | Explicit-rule share (%) |
|---|---|---|
Sort from highest to lowest. The first on the list is the candidate you take to the next budget meeting, with a price already known and a scope already written.
Basis
This course does not propose cutting suppliers. It proposes using the contract as a discovery document: it already contains volume, price, turnaround and criteria, which is exactly what an internal initiative born of intuition lacks.
When the inventory becomes an investment decision
If the contract at the top of your list has meaningful volume and an explicit-rule share above 60%, there is a case to assess. Prumo Discovery is TheNeil’s methodology for turning that candidate into a decision with baseline, feasibility and a recommendation.