Growing Customer Relationships through AI Expansion
In successful AI startups, contract size tends to grow over the life of a customer relationship.
A Vendor's Second and Third Contracts
Imagine a vendor that sells an AI solution to a company. The first project is modest in scope and budget. A year later, the same customer hands the vendor a project that is larger, more consequential, and more difficult than anything the vendor was asked to do at the start. This pattern shows up repeatedly in successful AI startups: the contract itself gets bigger as the relationship matures.
What do you think happens?
Before reading on: what do you think made the customer willing to sign a bigger, harder second or third agreement?
Reveal answer
The customer's trust in the vendor, which was built by the vendor having already delivered on the previous outcome.
The source states that expansion is a function of trust, and trust is a function of having delivered on the previous outcome. The customer does not extend the scope because the vendor asked for it; the customer extends it after the vendor has produced a result the customer can evaluate.
This article follows that pattern step by step. It traces how contract size changes across engagements, contrasts it with the seat-based growth familiar from SaaS, and explains why each delivered result makes the next, larger request possible.
Tracing Contract Size Across Engagements
The source describes a sequence of engagements in which each one is larger than the last. The first engagement is small, the second is larger, and the third is materially larger. By the third or fourth engagement, the vendor is working on problems the customer would not have been willing to hand over at the outset.
Read the timeline from left to right. Each box is one engagement, and each arrow is the step where a delivered result makes the next engagement possible. The growth is gradual and cumulative rather than a single jump. The specific project details in this walk-through are invented for teaching; the shape of the progression is the one the source describes.
Two Ways a Contract Can Grow
There are two different ways a contract can grow over time, and the difference matters. The comparison below shows the two patterns side by side.
SaaS expansion is growth in a contract mainly by adding more seats or users to a product that stays essentially the same. AI expansion, as described for successful AI startups, is growth in a contract by the vendor taking on harder and more valuable problems at the same customer site over time.
How Trust Carries the Next Engagement
The source gives the mechanism that links these steps. Expansion is a function of trust, and trust is a function of having delivered on the previous outcome. The vendor does not gain a larger contract by asking for one. The customer extends the scope only after the vendor has already produced a result the customer can evaluate.
Because each delivered result enables the next, larger engagement, the sequence is cumulative and cannot be skipped. A vendor that tries to begin with the hardest problem has no delivered outcome to point to, so the customer has little reason to trust it with that problem. The growth in contract size is therefore a visible trace of trust that has been earned one engagement at a time.
Following the Hypothetical Vendor Through Three Stages
A three-stage AI customer relationship
A hypothetical AI vendor serves one customer over three engagements. Explain why the third engagement is possible, and why it would not have been possible as the first engagement.
- 1Stage 1: modest first project
The vendor delivers a small, modest-scope result. The customer can evaluate it, and this is the first basis for trust.
- 2Stage 2: larger second project
Because the first result was delivered, the customer agrees to a larger and more consequential project. The vendor has a delivered outcome to point to.
- 3Stage 3: materially larger project
The second result also gets delivered and evaluated. Trust has now been built across two engagements, so the customer hands over a harder problem than the vendor was asked to take on at the start.
- 4Why the start would fail
If the vendor had asked for the third project on day one, it would have had no delivered outcome to point to, so the customer would have had little reason to trust it with that problem.
The third engagement is possible because of the delivered outcomes before it. Contract size grew in step with trust, not in a single jump.
Mistakes When Reading Contract Growth
Common mistake Treating AI expansion as the same as SaaS seat growth
What it looks likeLooking at a growing AI contract and concluding that the customer simply added more users.
Why it failsIn the AI pattern, growth comes from taking on harder and more valuable problems at the same customer site, not from counting how many people have access.
Do this insteadAsk what scope and difficulty of work the customer has entrusted to the vendor, not how many seats were added.
Common mistake Assuming the vendor can start with the hardest problem
What it looks likeA vendor proposes its most difficult project as the very first engagement.
Why it failsThe vendor has no delivered outcome to point to, so the customer has little reason to trust it with that problem.
Do this insteadBuild the sequence: deliver a result the customer can evaluate, then extend the scope.
Common mistake Expecting contract size to jump in one step
What it looks likeAssuming a relationship goes from a small first project straight to the largest possible one.
Why it failsThe source describes growth as gradual and cumulative, with each delivered result enabling the next, larger engagement.
Do this insteadLook for a series of engagements, each somewhat larger than the last.
Common mistake Confusing a customer's request for a bigger project with trust the vendor has not yet earned
Applying the Trust Sequence
A hypothetical AI vendor's first project was small and well received. The customer now proposes a project that is much larger and more difficult, with no further delivered results yet. Using the trust sequence from this article, explain whether the vendor should accept the larger project now, and what would need to happen first.
Hints
- Ask what the customer has actually evaluated so far.
- Recall that expansion is a function of trust, and trust is a function of having delivered on the previous outcome.
- Consider whether the required next step is a delivered outcome the customer can evaluate.
Key Takeaways
- In successful AI startups, contract size tends to grow over the life of a customer relationship.
- SaaS expansion grows mainly by adding seats or users to a stable product, while AI expansion grows by taking on harder, more valuable problems at the same customer site.
- Expansion is a function of trust, and trust is a function of having delivered on the previous outcome, so the growth is gradual and cumulative.
- A vendor cannot skip the sequence: without a delivered outcome to point to, the customer has little reason to trust it with the hardest problem.