Andrew Crossley

    How do startups use AI?

    SHORT ANSWER

    Startups use AI in three places: inside the product as a feature that completes a user task, inside operations to remove repeated manual work, and inside go-to-market for research, content and support triage. The highest-return use in a small company is usually operational — it lowers headcount pressure immediately, with no dependency on customers changing their behaviour.

    Answered by Andrew Crossley, Fractional Chief Product Officer · Updated 2026-08-01

    Why it matters

    Small teams have the most to gain and the least capacity to experiment badly. Choosing the wrong first AI project costs a quarter.

    Investors now expect a coherent answer to how AI changes your cost base or your product. 'We use ChatGPT' is not one.

    How it works in practice

    1. 1

      Start with repeated internal work

      Support triage, sales research, onboarding data entry, QA. Fast payback, no customer-behaviour risk.

    2. 2

      Then add one product-facing task

      Pick a step users already find tedious and produce a reviewable draft rather than an automatic action.

    3. 3

      Measure cost per successful task

      Not tokens, not calls. Cost per task that a human accepted without rework is the honest unit.

    4. 4

      Keep a human in the loop where it matters

      Anything customer-visible, financial or contractual keeps review until the evaluation data justifies removing it.

    Common mistakes

    • Adding a chatbot because competitors did, with no defined task.
    • Automating a broken process — you get the same mistakes faster.
    • Sending sensitive customer data to models without a data-handling policy.
    • Building infrastructure for scale you do not have.

    FROM EXPERIENCE

    Support triage as the first AI project

    Classifying inbound support messages by reason code, drafting a first response and routing the exceptions is a genuinely boring project with an immediate payback: faster response times and, more importantly, structured data on why customers contact you.

    That reason-code data is the best product roadmap input most startups have and never collect.

    Frequently asked

    What is the fastest AI win for a five-person startup?

    Support and sales research automation. Days to implement, immediate time saved.

    Do we need an AI engineer?

    Not initially. Most first projects are prompt, retrieval and workflow design on hosted models.

    How do we avoid data risk?

    Use business-tier APIs with no-training guarantees, redact identifiers, and write a one-page data policy before the first pilot.

    IN SHORT

    • Three places: product features, internal operations, go-to-market.
    • Operations first — fastest payback, least behavioural risk.
    • Measure cost per successful task and keep humans in the loop where errors are expensive.

    Work with Andrew

    A short call to work out whether the problem is strategy, scope or speed.

    Work with Andrew

    THE FRAMEWORK

    The Crossley Method: idea to first revenue in seven stages

    See the full method
    1. STAGE 1DiscoverWeek 1
    2. STAGE 2ValidateWeek 2
    3. STAGE 3PrototypeWeek 3
    4. STAGE 4Build MVPWeeks 3-4
    5. STAGE 5LaunchWeek 5
    6. STAGE 6First RevenueWeek 6
    7. STAGE 7ScaleOngoing

    MORE ANSWERS

    AI and product development

    Will AI replace product managers?

    No, but it is removing a large part of what product managers used to spend their week doing. AI already writes specs, summarises research, drafts tickets, analyses feedback and builds prototypes. What it cannot do is decide what a company should refuse to build, hold accountability for a metric, or persuade a room. Roles weighted towards documentation are shrinking; roles weighted towards judgement are becoming more valuable.

    Which AI tools should founders use?

    Founders need four categories, not forty tools: a frontier chat assistant for thinking and drafting, an AI-assisted build tool for prototypes and MVPs, a research and synthesis tool for customer and market work, and automation for repeated operational tasks. Pick one per category, use them daily for a month, and only add a fifth tool when a specific recurring task justifies it.

    How much does it cost to build an AI application?

    An AI application costs roughly £10,000–£30,000 to build as a focused MVP on hosted models, and £50,000–£150,000 for a production system with retrieval, evaluation and integrations. Running cost matters more than build cost: expect £0.01–£0.30 of model spend per task. If a customer generates a hundred tasks a month, price accordingly — AI products fail on unit economics more often than on build budgets.

    How can AI reduce operating costs?

    AI reduces operating costs by removing repeated, rules-light manual work: support triage and first-draft responses, data entry and reconciliation, research and reporting, and QA. Realistic savings in a small company are 20-40% of the time spent on those specific tasks, not of total headcount cost. The saving is real only if the reclaimed hours are redeployed rather than absorbed.