# AI Baseline

> Measure where your marketing or sales team's time goes and the KPIs it moves, check your Claude setup, and prove every AI change against it.

Pillar: Strategy & Advisory. Price: From £7,500, excluding VAT (all prices: https://kowalah.com/pricing.md).
Source: https://kowalah.com/services/ai-baseline

**Know where you're starting, so you can prove what changed.**

We record where your team's week goes and the KPIs that work moves, and check how your Claude account is set up. Every agent and skill we put in afterwards is measured against that starting point.

## Your CEO wants to see AI paying off. Nobody measured the before.

Your team has Claude, and a few people use it well. When the board asks what it has done for pipeline, conversion or retention, the answer is a handful of anecdotes, because nobody recorded where the time went before anything changed.
Without a starting point, every improvement is a claim, and the budget for the next one is harder to win.

## You're ready for a baseline when

- Your team has an AI goal with a date on it, and no way to measure it yet.
- Your CEO or CFO has asked what AI is doing for pipeline or retention.
- Your team uses Claude, and you don't know who has what connected, or which agents and skills are already running.
- You want to choose the first process to change from evidence, not from whoever shouts loudest.

## What we measure, and how.

1. **Where the week goes.** A short interview with each person records their repetitive work and the hours it takes, step by step, against the process it belongs to.
2. **The KPIs it moves.** Each piece of work is tied to the KPI it affects, such as speed to lead, MQL to SQL conversion, renewal rate or share of voice in AI answers, with a reading taken now.
3. **Your Claude setup, checked.** Accounts and permissions, connectors, and the agents and skills already in use, plus which connectors to approve and what to fix first.
4. **Goals you can track.** Your KPIs set up as goals in the [Kowalah Platform](/platform), so every improvement afterwards is logged against the same starting point.

## What you leave with

- A baseline of hours and KPIs, by person and by process.
- A ranked list of the processes where agents and skills would move a KPI first.
- A check of your Claude setup, with what to fix first.
- Goals set up so every later change is measured against the baseline.

## Where it leads

- https://kowalah.com/services/agent-pilot
- https://kowalah.com/services/ai-plan

## Questions

### Is this an attribution tool?

No. HubSpot, Salesforce and your analytics tools own attribution. The baseline measures whether AI work is moving the KPIs you already report on.

### How much of my team's time does it take?

One short interview per person, plus a session with you to agree the KPIs. We do the rest.

### Do we need Claude in place first?

It helps, because we check how it's set up. If you're still buying it, we baseline the work and the KPIs now and check the setup once it's live.

---

**Start with your number.** A free 45-minute planning session. Bring the target you're judged on and leave with the first three processes to change, and where a baseline fits. Book a planning session: https://kowalah.com/contact?from=offer-ai-baseline