Case study: building our own commercial system the way we build for clients

The client in this case study is us. We built the commercial foundation for Alea Applied AI the way we build for anyone else: positioning, offer architecture, website, operating workspace, prospecting infrastructure, outbound sequences, pipeline and content system. The figures cover the build period to 15 August 2026, and the short version is on the case study page.

Where it started

The business had broad automation capability and no clear, sellable operating proposition. That is a common position for a technical firm, and a weak one: capability without a proposition cannot be priced, targeted or proven.

The audience, the workflows, the offer, the proof model, the outbound system and the delivery path all needed to agree with one another before scale could be justified. Sending more email from a position that did not hold together would only have produced confusion faster.

The insight

Founders do not buy AI adoption. They buy fewer missed enquiries, faster follow-up, cleaner scheduling and less coordination overhead.

That reframed the offer from technology implementation to measurable operating outcomes with agreed acceptance criteria. It is also why the work we do for clients is held to criteria written before it starts: the commercial system and the delivery method rest on the same thesis.

What we did

Repositioned the business

Around applied AI systems for founder-led professional service firms. One audience, named, with workflows we could describe in its own language.

Designed the offer

A diagnostic that leads to a pilot, with written acceptance criteria and a path from there to implementation or a managed service. A buyer can see each step, and what would count as success, before committing to the next.

Built the market matrix

Five verticals across four regions: the United States, the United Kingdom, Portugal and Spain. Each cell of the matrix carries its own workflow pain, its own buyer language and its own campaign-specific sequences. Eighteen targeting cells across the approved matrix are built.

Built the operating system around it

The website, the operating workspace, the content system, the pipeline, the scorecards, and decision gates at day 30, day 60 and day 90. The gates exist so that a weak cell is cut on a date, by a rule, rather than defended.

Deployed the outbound infrastructure

Apollo infrastructure, contact sourcing, four-step sequences, controlled-volume sending and performance reporting. Volume was held down on purpose. The first sends were there to calibrate the system, not to fill a calendar.

The figures

Every figure below is verified. The activity figures cover 15 May to 15 August 2026.

  • Outbound infrastructure: 20 Apollo sequences built.

  • Targeting architecture: 18 targeting cells across the approved matrix.

  • Sequence-ready audience: 1,085 contacts added to sequences.

  • Controlled outbound activity: 236 sent, 234 delivered, 99.2 percent delivery.

  • Reached contacts: 148 contacts emailed.

  • Early replies: 3 email replies from 2 contacts.

  • Active-cell queue: 131 contacts scheduled.

  • Commercial operating model: decision gates at day 30, day 60 and day 90.

  • Market architecture: 5 verticals across the US, the UK, Portugal and Spain.

  • Demand capture surface: website, diagnostic action and proof-led offer live.

What these figures do not say

This case study does not claim won revenue, completed pilots or held meetings. Those figures are excluded from the outcomes on purpose. Three replies from two contacts is an early signal, and we report it as one.

The first performance data is treated as calibration. Delivery is healthy, response is early, and weak cells can be cut rather than rationalised.

What changed

The result is a functioning commercial system, not a brand refresh. Positioning, website, offer, pipeline, sequences, controlled sending, content and experiment governance now operate from the same thesis. It runs without anyone touching it, and we open it up on the diagnostic call.

Evidence

Verification sources: authenticated Apollo analytics and sequence inventory, the live operating workspace and the live website. The short version, with the figures at a glance, is on the case study page.