The foundation came first.
A 14-person specialist firm running integrated reports and sustainability disclosures for listed companies. Public AI tools were off the table for the work that mattered most. We rebuilt the rails before deploying anything on top.
The situation
The firm built integrated reports and sustainability disclosures for listed companies. The work was deep, the relationships were long, and the revenue was healthy. The infrastructure underneath the work had not scaled with the firm's growth.
The complication
Three disconnected systems were doing work that should have been unified: Asana for project management, Harvest for time tracking, and Excel for billing. Manual reconciliation across all three. A 500+ hour overspend on a major project was not detected until months after it had happened.
And the bigger problem: every public AI tool on the market was off the table for the work that mattered. Board-level information, unreleased disclosures, sensitive client material. None of it could legally or ethically be put through a public LLM. So the work that would benefit most from AI was sitting precisely outside its reach.
The question
The temptation, in a 14-person firm, is to deploy a quick AI tool and hope. We took the slower path: rebuild the operational foundation first, then layer AI where it would actually compound.
The answer
The engagement ran in two stages.
Foundation work first
- Migration from Asana + Harvest + Excel to Productive.io as a single Professional Services Automation platform
- Standardised project workflow: initiation, discovery, execution, review, financial close. One tracking thread per project.
- Real-time profitability visibility, replacing retrospective month-end reconciliation
Sovereign Intelligence Engine on top
- SIE deployed with PII scrubbing, private model hosting for the most sensitive work, intelligent routing for everything else
- The work that had been ringfenced from AI for confidentiality reasons (gap analyses, report drafts, document reconciliation, client research) started moving through the platform
- Team-wide AI workshop focused on the actual workflows we mapped, not generic prompt training
Outcomes
Discovery to strategy in four weeks. Numbers landing across the next two quarters.
- 60%+ reduction in administrative time on status reporting and data duplication
- 85-90% reduction in reconciliation effort; real-time profitability visibility
- R17,000/month eliminated in redundant platform costs
- 60-80% reduction in first-draft turnaround time on recurring report sections
- Aim: shorten the firm's 2-year onboarding and ramp-up period
The foundation came first. Rather than layering AI onto broken processes, we addressed the underlying operational fragmentation before deploying any AI tools.— Engagement note
Sound like a wrestle you recognise?
The first conversation is a discovery call. We will tell you honestly whether we can help, and what the first step would look like.