(
Operations & Scale · Retention
)
Eliminating Operational Drag for a Digital Logistics Platform
Identified where a fast-growing logistics SaaS was bleeding operator time and revenue, then built and deployed AI systems that removed the drag without removing human judgment from the decisions that mattered.
introduction
The platform connected SME merchants with last-mile delivery providers across three cities. They had grown to 340 active merchants and a team of 18. Revenue was ₦5.1M a month and rising. But the growth had a weight attached to it.
The founder was personally involved in merchant onboarding, exception handling, and weekly reporting. Two operations staff spent most of their time chasing status updates between merchants and delivery partners. Customer support was reactive and slow. The team was producing the output of a much smaller one. When we ran the operations audit, it was clear the business wasn't being held back by strategy, it was being held back by repetition.
the challenge
The core problem wasn't technology. It was that nobody had ever mapped which parts of the operation were genuinely judgment-dependent and which were just process-heavy tasks that a person happened to be doing. Merchant onboarding required collecting documents, running eligibility checks, creating accounts, and sending access credentials; none of which required human reasoning, but all of which were being handled manually by the same two people every time.
Delivery exception handling followed a decision tree that existed in the head of one operations lead. Weekly reporting to merchants was being compiled from four different sources and formatted manually by the founder's EA. Each of these was a discrete, solvable problem that together added up to a team running at well below capacity.
Solution
We ran a four-week audit before building anything. The audit identified nine repeatable processes consuming significant operator time. We prioritised three for the first build phase based on frequency, founder time impact, and complexity of automation.
Merchant Onboarding Automation: We mapped the full onboarding sequence and built an automated intake and verification workflow that collected required documents, ran eligibility checks against defined criteria, provisioned accounts, and sent personalised access sequences, all without manual intervention for standard applications. Edge cases flagged for human review.
Delivery Exception Routing: We documented the decision logic that the operations lead was applying in her head and built a rules-based routing system that handled the 80% of exceptions that followed a predictable pattern. The remaining 20% — genuine edge cases btw, were surfaced to the operations lead with context pre-loaded, reducing her decision time rather than replacing her judgment.
Automated Merchant Reporting: We connected the platform's transaction data, delivery performance metrics, and support logs into a single reporting pipeline and built automated weekly summaries sent directly to each merchant. The founder's EA was removed from the process entirely.
Result
14 hours of founder and senior operator time reclaimed per week
Merchant onboarding time reduced from an average of 6 days to under 18 hours
Delivery exception handling capacity increased by 3x with no additional headcount
Merchant-reported satisfaction scores improved 22 points in the first 60 days post-deployment
