NexitOS

Inspiration

NexitOS was inspired by the need to turn Inibsa’s historical sales data into clear daily commercial actions: not just showing what happened, but helping sales teams know who to contact first, why, and with what argument.

What it does

NexitOS identifies demand opportunities and churn risks at a customer-category level, so the same clinic can appear separately for Anesthesia, Biosecurity or Biomaterials. It generates a prioritized sales queue with urgency, economic impact, recommended channel, alert type and a clear explanation for each recommendation.

How we built it

We cleaned and joined sales, customer, product, potential and campaign data, created behavioral features such as recency, frequency, reorder cycles, opportunity gaps and volume drops, trained separate ML models for commodity purchase probability and technical churn risk, and combined those predictions with business rules to generate explainable priorities for the web.

Challenges we ran into

The main challenge was avoiding a generic dashboard and building something truly actionable. We also had to handle different product behaviors, incomplete potential data, inactive customers, and the need to explain every alert clearly to a commercial user.

Accomplishments that we're proud of

We are proud of building an end-to-end solution that goes from raw data to a usable commercial queue, separating commodity opportunities from technical churn risk and treating each customer-category as an independent commercial decision.

What we learned

We learned that prediction alone is not enough: a useful sales tool must combine probability, economic impact, urgency and explainability so that teams can act with confidence.

What's next for NexitOS

Next, NexitOS would integrate with CRM tools, capture feedback from sales actions, improve the models over time, add more external signals such as stockouts or margin data, and expand the system into a scalable commercial intelligence layer for multiple markets.