Julian Soler presses ahead with AI

Spanish grape juice and concentrated grape must producer Julian Soler is using Forterro’s MyForterro.ai platform to automate one of its most time-consuming daily tasks: working out the best recipe for every production run. The company expects the tool to save four to eight hours of calculation and review each day and to optimise production by up to 10%.
A daily calculation, automated
Julian Soler, the family-owned Spanish producer and exporter, is applying MyForterro.ai, Forterro’s AI platform for the industrial midmarket, to recipe formulation. The AI-powered recipe optimiser is built on Forterro’s Abas ERP solution and draws on the company’s own production, laboratory and quality data.
According to the announcement, the tool calculates every viable combination of raw materials. It then ranks the resulting formulations by cost and by how closely they meet each customer’s specification. The company expects the 10% production optimisation to come from factoring raw-material costs into every formulation.
Antonio López Soler, Julian Soler, said: “AI gives you in one minute what a person might take much longer to calculate.”
From hackathon to working tool
The project began at a two-day hackathon run by Forterro’s Abas teams, where Julian Soler chose to build a recipe optimiser around a genuine business need. The company then continued the work on MyForterro.ai, which connects leading AI models with the operational data held in Forterro’s software.
Julian Soler’s team configured the agent, adjusted the prompt and selected the AI models, supported by Abas specialists.
Why its own data matters
Because the optimiser runs inside the ERP, it works with the same production, laboratory and quality data the business already relies on, something a generic AI tool could not access, according to the announcement.
López Soler was clear on the role of that data: “Without the production, quality and laboratory data, we could not have implemented this solution.” He added that [MyForterro.ai] “is a tool that could help any company developing ingredient formulations improve its operational efficiency.”
Alice Taylor, head of AI, Forterro, described formulation as “a high-stakes decision that food and beverage manufacturers make dozens of times a day.” By connecting AI directly to the production, laboratory and quality data already held in Abas ERP, she said, “the team is replacing hours of spreadsheet work with ranked recommendations its people can act on.”
What the company expects
Julian Soler was founded in 1968 in Villanueva de la Jara, Spain, and sells grape-based ingredients in more than 60 countries across five continents. Its customers include manufacturers of soft drinks, wine, vinegar, confectionery and infant food.
The company expects the optimiser to deliver:
- Four to eight hours of potential daily time saving on recipe calculation and review
- Up to 10% forecast production optimisation through cost-optimised formulations
- Support for 30 to 40 production runs a day across laboratory, production and management
- Fewer manual errors and deviations, by predicting results before the industrial process begins
What comes next
Julian Soler plans to explore further uses of MyForterro.ai, including automating purchase delivery notes from PDFs, building agents that work from raw-material information, and applying AI to other Abas ERP functions.
The case centres on a single, repeatable decision: which raw-material combination best meets a customer’s specification at the right cost. Taylor’s remark that formulation is made “dozens of times a day” explains why the company chose to target it, and the stated expectation of 30 to 40 production runs a day points to the volume of work involved.






