How much money has the company lost over the last six months to stock-outs? Which customers have cut back their orders? Which products turn over too slowly, and where are the opportunities to grow sales?
These are questions sales managers and field agents need quick answers to, yet answering them usually means going through several reports, cross-checking the figures and sometimes hours of analysis in Excel.
The new Agent AI feature, available in NEXT SFA, changes the way users access and analyse commercial information. Instead of looking for data in predefined reports, users ask questions in plain language, in a chat window built into the application, and get answers based on the company's real data.
More than a virtual assistant, Agent AI becomes a commercial and operational analysis tool, able to spot trends, anomalies and situations that need action.
In NEXT SFA you no longer search, you just ask
A Sales Force Automation solution like NEXT SFA handles a considerable amount of information: orders, customers, products, stock, sales visits, transaction history and the activity of the sales agents. Thousands of records that seem to carry little value one by one, but turn into gold once they are put in context and looked at as a whole.
Almost everyone in business is used to traditional reports: a document that answers known, recurring questions. But what happens when a manager wants an analysis that does not already exist in the application? Or when the answer to one question raises three more? Then people from other departments spend their time gathering the data and putting it into a report. Agent AI removes much of that rigidity.
In practice, it is an AI chat window where the user can ask, for example:
- “Which are the 20 most important customers who cut back their orders over the last three months, compared with the same period last year?”
- “Which products had the most frequent stock-outs, and which customers were affected?”
- “Which sales agents had the largest growth in order value last quarter?”
- “Which products have high stock but falling sales?”
- “Which customers have not ordered in the last 60 days, although they used to buy regularly?”
The agent interprets the request, finds the relevant information and builds the answer from the available data and the user's access rights. The essential difference is that the user no longer needs to know which report holds the information, and does not have to search for it. They only need to know what they want to find out.
On top of that, the analysis is not limited to the current situation. Agent AI can use historical data, over months or even years, to identify changes and trends that could go unnoticed in day-to-day work.
Until now, getting complex information meant going through several reports or asking for additional analysis. With Agent AI, it becomes much simpler.
How does Agent AI work in NEXT SFA?
From the user's point of view, the experience is simple: a new chat window, built into NEXT SFA, where they can ask questions in plain language. Behind that interface is an architecture that connects an artificial intelligence model to the application's data. Agent AI works both on desktop and on mobile.
Agent AI can use ChatGPT, Claude or other compatible solutions. The connection to NEXT SFA data goes through an MCP (Model Context Protocol) server, which provides the integration between the AI model and the authorised data sources. This architecture gives controlled access to the data, without the user having to export it by hand or upload it into a separate AI application.
An essential point is that the access rights already set in NEXT SFA are respected. Agent AI works within the permissions of the user it is assigned to. A sales agent can only see the customers, orders and information they are authorised for, while a regional manager can analyse the whole area they coordinate.
MCP also makes it easy to extend the architecture to other data sources. For example, internal information can be complemented with public company data, from sources such as the Ministry of Finance, to add financial context to commercial analysis.
Implementation and licensing
Agent AI is available as an add-on for NEXT SFA, on a monthly subscription. It involves two separate components: the subscription for the Agent AI module in NEXT SFA, and the LLM service used to process the requests. Contracting the LLM service is the customer's responsibility, so they can choose the provider and model that fit their technical, commercial and security requirements.
The Next iSolutions team carries out the integration using the API access provided for the chosen AI service, together with the configuration needed to connect it to the application's data. This approach gives companies flexibility in choosing their AI technology and lets them adapt the solution to their own needs.
Would you like to see how Agent AI in NEXT SFA can analyse the commercial activity of your organisation? Contact the Next iSolutions team for a demo and for details on activation and integration.
