Skip to main content

AI Analytics

Structured sourcing data still leaves procurement teams with difficult decisions across cost, quality, risk, supplier concentration, savings and total cost of ownership. These questions are often answered using spreadsheets and manual analysis, making scenario testing slow, difficult to reproduce and hard to explain.

AI Analytics enables buyers to analyse RFx data using configurable AI actions and natural-language interaction. It helps buyers compare scenarios, evaluate trade-offs, prepare negotiations and generate executive-ready recommendations, while keeping procurement fully in control.

Using AI Analytics

  1. Log in to DeepStream.

  2. Navigate to the Request module.

  3. Switch to the Sent folder.

  4. Open the relevant sourcing event.

  5. Navigate to the Comparison tab.

  6. Click the black AI button in the bottom-right corner to open the AI Analytics panel.

  7. Choose how to start a conversation — you have two options:

    • Type your own question in the text box at the top and press Enter to send it.

    • Click a pre-built suggestion to run a ready-made analysis instantly.

      ✏️ Important to note: The number of suggestions adapts based on how many suppliers have submitted active bids.

  8. The chat window opens full-screen, on the left side showing the textual AI's response. You can continue the conversation by typing follow-up questions and/or inserting additional supporting documents. 

  9. When the AI returns a data report as part of its response, a download button (arrow icon) on the top right corner appears alongside it. Clicking it saves the report data to your device.

  10. On the right side of the screen, the Canvas displays a shareable visual representation of the AI-generated analysis.

    • Click the Share button (three-dot icon) in the top-right corner to generate a shareable link. The link opens the Canvas in an external webpage that can be shared with others. Recipients do not need to log in to DeepStream, as access is granted through the URL itself.

    • Click the Download icon in the top-right corner to export the Canvas as a PNG image.

    • Click the Full Screen button in the top-right corner to view the Canvas in a larger, full-screen display.

    • Click the Post-it icon to add your own notes to the Canvas.

    • Click the Brush icon to draw directly on the Canvas.

    • Use the Zoom In, Zoom Out, and Fit controls in the bottom-right corner to adjust the view.

  11. To close the chat, click the "Close this chat" button in the top-right corner. This returns you to the panel without losing your session.

  12. To return to a previous conversation, use the "Your recent chats" dropdown in the panel — it shows up to the 40 most recent sessions.

The AI analyses the selected RFx data and any supporting buyer-provided context to generate insights, scenario comparisons and negotiation recommendations. Buyers review the outputs before using them to support award decisions or negotiations.

Core Capabilities of AI Analytics

AI Analytics transforms sourcing data into explainable decision support. Its core capabilities include:

  • Natural-language analysis of individual and multiple RFx events.

  • Supplier comparison across cost, score, quality, risk, savings and coverage.

  • Alternative award and allocation scenario modelling.

  • Qualification, concentration, coverage and policy constraint analysis.

  • Identification of conflicting or infeasible constraints.

  • Sensitivity analysis and quantified trade-offs.

  • Supplier-specific BAFO and negotiation strategies.

  • Negotiation targets, leverage points and fallback positions.

  • Executive-ready Decision Packs, negotiation roadmaps and board papers.

  • Clear separation of native RFx data, uploaded context, assumptions and AI interpretation.

Best practices

  • Clearly define the decision objective and any constraints before running the analysis.

  • Validate important calculations before using them to support business decisions.

  • Use AI-generated insights to support buyer judgement, not replace procurement governance.

  • Review recommendations and assumptions before progressing to award or negotiation.

Disclaimers

  • It is not autonomous award execution.

  • It does not present assumptions as confirmed facts.

  • It does not replace procurement governance or delegated authority.

  • Award and negotiation recommendations remain subject to buyer review and approval.

Got a question or need clarification? Send us a message by clicking the Help chat button in the bottom right - our AI assistant, FIN, will help you right away, and connect you to our team if needed.

Did this answer your question?