> ## Documentation Index > Fetch the complete documentation index at: https://docs.alvahq.com/llms.txt > Use this file to discover all available pages before exploring further. # Understanding Email Analysis > See how Alva reads your emails to surface commitments, sentiment, and next steps. When you [connect your email](/email/connecting-an-email-account), Alva analyses every conversation automatically. This page explains what Alva looks for and where the results appear. ## What Alva analyses For every email, Alva evaluates: ### Sentiment How positive or negative the conversation is. Sentiment is tracked over time, so you can see if a deal's tone is improving or deteriorating. ### Intent What the sender is trying to do: * **Request** — asking for something (a meeting, information, a proposal) * **Question** — asking a question that needs an answer * **Commitment** — promising to do something * **Update** — sharing information or a status change * **Objection** — raising a concern or pushback ### Urgency How time-sensitive the email is — low, medium, or high. ### Buying signals Phrases and patterns that suggest the prospect is moving towards a purchase: * Asking about pricing or contracts * Requesting a demo or trial * Involving additional stakeholders * Discussing implementation timelines ### Risk signals Patterns that suggest the deal might be in trouble: * Delays or postponements * Reduced engagement * Mentions of competitors * Budget concerns * Stakeholder changes ### Questions Alva detects questions asked by prospects and tracks whether they've been answered. Unanswered questions are flagged — these often represent blockers that need attention. ## Direction-aware analysis Alva analyses **inbound** and **outbound** emails differently: * **Inbound emails** (from prospects) are analysed for intent, buying signals, and questions * **Outbound emails** (from your team) are analysed for what was promised, what questions were addressed, and follow-up commitments This distinction matters because the same phrases carry different meaning depending on who said them. ## Methodology evidence If you're using a sales methodology (MEDDPICC, BANT, or SPICED), Alva extracts **methodology evidence** from emails. For example: * A prospect mentioning their budget → BANT "Budget" evidence * A prospect describing their pain point → MEDDPICC "Identified Pain" evidence * A prospect discussing their decision process → MEDDPICC "Decision Process" evidence When Alva finds evidence, it appears as a **suggestion badge** on the deal's methodology tracker. Click to review and apply it with one click — no manual typing needed. ## Where results appear ### On email records Each email in the system shows: * **Direction badge** — inbound or outbound * **Question count** — how many questions were detected and how many are addressed * **Entity links** — direct links to the related deal, contact, and company ### On deal pages The deal detail page includes an **Email Intelligence Panel** summarising: * All buying signals detected across emails * All risk signals * Outstanding commitments * Unanswered questions * Sentiment trend over time ### On the Emails page Go to **Inbox** in the sidebar to see all emails with their analysis results. You can filter and search across all conversations. ## Chronological processing Alva processes emails in chronological order within each thread. This ensures it has proper context — it knows what was said earlier before analysing a later reply. This is especially important for question tracking (knowing which questions have been addressed by subsequent emails). ## How it feeds into deal health Email analysis is a major input to [deal health scoring](/crm/managing-deals). Deals with positive sentiment, active engagement, and no unanswered questions score higher. Deals with risk signals, negative sentiment, or long gaps between emails score lower. This all happens automatically — you just need to keep having conversations, and Alva keeps your CRM intelligence up to date.