Advisors can spend hours a day in their inbox while also juggling meetings, planning work, follow-ups, and client requests. Depending on complexity, a single client email can take anywhere from 5 to 45 minutes to answer thoughtfully.
The Vega automatic email draft feature is built to remove that friction.
The goal isn’t to automate communication for the sake of it. The goal is to give you time back while preserving your voice, judgment, and client relationships.
Automatic email drafts
How Vega generates email drafts
Vega can automatically draft responses to client and prospect emails and place them directly into your email drafts folder in Outlook or Gmail.
This feature has been part of Vega for some time, but how does it actually work? Let's address the two questions advisors frequently ask the Vega team:
How and when does Vega decide to draft a response?
How does Vega continuously improve its draft emails?
Below is a clear breakdown of how it works.
Vega runs quietly in the background and does not require you to change your existing email workflow.
To fully leverage this feature, you just need to connect your email client (Outlook or Gmail) to Vega. Instead of starting from a blank page when responding to client emails, you can start from Vega’s automatic draft before editing it and sending your response to your contact.
Vega follows a simple, structured process before placing a draft in your drafts folder.
Step 1: Contact identification
When a new email arrives, Vega checks whether the sender matches one of your contacts. This includes contacts from your connected CRM, contacts you manually added to Vega, or contacts Vega identifies through intelligent detection. Vega only drafts responses for recognized contacts, not for spam, vendors, or marketing emails.
Step 2: Response assessment
Vega evaluates whether the email actually requires a response. In most cases, it does. In some cases, it does not, for example brief confirmations, closing messages, or automated notifications. Vega has demonstrated strong accuracy, around 95%, in determining whether a reply is necessary.
Step 3: Draft creation
If a response is required, Vega generates a draft using relevant context. The Contacts agent is always involved, reviewing past emails, relationship history, and your writing patterns. Depending on the content of the email, additional agents may be used, such as those referencing prior meetings, financial data, or market context.
The resulting draft is then placed directly into your email drafts folder for review. You remain fully in control. Vega prepares the response, but you decide what gets sent.
Vega will also notify you by email once the email reply is in your drafts folder and ready for your review. This notification is on by default but can be turned off in your Vega settings.
Customizing your writing style
This is where Vega really becomes your AI email assistant.
How Vega recreates your writing style
We often think we have a single, distinct writing style. In practice, writing style is highly dependent on context and on who we are communicating with, which makes it difficult to define explicitly.
At Vega, we see this clearly with advisors. The way you write to a new prospect is not the same as how you write to a client you’ve worked with for ten years, and it’s certainly different from how you write to a client who’s also a close personal connection. Tone, formality, structure, and even word choice naturally shift based on the relationship.
Because of that, Vega does not try to learn one universal writing style for you.
Instead, when drafting an email, Vega looks at your relationship with the specific contact you’re replying to.
Two factors matter most here:
Contact context: Knowing who the recipient is makes a meaningful difference. Whether they are a prospect, an existing client, or a long-standing relationship influences tone, pacing, and level of detail. More relevant context generally leads to better drafts.
Conversation history: Past emails and prior meetings provide strong signals. Vega looks for consistent patterns such as how you open emails, how you close them, whether you use nicknames, how formal you are with that person, and how you typically frame responses in that relationship.
When a new email arrives that requires a response, Vega processes this information automatically. In under two seconds, it extracts a small set of relationship-specific signals from CRM data, past emails, and your preferences. Those signals are then used to guide the draft generation so the response fits the relationship, not just the content of the message.
Customize your writing style preferences
We’ve found that while many users have a good sense of their preferred writing style, the best results come when that feedback is shared directly with us and incorporated into Vega’s model in the background. We know how Vega behaves under the hood and how to translate writing preferences into guidance the system can apply consistently.
While Vega already learns from your past email conversations, we can also make deliberate adjustments so its drafts align more closely with how you write from the start.
On our side, this is done by adding a small set of writing rules to your Vega profile. For example:
Tone and cadence: short, direct, human, and warm without sounding cheesy.
Openings: “Hi [First name],” plus a natural one-line opener only when it fits the context.
Closings: a brief, friendly closing line that sounds like you, not like a help article.
Style: avoid long paragraphs and jargon, prefer plain language and shorter sentences, avoid phrases you never use, lean into ones that feel natural to you.
You don’t need to write prompts or templates for this. Just describe your preferences in plain language.
If you have patterns like this in mind, you can send that guidance to [email protected], and we’ll translate it into instructions on our side so future drafts align more closely with how you actually write.