Your clients

Market-linked holdings your clients share with you · as of 21 Oct 2026

Needs attention

Clients

Client

Market-linked value

Status

Last logged in

Last contact

Next follow-up

What this mockup is

MFD evaluation demo. Sample clients only: no login, nothing saved, nothing sent. Notes, tags and follow-ups you add vanish on reload.
Return figures are illustrative. Total Return = (value − invested) ÷ invested. Annualized Return is a value-weighted sample, not the real solver.
"Last logged in" is sample data. Real visit logging is a separate feature (dropped from this scope).

AI, WhatsApp and pipeline

Talking points and drafts are rule-based stand-ins for the AI. They state facts from shared data and never recommend a product.
"Shared by client" events: a client can share a date from a private asset (FD maturity, loan end) without the amount. To decide whether that's worth building.
WhatsApp: the demo only logs the message to Notes. A real build needs the WhatsApp Business API or a click-to-chat link.

Pipeline

Add prospect works: name, source, referral, potential, next step. The lead lands in Lead.
Meeting done → hands off to Cue's prospects helper. It logs decisions and follow-ups straight away (sample content). Change the outcome to Needs time or Not interested and the follow-ups change with it. Undo is there if it was a mistake.
Send invite → turns the prospect into an invited client, carrying over the decisions and follow-ups.
Reviews: each client has a cadence (monthly, quarterly, half-yearly or yearly), changeable on their Profile. "Due" is computed from it.

Cue, calls and voice

Cue is the relationship agent the MFD sees. Five helpers work behind it (client profile, prospects, scheduling, follow-ups, reports). Their rules live in Settings → Cue, each set to Auto, Draft for me or Off.
On a client's Relationship tab, Schedule call is one click. Cue picks a slot, sends the invite, sets the brief, and later logs the transcript, note and follow-ups.
Defaults: internal work and calendar invites are Auto; free-text WhatsApp messages are Draft for me.
Video calls: in-app call with live transcript. Transcription needs the client's consent, ticked at the start. The summary and follow-ups are drafts the MFD confirms before saving.
Voice notes: after a phone call, record a quick recap. It's transcribed into a note with follow-ups. (Voice notes were parked on 21 Oct; back in by request.)
Everything is simulated: no calendar, meeting, WhatsApp or speech service is connected.

Benchmarks and daily movers

Each holding is compared with its own benchmark (Nifty 50 TRI, Nifty 500 TRI, Midcap 150, Smallcap 250, CRISIL bond/liquid/hybrid, gold, Nasdaq-100). The client's "matched benchmark" weights those by value. Bank FD and inflation are shown as reference points.
All index figures are samples. A real build must work out the benchmark return using the client's own investment dates (same cash flows run through the index), or the comparison misleads. That's money math, so it needs its own spec.
Today's movers use sample daily changes. Real figures would come from the daily price refresh.

Analyser and Report tabs

Analyser is computed, not AI: allocation, equity vs debt against the risk-profile range, market cap (estimated), Total Return by holding, fund house exposure. Always up to date, so no review step.
Report is generated on demand: an AI summary plus findings. It's a draft until the MFD edits and approves it; then it can be shared or added to the review.

Wealth analyser (findings)

Rule-based checks stand in for the AI: equity share vs risk profile, concentration, fund overlap, small/mid share, holdings below invested, goal pace, SIPs.
Human in the loop: findings start as an AI draft. The MFD edits or removes each one and approves before sharing. Approved findings also appear in the review.
Overlap is a name-based estimate. A real build needs fund portfolio holdings data.

Tax view and reviews

Tax view uses the current ₹1.25 L yearly tax-free limit on long-term equity gains. "Booked so far" and holding periods are sample data. A real build would use purchase dates from transactions.
Review (monthly, quarterly, half-yearly or yearly, set per client): a branded report with an AI-drafted summary the MFD edits before sharing. It covers market-linked holdings only.

White label

Branded as a sample firm, VasuniWealth. Change the name, logo, colour and advisor profile from the avatar menu → Firm profile & branding. Changes are demo-only.
Brand colour changes buttons and highlights only. Gains, losses and status colours stay fixed so returns always read the same.
ARN / EUIN values are placeholders.

Open questions (to raise, not decide)

Bonds & ULIPs in the MFD set. Use the compare toggle at the bottom of these notes to see the book with and without them. Vasuni has no ULIP type; shown here under Life Insurance.
Client consent + SEBI. Does an MFD seeing client data need explicit client consent, and does it raise distributor-vs-adviser questions? Needs a ruling before any real build.
Lapsed seats. Should an MFD still see a lapsed client's holdings (frozen, hidden, or live)? Mockup shows them frozen and excludes them from book value.
Access model. Can group_invitations / portfolio_grants / consent_ledger model MFD-to-client access? Unverified.
"Powered by Vasuni". Can a firm hide it, or is it always shown (and in the client app too)?
Where branding applies. Advisor dashboard only, or also invites and the client's own app?
ARN display. Check whether AMFI rules require the ARN on client-facing screens.
AI and the distributor line. Do AI talking points or drafts cross into advice? Who is responsible for what an AI draft says once the MFD sends it?
Tax-loss harvesting and the distributor line. Showing losses and unused limits is factual. Suggesting which units to sell may count as tax or investment advice. Where should the line sit?
Recording consent and retention (DPDP). How consent is captured and stored, how long recordings and transcripts are kept, and whether the client can see or delete them. Does SEBI/AMFI expect distributors to keep call records?
Build or integrate: video (own vs Google Meet / Zoom / Jitsi), calendar (Google / Outlook), WhatsApp Business API, speech-to-text that handles Indian English and Hinglish.
Agent autonomy. Which steps may ever run without approval?
Seat count. Mockup counts only active clients as seats; trials aren't billed yet.

Your clients

Needs attention

Clients

What this mockup is

AI, WhatsApp and pipeline

Pipeline

Cue, calls and voice

Benchmarks and daily movers

Analyser and Report tabs

Wealth analyser (findings)

Tax view and reviews

White label

Open questions (to raise, not decide)

Compare: bonds & ULIPs in the MFD set (internal)