Legacy SaaS dossier · docsend

DocSend

CategoryDocument trackingDocSend Advanced (Dropbox upsell)Site ↗

A short-link generator with page-by-page view tracking. A weekend project sold as a sales-enablement category.

01 · Data

Data the SaaS sits on

S3 (PDFs)Yours

PDF.js rendererPublic

Resend (alerts)API

Internal Build Score

0.84

Feasibility of a useful internal substitute for the actually-used 20% of this SaaS, built with Claude and your own data layer.

IBS = 0.25·U + 0.25·C + 0.20·D + 0.15·I + 0.15·Mthis record · 84%

U
User adoption gap
weight 0.250.60
How much licensed surface area sits idle.
C
Core simplicity
weight 0.250.97
How small the actually-used 20% is when re-implemented honestly.
D
Data portability
weight 0.200.85
How clean the exit is, exports, formats, audit lock.
I
Integration burden (inverse)
weight 0.150.92
How few external systems must keep working through the SaaS.
M
Moat erosion
weight 0.150.95
How much the original differentiator has been commoditised by LLMs + cheap infra.

Editorial heuristic. Not investment advice, not a procurement recommendation.

03 · Feature usage

Commonly used vs. rarely touched

Commonly used

Upload PDF, get a short link
Per-recipient link with email gate
Page-by-page view duration
Email alert when a deck is opened
Revoke / expire link

Rarely touched

Spaces, folders of links
Data rooms competing with a real VDR
eSign add-on inside the Dropbox suite
Per-page commenting nobody uses on a pitch deck
Branded portals on the top tier
Custom domains as an enterprise SKU

Build it yourself→

Internal build

Build it yourself

01 · Stack & flow

Next.js

Postgres

PDF.js

Resend

Internal build map

Data in

PDF in S3

Recipient link

Viewer heartbeats

Agent layer

View summariser

Logic

email gate

5s page heartbeat

first-open alert

14-day roll-up

Outputs

Events table

Open alert

Engagement digest

Build time: 1 weeks
Run cost: $0.25/seat·mo
vs. SaaS: $15K / yr

Build outline

S3 upload + signed URL per recipient
PDF.js viewer with email-gate intercept
Heartbeat every 5s with current page
Events table: open / page / time-on-page
Email + Slack webhook on first open

02 · Claude prompt

First, map what you actually use (optional)

Step 1 · usage audit prompt

// DocSend usage audit, scope the rebuild Input: the document list with per-link view counts, the spaces / data rooms, and the team's send volume by user over 90 days. ## Do 1. Keep the link + view-tracking feature, that is the actual product. 2. Mark spaces / data rooms as active or abandoned by last view. 3. Flag any data room with NDA / gating needs, those may warrant a real VDR rather than a rebuild. ## Output JSON: { keep_features[], active_spaces[], vdr_candidates[], drop[] }.

Step 2 · rebuild the core

// Deck-view summariser, sales digest You receive (a) one prospect's view events for the last 14 days (timestamp, page, duration_ms), and (b) the deck's page index (slide titles). ## What you must do 1. Identify the slides the prospect lingered on (> 60s) and the slides they skipped (< 5s). 2. Produce one sentence: "Spent most time on …; skipped …; opened N times across M days." 3. Return a confidence flag: "strong signal" / "weak signal" / "noise", based on total session count and time depth. ## What you do not do - Do not infer intent ("they're going to buy"); only describe behaviour. - Do not paraphrase slide content the prospect didn't view. ## Output JSON: { summary, confidence, hot_slides[], skipped_slides[] }.

03 · Result

Summary for prospect ACME, Series A deck

deck:series-a-v3#prospect:acme

Spent most time on slides 5 (Market) and 11 (Pricing); skipped Team; opened 4× across 6 days. Strong signal.

← Legacy SaaS index AI wrappers index →