SAASPOCALYPSEverdict #NOTION-7653
scanned 2026.05.04 · 18:21
subject of investigation

notion.com

AI workspace: docs, wikis, projects, agents
verdictCONTESTED
wedge score
48
/100
wedge thesis

the door is niche depth: Notion is a horizontal everything-tool, and any vertical slice — eng wikis, personal PKM, startup CRMs — can be out-featured by a focused tool before Notion's roadmap committee even schedules the meeting.

real walls — pick your flank·ship in 10–14 weeks·run for $22.00 + usage
the doorregulatory
wedge

where the walls are.

methodology →
the door

no regulatory wall — SOC 2 doesn't count.

watch out

their distribution is fortress-grade — they own their brand SERP end-to-end.

capital
3.0/10
investment the incumbent had to make
why this scoremedium confidenceNotion is a pure software product with no meaningful physical infrastructure, inventory, payments risk, or compliance...

Notion is a pure software product with no meaningful physical infrastructure, inventory, payments risk, or compliance teams. Their infra is commodity cloud (Vercel/AWS). The main capital consideration is engineering headcount to maintain a massive horizontal surface, but an indie builder attacking a vertical slice sidesteps most of that spend entirely. No enterprise implementation services, no hardware, no regulated capital requirements.

  • Detected stack is Vercel + Supabase + Cloudflare — all commodity, low-cost, replicable by a solo builder for ~$22/mo per the report.
  • No signals of proprietary data centers, compliance teams, or regulated financial/health obligations.
  • Plus plan at $10/user/mo implies pure SaaS margin structure with no capital-intensive delivery.
technical
6.0/10
depth of the underlying engineering
why this scorehigh confidenceThe block editor at scale is genuinely hard — virtualized lists, optimistic updates, zero-flicker navigation on...

The block editor at scale is genuinely hard — virtualized lists, optimistic updates, zero-flicker navigation on 10,000-block pages is months of tuning. Real-time multiplayer via CRDT (Yjs) with correct conflict resolution is a known engineering nightmare. However, TipTap/BlockNote commoditize 80% of the editor primitives, and Yjs + PartyKit/Supabase Realtime are off-the-shelf CRDT solutions. The moat is in the polish and performance profile, not in novel algorithms. A vertical attacker can ship 'good enough' collab and editor without matching Notion's full performance envelope.

  • Report explicitly labels real-time multiplayer editing 'hard' and 'The Notion feel at scale' as 'nightmare' — months of tuning, not a feature.
  • TipTap/BlockNote handle 80% of editor primitives out of the box, reducing the technical barrier for an attacker.
  • Yjs + Supabase Realtime / PartyKit are open-source, well-documented CRDT solutions available to any builder.
network
4.0/10
users compound users
why this scoremedium confidenceNotion has a large template gallery and a community ecosystem, but it is not a true marketplace or multi-sided...

Notion has a large template gallery and a community ecosystem, but it is not a true marketplace or multi-sided network. There is no meaningful liquidity problem to solve — pages don't require other users to have value. Viral loops exist (shared pages, team invites) but are standard SaaS referral mechanics. The 100M user base creates some social proof and template corpus, but a vertical attacker doesn't need to match that breadth.

  • Notion's template gallery and community are real but not a locked marketplace — templates are freely copyable and not network-dependent.
  • No app ecosystem or partner integrations cited as a structural moat; integrations are standard (Slack, GitHub, etc.).
  • Viral loops are standard team-invite SaaS mechanics, not a liquidity-dependent marketplace.
switching
5.0/10
stickiness of customer data + workflow
why this scorehigh confidenceSwitching costs are real but moderate. Teams accumulate significant structured data — nested pages, databases, wikis...

Switching costs are real but moderate. Teams accumulate significant structured data — nested pages, databases, wikis — that is painful to migrate. Notion does offer export (Markdown, CSV, HTML), which meaningfully reduces lock-in. Workflow lock-in is genuine for teams that have built internal tools on Notion databases, but it's not deeply integrated into billing systems, ERP, or regulated workflows. The pain is organizational inertia and migration effort, not technical impossibility.

  • Notion supports Markdown, CSV, and HTML export — data is not fully trapped, reducing switching cost.
  • Teams build nested page hierarchies and relational databases that are non-trivial to migrate, creating real but surmountable friction.
  • No deep ERP, billing, or regulated system integrations cited that would create approval-chain lock-in.
data
3.0/10
proprietary data accumulates over time
why this scoremedium confidenceNotion holds a large corpus of user-generated content (pages, databases, templates), but this data belongs to users...

Notion holds a large corpus of user-generated content (pages, databases, templates), but this data belongs to users and is exportable. There is no evidence of a proprietary behavioral data flywheel that trains unique models or a fraud/risk dataset. Their AI features (Q&A, writing assist) are built on top of OpenAI APIs — they are not training proprietary foundation models. The template gallery is a mild corpus advantage but is largely public and copyable.

  • AI features described as 'streaming completions via OpenAI' — no proprietary model training indicated.
  • User data is exportable (Markdown, CSV, HTML), meaning the corpus is not non-exportable or proprietary.
  • Template gallery is publicly accessible and copyable — not a locked dataset.
regulatorydoor
1.0/10
real licenses, not SOC 2 theater
why this scorehigh confidenceNotion is a general-purpose productivity tool with no regulated duties. No HIPAA BAA, no FINRA, no KYC/AML, no money...

Notion is a general-purpose productivity tool with no regulated duties. No HIPAA BAA, no FINRA, no KYC/AML, no money transmission, no clinical/EHR data obligations. SOC 2 compliance is present (enterprise tier) but per the rubric, SOC 2 alone is low. An indie builder can obtain SOC 2 Type II within 6-12 months via Vanta/Drata. There is no regulatory moat here.

  • No healthcare, financial services, or payment processing obligations detected.
  • Product is a general-purpose workspace tool — no licensed regulated duties.
  • SOC 2 is the likely compliance ceiling, which the rubric explicitly scores as low.
distribution
9.5/10
brand SERP grip, knowledge graph, news flow
take

the blunt take.

Notion is the Swiss Army knife that convinced 100M users to stop using scissors. The wedge isn't beating it — it's picking one blade and making it sharper than Notion will ever bother to.

Their horizontal surface area is the vulnerability: every new feature (agents, mail, calendar) stretches the product thinner. A focused contender — say, a dev-team wiki with native code blocks and GitHub sync, or a personal PKM with actual graph traversal — can own a niche Notion will always treat as a checkbox.

cost

cost of competing.

what they charge
Plus plan
$10
/ user/mo
billed annually; $12/mo if monthly. Free tier exists with block limits.
annual:$96
what running yours costs
01 · Vercel Pro (SSR + edge needed for collab)$20.00
02 · Supabase free (Postgres + realtime)$0.00
03 · Cloudflare R2 (file/image uploads)$1.00
04 · Resend free tier (notifications)$0.00
05 · LLM API (OpenAI for AI features)??? — scales with usage
06 · Domain$1.00
07 · Sentry free tier (error tracking)$0.00
TOTAL / mo$22.00 + usage
▸ break-even:immediately for solo users — Notion's free tier exists, but the Plus plan at $10/seat/mo breaks even at 3 seats ($30 vs ~$28/mo self-hosted)
build

what you're up against.

2 weeks block editor core · 3 weeks nested pages + database views · 2 weeks AI integration · 2 weeks real-time collab · 3 weeks polish + niche-specific differentiator
easy
medium
hard
nightmare
01
easy
Basic block editor
TipTap or BlockNote handles 80% of Notion's editor primitives out of the box. Paragraphs, headings, lists, code blocks — one afternoon.
02
easy
Nested pages + sidebar nav
Recursive tree in Postgres, drag-and-drop with dnd-kit. Straightforward CRUD with a tree-ish schema.
03
medium
Database views (table, kanban, calendar)
The schema is flexible properties on page rows. The UI is the work — each view type is its own component surface.
04
medium
AI integration (writing assist, Q&A)
Streaming completions via OpenAI are easy. The hard part is scoping context (which pages feed the prompt) without blowing token budgets.
05
hard
Real-time multiplayer editing
Yjs + Supabase Realtime or PartyKit. Conflict resolution is the nightmare inside this hard. Works until two people edit the same block simultaneously.
06
nightmare
The Notion feel at scale
Notion's editor is fast, keyboard-navigable, and handles 10,000-block pages without jank. Replicating that performance profile — virtualized lists, optimistic updates, zero-flicker navigation — is months of tuning, not a feature.
stack

their position.

detected signals· measured
hostingVercelframeworkNext.jscdnCloudflare
recommended stack · inferred
inferNext.js 15 + TipTap (block editor)inferSupabase (Postgres tree schema + Realtime for collab)inferYjs (CRDT for multiplayer)inferCloudflare R2 (asset storage)inferOpenAI API (AI features)
rivals

who else has tried this.

option A
AppFlowy (self-host)
Open-source Notion clone, Rust + Flutter, Docker-deployable. Already has databases, kanban, docs. Start here before building from scratch.
option B
Obsidian (free, local-first)
For PKM use cases, Obsidian is free, offline, and has a 1000-plugin ecosystem. Hard to compete with for personal notes.
option C
Outline (self-host)
Open-source team wiki, Markdown-native, GitHub/Slack integrations. Covers the knowledge-base wedge cleanly.
compare

similar scans.

same shape - different moat
ready to wedge in?
Get the wedge plan. Cancel some plans.
▸ generated with love, by a heartless robotverdict v2.5 · saaspocalypse.dev