You'll need to sign back in to access the Admin Panel.
CEO & CO-FOUNDER · STRATEGIC NOTES
Alex's Vision
Everything Alex has written — from hand-sketched February notes to June strategic memos — synthesised into one page. This is the direction he wants the product, company, and team to go.
📅 Sources: Feb 2026 → Jun 2026 · 12 documents · 6 themes
01
The Core Thesis — This is Not a SaaS
Alex's most repeated, most emphatic point across every document: you are not building software.
A
"What you are really trying to value is not a SaaS product. Most founders accidentally value only the software layer and completely miss the value of the data layer, intelligence layer, and network layer. Artemis potentially contains all four."
🗄️
1. Software Asset
Dashboards, analytics, Art-E, ingestion systems, recommendation engines, discovery. Creates subscription revenue and retention. Most companies stop here.
📦
2. Data Asset
150,000+ structured posts, 3.5M labelled comments today. Every post, reaction, comment, click becomes AI-ready. Raw data is worthless. Structured, labelled, continuously-updating data is exponentially valuable.
🧠
3. Intelligence Asset
The models that turn data into predictions, content recommendations, audience matching, trend detection. Gets stronger with every customer, every post, every interaction.
🕸️
4. Network Asset
Every creator, agency, marketplace becomes a node. As the graph grows, recommendations improve, switching costs rise. This is where valuation accelerates dramatically.
02
The Compounding Flywheel
Alex's single most important economic insight: every customer makes the platform more valuable for every future customer.
A
"Most SaaS businesses grow linearly. Artemis has the potential to grow exponentially because every customer increases the future value of the dataset. This is where traditional SaaS valuation starts to become inadequate."
150K+
structured posts already in the warehouse
3.5M
labelled comments — AI-ready today
600K
posts + comments per creator per year
1.2M
annual data points from a single agency (20 pages)
03
The Valuation Ladder
Alex's 7-scenario model. Revenue is not the primary driver — data, network, and intelligence are.
A
"A pre-revenue Artemis with completed infrastructure, working product, active creator adoption, millions of labelled interactions, and functioning intelligence models could reasonably justify a valuation in the low millions on replacement cost and future strategic value alone — not current revenue."
04
The Product Philosophy — One Question
From Alex's June 2026 planning note and the System Redesign. His most pointed critique of product direction.
"If you execute your current version → users will try it, feel overwhelmed, and leave. If you execute the simplified version → users will rely on it daily."
— Alex, System Redesign (June 2026)
❌ Not Building
Hootsuite competitor
Analytics dashboard
AI tool
Scheduling platform
Reporting suite
A collection of features
✅ Building This
A real-time decision engine for attention and behaviour
A system that tells users what to do next
A daily habit that reduces thinking
An intelligence layer that creators can't live without
A platform where every output = insight + action
The Correct 3-Layer Structure
Layer 1
Art-E
Decision Engine
NOT a chatbot
The interface to the system
Explains everything
Recommends actions
Everything routes through it
Layer 2
Alerts
Re-engagement Engine
Only fires when action needed
Problem + implied solution
Urgency + curiosity
Opens Art-E on tap
Layer 3
Feed
Passive / Scrollable Brain
Short, clear, actionable
Every item = "why does this matter?"
No raw metrics
No dashboards as default
05
The Artemis Habit Loop
Alex's blueprint for daily retention. This loop must repeat every day.
Art-E: "Post this. At 6pm. Target this cluster." One clear next step.
→
05
🏆
Reward
Engagement increases. User feels in control. Sees cause → effect. Trusts the system.
Alert quality: the difference
Bad Alert
"Your sentiment is down"
Observation only. No action. Creates anxiety, not engagement.
Good Alert
"Your last 3 posts are underperforming — here's what to fix"
Problem + implied solution. Creates urgency. Opens Art-E naturally.
5b
The MVP User Flow — Alex's Own Diagram
Hand-drawn in his June 3 "Path to Simplicity and MVP" note. This is the exact journey he wants a new user to experience.
A
"If a user enters, gets a 2-year breakdown of analysis from Art-E, is told how to optimise, shown hidden patterns, and presented with campaign and collaboration opportunities — then there is no confusion. They need zero new skills, just the ability to use social media and connect to Artemis. I believe this is success." — Path to Simplicity and MVP (June 3, 2026)
"We need to get out of the thought process of 'more features are better' and start thinking that less is more. More intelligence, less features. The most important question right now: what is going to be valuable to our first set of target users, and what will get them to invite others? Anything else at the moment is noise."
— Alex, June 3, 2026
5c
The Variable Rewards Model
Alex's psychological-engagement framework from the June 3 note — how to make Artemis a habit that keeps users coming back.
🔄
Rewards of the Tribe?
Changing AI feed. A constantly-changing intelligence feed so the user never sees the same thing twice when they open the app. Novelty drives the daily return.
⭐
Social Rewards
Starring / saving pages for collab. Gives a user validation that their page and audience are sought after by others. Status as a reward.
🔍
Rewards of the Hunt
Searching for intelligence + actionable insights, and relevant pages in the Discovery feed. The reward is psychological — finding what they're looking for, so they can decide to act.
🏅
Rewards of the Self
Validation from AI. Scores for completing Art-E suggestions, rewards for mastery and competence within the platform. Progress the user feels.
🕊️
Autonomy
No new skills needed. Never forcing use. They want to use it. The product earns the open, it doesn't demand it.
⚠️
The Warning
"Limited variability becomes boring and predictable. To maintain interest we need rewards that vary — so they're not predictable — in a super-easy, super-simple, but variable and attractive way."
A
Note for the team: most of these reward mechanics are staged for after MVP launch in the current plan (gamification sits in the Later track). Alex flags them here as the retention engine the product should grow into — the "why would they open it tomorrow?" answer.
06
The Intelligence Architecture (Graph Theory)
From Alex's Graph Theory V3 document and Sprint 3/4 input notes. This is the technical moat he's been describing since February 2026.
A
"Traditional analytics tools look at individual metrics. Graph-based systems ask deeper questions: who interacts with whom, which communities form, which users bridge groups, and how ideas move through the network. Standard metrics measure performance. Graph theory explains why that performance occurs."
Betweenness Centrality
Finds the bridge nodes — users who connect otherwise separate communities. These are the Super-Sharers Alex identified in Sprint 3. High betweenness = high virality potential.
Louvain Community Detection
Groups users by behavioural interaction patterns, not demographics. "Segments everything" — content archetypes, audience reactions, posting patterns, niche fingerprints.
Node2Vec + GraphSAGE
Converts the entire creator graph into ML-ready vectors. Enables cross-page recommendations and the Master System — the platform watching itself and learning globally.
A
"Artemis creates fingerprints of content, audiences, and reactions. If a segment crosses over 80% match to another page's fingerprint, Artemis can give in-depth insights about that crossover. The system learns anonymised, and gives positive insights to all users with similar patterns." — Feb 2026, Sprint 3 input notes
07
The Facebook OAuth Strategy
Alex's most immediately operational note. The approval is about trust, not tech.
A
"Meta is not primarily evaluating your technical sophistication. They are evaluating risk, legitimacy, compliance, and intent. The entire approval process is essentially a trust-verification system."
The Permission Ladder — Incremental, Not Maximalist
Phase 1 NOW
Login authentication
Page identity
Page selection
Read-only analytics
Creator discovery
Easiest to justify. Maps to built features. Gets us in the door.
Phase 2 POST-TRUST
Publishing permissions
Messaging access
Agency workflows
Page management
Only request once trust + usage + business verification are established.
Phase 3 SCALE
Marketplace integrations
Collaboration automation
Monetisation tooling
Ad account access
Only viable after Meta becomes cooperative through demonstrated stable use.
Alex's hand-written sketches and early Sprint 3 input. Many ideas from here are still core to the product.
📱
UX Vision (Feb 2026 sketch)
Home screen: KPI executive dashboard first, swipe up for trends, swipe up again for virality and sentiment. Swipe-based navigation, single clicks, simplicity. "Always assume our users are non-tech — simple, easy-to-use dashboards cater to everyone."
🎡
Discovery Flywheel (Feb 2026 sketch)
Discovery feed as a "constantly slow-moving flywheel" showing every analysed page with metrics. In early days, show all pages. As more subscribe, show "PLY — People Like You" — matching audiences, consent, niche. Centre card bold, fades as wheel turns.
🤖
Art-E Vision (Feb 2026)
Art-E should know what screen you're on and give context-relevant advice. Floating on mobile, full-screen on demand. On the Discovery feed: "Yes, audience match 76%, very positive, X engagements — is this a good fit for collaboration?" Art-E as your always-on advisor, not a separate tool.
💡
The Sentence (Feb 2026)
"Imagine Microsoft Copilot for your social media data — conversational like ChatGPT or Claude on an AI-integrated platform that increases its knowledge with every post and every interaction. A platform that learns everything, and uses all internal questions, answers, and platform usage to optimise your business strategy."
A
"We are a predictive, multi-client intelligence layer that learns across first-party data and amplifies human decisions in real time. Competitors explain the past. We predict the future. Static ML and no cross-client learning — each competitor client is siloed. We gain from every client." — Sprint 4 input notes, April 2026
09
Competitive Position
Alex's competitor analysis — where the gaps are and how to exploit them.
Where Competitors Fail
Static rules, no continuous learning
Report on the past, not the future
Each client siloed — no cross-learning
Hourly/daily data polling
Poor behavioural segmentation
No predictive API
No network effects
Artemis Advantage
Self-learning per post, page, niche
Predicts before the algorithm decides
Cross-client network learning
Minutes-level real-time velocity
Behavioural fingerprints, not demographics
Predictive API + data licensing (future)
Gets harder to replicate over time
A
"Potential licensing model: we can sell predictive indices, benchmarks, forecast outputs, API access to machine ratings, behavioural trend datasets to hedge funds, ad networks, media buyers, brand strategists, political analysts, AI labs. A much higher-margin product than SaaS — and not served well by competitors."
10
The Story We Tell
From Alex's April 2026 notes — the emotional and philosophical dimension of the brand.
"We believe in change. Change of social media with data-driven decisions. Professional, affordable tools. Taking power back from the big players. Most people find change hard — change is painful — we make it easy. Artemis AI is the only platform that truly believes in change: change your decision making, change your strategy, change your life. Are you ready for change?"
— Alex, Sprint 4 input notes (April 2026)
📊
The Landing Page
"Don't overcomplicate. People don't care what we do behind the scenes — they only care that it works. Use proven use cases, short clips, screenshots. Show the system saving a brand. Show real-time counters of posts analysed, comments processed — not static numbers."
✨
The First Experience
"Once analysis loads, Art-E says: 'This may look complicated — let me break it down for you.' Showcase the complexity, then immediately explain it in plain English. This is the first major inflection point — the user must feel: I want to share this."
🎯
Customer Respect
"The customer needs to feel important. Show appreciation of effort. 'Great work, yesterday's reach was fantastic.' Consistency and positivity. Positive nudges: 'You've implemented 15 recommendations with 21% uplift — what can I do for you today?'"
11
The North Star
If Alex could boil it all down to the essential directives for the team.
Product
"We are removing complexity. Art-E is the main product. Every output must include an action. We prioritise importance, not volume. We design for daily use. Ask: why would they open this tomorrow?"
Company
"This is not a SaaS. Four compounding assets. Density before scale. Every new creator you onboard is not just a subscription — it is a long-term increase in the intelligence and strategic value of the entire platform."
GTM
"Narrow wedge first. Dense niche. Minimal OAuth permissions. Conservative public language. Invest in the network moat underneath while staying operationally humble on top."
Long-term
"The ultimate opportunity is not to build another creator tool. It is to create the intelligence layer for the creator economy. A platform where the value compounds daily through data, models, and network growth."
£1-5M
valuation at sprint completion (no revenue needed)
28 Nov
MVP launch — the starting gun, not the finish line
£5-20B
ceiling scenario at 1M clients with network effects