Commercial Intelligence Architecture

Learning sales intelligence, I realized that in the job market, the product is me. Engineers do technology watch to anticipate their domain. But if I’m the product, I need to monitor the market itself: position myself in the right place at the right time.

I started searching for B2B SaaS Tech AI opportunities in France on WelcomeToTheJungle, LinkedIn, reading Maddyness for funding announcements, following founders on Twitter/X. As an engineer, I applied my reflex: think in systems, data flows, input → output pipelines.

2h Scrolled, Zero Applications

2 hours scrolled daily, zero qualified applications.

What I’m missing:

• Funding announcements (which company just raised €10M and is hiring?)
• Positioning shifts (who’s pivoting to agentic AI?)
• Hiring signals (which sales teams are structuring?)
• Tech trends (LLM, RAG, autonomous agents)

The System: Targeting, Positioning, Timing

The solution: structure this chaos to answer three questions:

• Targeting - Which companies to approach?
• Positioning - How to present myself?
• Timing - When to act?

Concretely, Delos announces a €10M Series A on Monday. By Wednesday, they’re in my Target List. By Friday, I’ve understood their LLM stack. In 10 days, I contact them during their commercial structuring phase.

“We will recruit sales forces to implement our AI employees in France but also internationally”

Container Architecture: The Three-Stage Pipeline

The pipeline has three stages. Collection: I gather signals (funding, jobs, pivots). Analysis: I understand patterns (who’s scaling, which GTM works). Action: I act with context. Today, it’s 80% manual. But it works: I’m finally taking action with structured insights.

1. Collection (Input)

I’ve structured by signal type:

Signal	Sources	Status	Notes
Funding	Maddyness, EU-Startups, Scaleup Corner	⚠️ Unfiltered RSS	50+ items/week, 90% off-target (US/non-tech)
Hiring Signal	LinkedIn Jobs, WelcomeToTheJungle, HelloWork	❌ Manual	Ad-hoc search 2-3x/week, missed opportunities
Company Intel	Websites, LinkedIn, blogs	❌ Manual	1h/company (pricing, stack, clients, ICP)
Tech Trends	Twitter/X, Google Alerts	⚠️ Semi-auto	Alerts: high noise. Twitter: manual but fast
Market Data	Dealroom, Crunchbase, Pappers	❌ Manual	Dealroom freemium limited. Pappers: FR only

The table shows the system. Here’s a real example.

Week of October 7. Reecall captured via Maddyness. Here are my notes after 1h:

• Maddyness: Reecall raised €3.5M on 09/24/2026 from Founders Future, Holnest, Mesh and Evolem, and Francis Nappez (BlaBlaCar) joins the board.
• DealRoom: The funds will finance the platform, new hires and European expansion, with no specific positions identified to date.
• Scaleup Corner: After a 2025 pivot to white-label, Reecall claims 4x revenue growth with Alcatel-Lucent, Simplify and Obat as clients (Brief IA).
• Grand View Research: The AI voice agent market would grow from $3.5B in 2026 to $35.2B in 2033, according to Reecall’s press release (Planet Fintech).

Four signals. Which to prioritize? Which to ignore? What does this say about Reecall?

2. Analysis (Processing)

Let’s take Reecall again. How to transform these four signals into a decision?

Three questions structure the analysis:

• Targeting - Contact them?
• Positioning - How to present myself?
• Timing - When to act?

Filtering examples:

Source	Signal	Verification	Decision
Maddyness	€3.5M + named investors	✅ Precise names	Targeting (traction) + Timing (post-raise)
Scaleup Corner	4x revenue + clients	✅ Named clients	Positioning (B2B enterprise)
Grand View	$35B market	❌ Marketing press release	Ignore

Current state: This filtering takes 30min per company. Template applied to 2-3 analyses/week. 9 remain in backlog.

3. Action (Output)

Intelligence note October 10: Reecall added to my Target List, high priority.

Targeting confirmed:

• Company profile research: successful 2025 white-label pivot, enterprise B2B clients (Alcatel-Lucent, Simplify, Obat)
• Theorized ICP: companies with high-volume customer support, need for AI voice automation

Positioning identified:

• Approach angle: understand their white-label GTM post-pivot, not selling product
• Structuring in progress (Francis Nappez on board)

Timing and next step:

• Active post-raise window (Series A 09/24/2026)
• Contact point: search in progress, not yet identified
• Outreach planned: week of October 21

After 2 Weeks

The base pipeline exists: collection format, analysis framework, structured intelligence notes.

What’s missing:

• No automated collection: RSS filtering and manual reading
• No automated enrichment: company research (product, clients, jobs)
• No method to search important contacts

Week 1 insight: Reduce job board scrolling. Pivot to cold outreach: prospect research, calls, meetings. Side effect discovered: it’s sales training.

Current vs target state:

Dimension	Today (2 weeks)	Target
Collection	5-10 companies/week, manual reading 2-3x	Filtered RSS feeds, automated alerts
Analysis	3 companies analyzed, 9 backlog, 1h/company	Scoring framework, 0 backlog
Action	3 intelligence notes created, 1 cold outreach sent	Active outreach pipeline, 2-3 contacts/week

Systemic Vision: Vicious and Virtuous Cycles

The real problem isn’t lack of information, but the system’s dynamics.

My vicious cycle, week 1:

1. Scrolling job boards 2h/day. Maddyness, EU-Startups, LinkedIn…
2. Scattered notes, no clear criteria.
3. Brain saturated.
4. Result: zero complete analysis, zero application sent.

The more signals I collect, the more I overload my brain, the less I act.

The pattern explaining this dynamic:

Where I am after 2 weeks:

Same system, different entry point.

• Week 1: red path (scattered notes → overload → blockage).
• Week 2: green path (template → visible backlog → planned action).

The goal isn’t to build the perfect system, but to shift from vicious to virtuous cycle. First outreach sent, first loop initiated. Now: accelerate so Outreach → Insights → Refined criteria spins faster.

Next Step: Automate Collection and Research

If I’m the product in the job market, then I need to do market intelligence like a SWE does technology watch. Not passive scrolling, but a system: Collection → Analysis → Action

The current bottleneck:

The backlog grows faster than I can process it. 9 companies waiting = 9h manual research.

The pipeline works end-to-end, but doesn’t scale: 1h/company to understand product, ICP, find contacts.

Next article: Automate collection and research

How to go from 1h to 15min per company? Again, decisions and structure.

The goal isn’t perfect automation. It’s accelerating the loop: go from 1 outreach/week to 2-3/week so my insights and market knowledge refine faster.

The next post will contain code, script examples, prompts, and scraping patterns I use for the French B2B SaaS Tech AI market.