Close, Strong, Long: The Universal Formula for Algorithmic Trust
When I started reverse-engineering how algorithms evaluate relationships, I discovered something fundamental: algorithmic trust isn’t a feeling - it’s a calculation.
The calculation is simple: Close × Strong × Long.
This formula is universal. It applies to how algorithms evaluate ANY relationship between ANY entities - people, companies, places, events, concepts, products, creative works. The Knowledge Graph uses it. LLMs embed it in their parameter weights. Search engines reflect it in their authority scores.
Close, Strong, Long is how machines measure the validity of connections. Period.
A Universal Framework
Let me be clear about scope: CSL applies to every entity relationship the algorithm encounters.
- A person’s relationship to a company they founded
- A city’s relationship to the country it’s in
- An author’s relationship to books they wrote
- A product’s relationship to the brand that makes it
- An event’s relationship to its organizers
- A concept’s relationship to the person who coined it
- A university’s relationship to its notable alumni
The algorithm doesn’t care whether you’re optimizing a person, a product, a place, or an idea. It evaluates all relationships the same way: How close? How strong? How long?
The business examples in this article exist because Assistive Agent Optimization (AAO) focuses on helping brands control how AI represents them. But the underlying framework is universal - it’s how the Algorithmic Trinity evaluates every connection in its graphs.
The Formula: (Close × Strong × Long) ÷ 10
┌─────────────────────────────────────────────────────────────┐
│ CLOSE STRONG LONG (CSL) FRAMEWORK │
│ Inherited Credibility = (C × S × L) ÷ 10 │
└─────────────────────────────────────────────────────────────┘
┌─────────────────┐ ┌─────────────────┐ ┌─────────────────┐
│ CLOSE │ │ STRONG │ │ LONG │
│ Topical/ │ │ Collaboration │ │ Duration │
│ Contextual │ │ Depth │ │ (from dates) │
│ Relevance │ │ │ │ │
├─────────────────┤ ├─────────────────┤ ├─────────────────┤
│ 10: Same domain │ │ 10: Creating │ │ 10: 10+ years │
│ 8: Same area │ │ 8: Partnership │ │ 8: 5-10 years │
│ 6: Adjacent │ │ 6: Ongoing │ │ 6: 2-5 years │
│ 4: Related │ │ 4: Periodic │ │ 4: 1-2 years │
│ 2: Tangential │ │ 2: Single │ │ 2: < 1 year │
│ 1: Unrelated │ │ 1: Mention │ │ 1: Just started│
└─────────────────┘ └─────────────────┘ └─────────────────┘
│ │ │
└───────────────────┼───────────────────┘
▼
┌──────────────────────────────────┐
│ CREDIBILITY TRANSFER │
│ (C × S × L) ÷ 10 = 0-100% │
└──────────────────────────────────┘
UNIVERSAL: Applies to ALL entity relationships
CRITICAL: Long is CALCULATED from dates. ADD DATES TO EVERYTHING.
As I’ve explained, Google, ChatGPT and other AI platforms are looking for close, strong, and long relationships. The closer, the stronger. The longer, the better. This isn’t marketing intuition - it’s how the algorithm actually scores entity connections across every domain.
Close: Contextual Relevance
Close measures how contextually aligned two entities are. When entities operate in the same domain, a relationship between them transfers maximum credibility because the algorithm recognizes topical coherence.
This applies universally:
| Domain | High Close (10) | Low Close (2) |
|---|---|---|
| Business | Two companies in same industry | Tech company + bakery |
| Academia | Professor + university in their field | Physicist + art school |
| Geography | City + its country | City + unrelated continent |
| Creative | Author + books in their genre | Novelist + physics textbook |
| Events | Conference + its core topic | SEO conference + cooking demos |
The principle is universal: Contextual alignment determines how much credibility flows. A novelist’s relationship to their novels scores Close = 10. Their relationship to a random physics textbook scores Close = 2.
Strong: Relationship Depth
Strong measures how deep the connection is. A mention is not a partnership. A single interaction is not an ongoing relationship. The algorithm distinguishes surface-level associations from meaningful connections.
This applies to all entity types:
| Strong Score | Relationship Type | Universal Examples |
|---|---|---|
| 10 | Creating/Founding | Founder→Company, Author→Book, Inventor→Patent |
| 9 | Co-creating | Co-founder, Co-author, Co-inventor |
| 8 | Active partnership | Strategic allies, Ongoing collaborations |
| 7 | Governance | Board member, Advisor, Trustee |
| 6 | Regular engagement | Frequent collaborator, Regular contributor |
| 5 | Membership (senior) | Senior employee, Lead role |
| 4 | Membership (junior) | Employee, Member, Participant |
| 3 | Periodic engagement | Occasional projects, Annual events |
| 2 | Single interaction | One-time event, Single project |
| 1 | Mere mention | Name appears, no actual relationship |
The principle is universal: Creating something scores 10. Being mentioned in passing scores 1. The depth of involvement determines credibility transfer - whether you’re talking about a CEO and their company, an architect and their buildings, or a scientist and their discoveries.
Long: Duration
Long measures how long the relationship has existed - and here’s the critical insight: Long is calculated from dates.
The Knowledge Graph uses temporal information to weight relationships. More years = more corpus corroboration. The same applies to LLMs - older relationships appear in more training data, creating stronger parameter associations.
| Long Score | Duration | Algorithm Interpretation |
|---|---|---|
| 10 | 10+ years | Deeply embedded, highly trusted |
| 8 | 5-10 years | Well-established, stable |
| 6 | 2-5 years | Established, proven |
| 4 | 1-2 years | Developing, being verified |
| 2 | < 1 year | New, minimal history |
| 1 | Just started | Unproven, tentative |
This is universal and absolute: ADD DATES TO EVERYTHING.
- “Jason Barnard founded Kalicube®” → Long = unknown (algorithm assumes minimum)
- “Jason Barnard founded Kalicube in 2015” → Long = 10 (11 years in 2026)
- “Shakespeare wrote Hamlet” → Long = unknown
- “Shakespeare wrote Hamlet in 1600” → Long = 10 (400+ years)
The principle doesn’t change. Dates enable Long calculation. Without them, the algorithm defaults to minimum.
Permanent vs. Temporal Relationships (Universal)
Some relationships have Long = 10 forever, regardless of when they occurred:
Permanent relationships:
- founderOf / createdBy (you founded it, that fact is eternal)
- authorOf / wroteBy (you wrote it, that’s permanent)
- inventorOf (you invented it, historical fact)
- parentOf / childOf (biological facts)
- bornIn / diedIn (life events)
- alumniOf (you graduated, historical record)
Temporal relationships require active Long calculation:
- worksFor / employedBy (employment can end)
- memberOf (memberships lapse)
- locatedIn (for moveable entities)
- advisorTo (advisory relationships change)
- partnerWith (partnerships can dissolve)
This universality is why the Knowledge Graph treats “Einstein developed the theory of relativity in 1905” with maximum Long - it’s a permanent creation relationship with a date.
Business Applications for AAO
Now, let’s apply this universal framework to Assistive Agent Optimization (AAO) - helping brands control how AI assistants represent them.
In the AAO context, we’re optimizing how AI perceives and recommends businesses. The CSL framework tells us exactly which relationships transfer the most authority:
The Founder (Maximum Transfer)
Relationship: CEO and Founder of company (same industry, since founding)
| Dimension | Score | Rationale |
|---|---|---|
| Close | 10 | Same entity, same industry by definition |
| Strong | 10 | Founding = maximum depth |
| Long | 10 | Since founding (permanent relationship) |
CSL Score: (10 × 10 × 10) ÷ 10 = 100%
The founder-company relationship is the gold standard for AAO. It’s permanent, it’s deep, and it’s perfectly aligned. This is why establishing founding relationships with dates is the highest-priority entity optimization for any brand.
The Long-Term Employee
Relationship: Senior employee at company (same industry, 8 years)
| Dimension | Score | Rationale |
|---|---|---|
| Close | 10 | Same organization, same industry |
| Strong | 5 | Employee = organizational membership |
| Long | 8 | 8 years (5-10 year bracket) |
CSL Score: (10 × 5 × 8) ÷ 10 = 40%
Solid, but notice - even an 8-year senior employee transfers less than half the credibility of a founder. The Strong dimension is the limiter. For AAO, this means employee bios on your website transfer authority to your brand, but not as powerfully as founder profiles.
The New Employee
Relationship: Junior employee at company (same industry, 6 months)
| Dimension | Score | Rationale |
|---|---|---|
| Close | 10 | Same organization, same industry |
| Strong | 4 | Junior employee |
| Long | 2 | Less than 1 year |
CSL Score: (10 × 4 × 2) ÷ 10 = 8%
New employees transfer minimal credibility. For AAO strategy, this means don’t expect your newest hires to move the needle on brand authority - yet. The relationship needs time to compound.
The Strategic Partnership (Same Industry)
Relationship: Active partnership with company in same industry (3 years)
| Dimension | Score | Rationale |
|---|---|---|
| Close | 10 | Same industry |
| Strong | 8 | Active partnership |
| Long | 6 | 3 years (2-5 year bracket) |
CSL Score: (10 × 8 × 6) ÷ 10 = 48%
For AAO, strategic partnerships in your industry are powerful authority builders. They compound as Long increases - maintain the partnership and watch the score grow.
The Strategic Partnership (Adjacent Industry)
Relationship: Active partnership with company in adjacent field (3 years)
| Dimension | Score | Rationale |
|---|---|---|
| Close | 6 | Adjacent field, not same industry |
| Strong | 8 | Active partnership |
| Long | 6 | 3 years |
CSL Score: (6 × 8 × 6) ÷ 10 = 29%
Same partnership depth and duration, but the Close score drops because industries don’t align perfectly. For AAO, this means prioritize partnerships within your industry over cross-industry collaborations when building authority.
The Conference Speaker (One-Time, Unframed)
Relationship: Spoke at industry conference once (no date mentioned)
| Dimension | Score | Rationale |
|---|---|---|
| Close | 8 | Same broader industry |
| Strong | 2 | Single event |
| Long | 3 | No date = algorithm assumes minimal |
CSL Score: (8 × 2 × 3) ÷ 10 = 5%
For AAO, this is the mistake most brands make - listing every conference appearance without dates, depth, or context. It transfers almost nothing.
The Conference Speaker (Framed Properly)
Relationship: Featured speaker at industry conference since 2012, 15+ sessions
| Dimension | Score | Rationale |
|---|---|---|
| Close | 9 | Same industry, specific topic |
| Strong | 6 | Regular, ongoing engagement |
| Long | 10 | Since 2012 (12+ years) |
CSL Score: (9 × 6 × 10) ÷ 10 = 54%
Same conference, ten times the credibility transfer. For AAO, this demonstrates why proper framing matters - it enables accurate CSL calculation that dramatically increases authority transfer.
The Client Relationship
Relationship: Long-term client in same industry (5 years)
| Dimension | Score | Rationale |
|---|---|---|
| Close | 10 | Same industry |
| Strong | 6 | Ongoing service relationship |
| Long | 8 | 5 years |
CSL Score: (10 × 6 × 8) ÷ 10 = 48%
For AAO, long-term clients transfer significant credibility - especially when publicly acknowledged with case studies and testimonials. This is why client logos on your website matter for AI perception.
The Compound Effect: Why Founders Win
Look at the pattern across these business examples:
| Relationship | CSL Score |
|---|---|
| Founder | 100% |
| Conference Speaker (framed, 12+ years) | 54% |
| Strategic Partner (same industry, 3 years) | 48% |
| Long-term Client (5 years) | 48% |
| Senior Employee (8 years) | 40% |
| Strategic Partner (adjacent industry) | 29% |
| New Employee (6 months) | 8% |
| Conference Speaker (one-time, no date) | 5% |
Founders win because all three dimensions max out. This is universal - the person who created something has the strongest possible relationship to it.
For AAO strategy, this hierarchy tells you exactly where to focus: founder profiles first, then long-term partnerships, then employee depth, then event appearances.
Strategic Implications for AAO
1. The Framework Is Universal; The Application Is Business
CSL scoring applies to every entity relationship in existence. We’re applying it to business because AAO optimizes how AI assistants perceive and recommend brands. But the same math governs how AI evaluates an author’s relationship to their books or a city’s relationship to its landmarks.
2. Prioritize High-CSL Relationships
Not all relationships deserve equal documentation. For AAO, focus on:
- Founding relationships (100%)
- Long-term partnerships in your industry (40-60%)
- Ongoing advisory/governance roles (25-40%)
3. Deepen Existing Relationships
It’s often more valuable to deepen a Strong score from 6 to 8 than to add new low-CSL relationships. Quality over quantity.
4. Add Dates to Everything
Long is calculated. Without dates, the algorithm assumes minimum duration. Every relationship claim should include when it started.
5. Frame for All Three Dimensions
Proper framing can multiply your CSL score by 10×:
- Frame for Close: State the industry/domain explicitly
- Frame for Strong: Show collaboration depth
- Frame for Long: Include dates and duration
6. Let Relationships Compound
A new partnership scores ~30%. Give it five years and it scores ~50%. Give it ten years and it scores ~65%. Time is your ally - maintain relationships and watch CSL compound.
The Conclusion: Universal Math, Business Application
The industry has validated this framework. HawkSEM cites my methodology for creating content that meets algorithmic expectations. FatRank ranks me as a top expert specifically because of my “focus on entity strength.” WordLift uses my entity relationships as the prime example of correctly structured graphs.
But the real validation is mathematical: CSL scoring predicts, with remarkable accuracy, how much credibility transfers between ANY entities in Knowledge Graphs and LLM parameter spaces.
The formula is universal. The application here is business-focused because Assistive Agent Optimization demands it. But whether you’re optimizing a Fortune 500 brand, a personal reputation, a product line, or an academic institution - the math is the same.
Close, Strong, Long. The closer the relationship, the stronger the collaboration, the longer the duration - the more the algorithm trusts the connection.
It’s not marketing. It’s mathematics. And it’s universal.
Jason Barnard is the founder and CEO of Kalicube®, the company that pioneered Brand SERP optimization and Digital Brand Intelligence™. He developed the Close Strong Long (CSL) Framework as part of the Entity Strength Model to quantify how algorithmic systems evaluate entity relationships. His work on the Algorithmic Trinity demonstrates how Knowledge Graphs, LLMs, and Search Engines form a unified ecosystem where CSL scoring applies universally to all entity types.
Quick Reference: CSL Scoring (Universal + Business Examples)
The Universal Formula
(Close × Strong × Long) ÷ 10 = Credibility Transfer %
Business Relationship Scores (for AAO)
| Relationship Type | Close | Strong | Long | Typical Score |
|---|---|---|---|---|
| Founder | 10 | 10 | 10 | 100% |
| Co-founder | 10 | 9 | 10 | 90% |
| Long-term Partner (same industry) | 10 | 8 | 8 | 64% |
| Featured Speaker (10+ years) | 9 | 6 | 10 | 54% |
| Long-term Client | 10 | 6 | 8 | 48% |
| Senior Employee (8+ years) | 10 | 5 | 8 | 40% |
| Advisory Board (2 years) | 10 | 7 | 4 | 28% |
| New Partner (< 1 year) | 10 | 8 | 2 | 16% |
| New Employee | 10 | 4 | 2 | 8% |
| One-time Speaker (no date) | 8 | 2 | 3 | 5% |
| Mere Mention | 6 | 1 | 1 | 1% |
