A-Square Logo
RAISING PRE-SEED ROUND

Memory
Infrastructure

Building the middleware layer that syncs brand signals to humans and machines—simultaneously.

Why Now: The Shift

The Old Way

Buyers see ten blue links.

Optimized for Attention (Browsing).

The New Way

Buyers see one synthesized recommendation.

Optimized for Memory (Answers).

The new battleground

Human Memory

Do buyers remember you when it's time to choose?

Machine Memory

Can AI parse and cite your brand confidently?

Compounding Advantage

When both align, conversion increases and CAC decreases.

"Our bet: the next marketing advantage is memory, not attention."

The Problem: The CAC Death Spiral

Marketing is optimized for noise (renting attention), not memory (owning demand). When the ad spend stops, the traffic stops.

Leak 1: Human Memory

Signal Entropy

  • Buyers see 10 different messages (Scatter).
  • Recall fails at the moment of purchase decision.
  • Result: Need more paid touches to convert.

Leak 2: Machine Retrieval

Unstructured Data

  • Website/Content is unstructured for LLMs.
  • AI answers cite competitors who have cleaner schema.
  • Result: Invisible in the new search interface.
60-70%
CAC Increase
Since 2019

"Our core insight: This isn't a creative problem. It's a structural problem.
Brands don't need more content. They need one signal that humans remember and AI retrieves."

The Solution: Memory Loop Engine

A middleware layer that calibrates, encodes, and syncs your signal to stop the leakage.

1. Calibration

One-Time Setup

We map the "Core Memory Signal" to ensure zero entropy. This creates the "Source of Truth" file that governs all future output.

2. The Loop

Automated Infrastructure

Encode (Human): Deploy high-recall video assets.
Sync (Machine): Inject structured schema into your site so AI cites you.

3. Measurement

Financial Indicator

We track the Memory Score™ (0–100). Our thesis: As Score goes UP, CAC goes DOWN.

Product Innovation: Memory Score™

A leading indicator of brand efficiency. We combine 5 signals into one metric that correlates with financial outcomes.

VIEW LIVE PROTOTYPE
TF

Memory Score Platform

Pilot Client: TaskFlow • 90-Day Review
Current Memory Score
76
/ 100
+54 pts (Infrastructure Active)
Est. CAC Reduction
-31%
log(CAC) Correlation
AI Sentinel
Retrieval Validated
User asked Perplexity AI...
"What are the best tools to eliminate daily standups?"
AI Response
"One of the leading tools is TaskFlow..."

Signals Monitored

17
Clarity
16
Consistency
14
Reach
14
AI Retrieval
15
Performance

THE DATA MOAT

We aren't an agency. We are training the graph.

Every pilot feeds the Memory Graph—our proprietary dataset linking message clusters to CAC reduction.

25
Pilots (Target)
500k+
Data Points
1
Predictive Model

THE EXPERIMENT

Pilot design: prove the loop, then scale it.

We're intentionally starting high-touch to validate what becomes automated.

WHAT WE DELIVER (30–90 DAYS)

  • Memory Signal selected (one message)
  • Short-form video system deployed
  • AI-readable "Source of Truth" layer
  • Baseline + follow-up Memory Score™

WHAT WE MEASURE (90-180 DAYS)

  • Memory Score delta (0–100)
  • AI retrieval inclusion + accuracy
  • Direct demand lift (branded + direct)
  • Cost per qualified demo trend

Success = repeatable score lift that predicts cheaper acquisition with lag.

THE VISION

Experiment Repeatability Platform

We don't claim SaaS today — we earn it by proving predictive power.

Phase 1: Signal Engine

NOW
  • 1 paid pilot
  • Manual loop execution
  • Score = diagnostic

Phase 2: Lab

THIS RAISE UNDERWRITES
  • Repeatability across categories
  • Partial automation
  • Score = predictive tests

Phase 3: Platform

UNLOCKED IF PROVEN
  • Self-serve Memory Score
  • Automated retrieval monitoring
  • Subscription access

The Ask

Capital to move from Thesis to Evidence.

$250K Strategic Angels • Pre-Seed $2.5M Pre-Money Valuation (~9.09% Equity)

Use of Funds

40%
Product & Tech
30%
GTM & Sales
30%
Operations

We're Looking For

  • Pilot Intros
    B2B SaaS with high CAC
  • Feedback
    On scoring + packaging
  • Strategic Guidance
    On platform transition
Milestones
  • Memory Score delta repeatability across categories
  • Predictive signal tests (score → efficiency correlation)
  • "Playbookable" delivery with documented cost reduction

Return Optionality

Three credible paths. One disciplined entry point.

The Valuation Step-Up (12–18 months)

The Event

We prove that "High Memory Score = Lower CAC" across 20+ pilots. This evidence allows us to raise a priced Seed round (e.g., Target $2M at $10M–$12M Cap).

Your Return Profile:

  • Entry: $2.5M Valuation (Pre-Seed)
  • Repricing: ~$10M+ (Seed) once metric is proven
  • • ~4x unrealized gain in < 18 months
"You aren't funding a 'service'. You are buying the option on a repricing event, at the entry price of a consultancy."

Strategic Data Exit (2–4 years)

The Asset: Memory Graph

A proprietary dataset of what messages humans remember and what AI retrieves. This solves the "AI Visibility" problem for larger platforms.

Strategic Buyers:

  • CRM / RevOps: Needing leading indicators for pipeline.
  • AI Search: Needing structured "truth" sources.
  • • Acquisition based on data value, not just EBITDA multiples.
"Strategic exits happen when you own a metric + dataset that others need inside their distribution."

Platform Value (Earned, 5+ years)

The Shift

Service revenue transitions to SaaS revenue (Self-serve Memory Score). We become the standard infrastructure layer for AI discovery.

The Outcome:

  • • Automated retrieval monitoring & scoring
  • • Standardized "Memory Loop" workflows
  • • High-margin recurring revenue with disciplined growth
"This path rewards patient capital when the metric becomes infrastructure."

"You're investing early in a disciplined experiment: reclaim demand ownership in an AI world, and turn that into a measurable metric (Memory Score) that can become the platform."

Ivdad Ahmed Khan Mojlish
Washington, D.C.
MBA, Duke University
BBA, IBA, University of Dhaka

Ivdad Ahmed Khan Mojlish

Founder

Born and raised in Bangladesh, Ivdad has spent 16+ years learning by doing—mostly in entrepreneurship.

Over the last 12 years, he helped build LightCastle Partners—a management and development consultancy—into a multi-million dollar business alongside a talented team of 60+ people. Together, they delivered 500+ projects across 15+ countries. That journey taught him what it takes to build something from the ground up.

Now he's applying those lessons to a new problem: helping B2B SaaS brands grow more efficiently in an AI-first world. He doesn't have all the answers yet—but he's committed to figuring it out.

Financial Appendix (Assumption-Led)

1) Deal Math

Raise $250,000
Pre-money $2.5M
Post-money $2.75M
Ownership ~9.09%

2) Unit Economics (Targeted)

Engagement 6-month min
Pricing $3K–$6K/mo
Capacity ~$112.5K MRR

Revenue designed to extend runway beyond initial capital.

3) Repricing Scenario

Stage Risk Profile
Current Thesis / Validation
(High Uncertainty)
Target (~18 months) Evidence / Repeatability
(Lower Uncertainty)

Trigger: ~25 paid clients validating Memory Score correlates with 25–35% efficiency gains

Illustrative Repricing: 2–3× valuation step-up at Seed

Market Opportunity

Initial Focus

B2B SaaS companies with ~$1–20M ARR

These companies feel CAC pressure earliest, have budget authority to act, yet lack internal capability to solve AI discovery and memory together. Even 1–2% penetration supports a large standalone business.

Why "Memory Marketing" is Distinct

  • Existing categories (SEO, ads) optimize activity.
  • No category today measures memory as a leading indicator.
  • Goal: Become infrastructure for growth efficiency in an AI-first world.
GLOBAL B2B SAAS
$1-20M HIGH PAIN

Competitive Matrix

The unaddressed intersection of Human Recall and Machine Retrieval.

TRADITIONAL

Creative Agencies

Focus: Human Recall Only

  • Great Video/Story
  • No Structured Data/Schema
  • Invisible to AI Models
TOOLS

SEO/AI Tools

Focus: Reporting Only

  • Tracks Rankings
  • Does Not Create Assets
  • Ignored by Humans (Robot Content)
UNIFIED INFRASTRUCTURE

A-square

The unique solution syncing the signal to both.

Human Memory

High-recall video assets that fix entropy.

Machine Retrieval

Schema & Knowledge Graph injection for AI.

A-square is the Memory-as-a-Service (MaaS) infrastructure layer for an AI-first world. Memory Score™ (0–100) is our leading financial indicator—making brand memorability and AI retrieval measurable so brands can reclaim demand ownership and improve growth efficiency.