How BigAcademy Built a Full US Marketing Operation in 2 Weeks

Q: Do you need technical skills to use Junior for marketing?

No. BigAcademy's founder John Wu is an education expert, not a marketer or engineer. He directed Junior entirely through plain-language Slack messages, often in Chinese. Junior handled the technical execution: code, deployment, email infrastructure, and analytics.

TL;DR

BigAcademy.ai had a world-class K-12 literacy product and zero US market presence. In 2 weeks, one non-technical founder used Junior to build the entire go-to-market operation: whitepaper, website, 2,000+ emails, CRM, and GEO articles. No agency, no marketing hire, no engineering team.

In this post:

The Starting Point: Great Product, No Market Presence
Phase 1: Product Deep Dive and Content Foundation
Phase 2: A Full Marketing Website, Built from Scratch
Phase 3: Email Marketing at Scale
Phase 4: GEO Optimization for AI Search Engines
Results: What 2 Weeks Produced
How the Collaboration Worked
Best Practices from This Case Study
FAQ

The Starting Point: Great Product, No Market Presence

John Wu is the co-founder of BigAcademy.ai, an AI-native K-12 literacy platform. The product is serious: a Socratic AI tutor, 20,000+ articles, a 6-trait writing coach, vocabulary system covering 6,934 words across 7 test standards, and 300,000+ families already using it.

But in the US market? Nothing. No whitepaper. No landing page. No lead list. No email campaigns. No SEO presence. No organic visibility on Google or AI search engines.

John is an education expert. He understands his product and his users deeply. He is not a marketer, not a designer, and not an engineer. Hiring a marketing agency would mean weeks of briefings, misaligned deliverables, and a feedback loop measured in days. Hiring in-house would take months.

He hired Junior instead. One AI employee. $2,000 per month. Full access to email, browser, SSH servers, file system, and APIs.

Here's what happened over the next 2 weeks.

Phase 1: Product Deep Dive and Content Foundation

Before writing a single word of marketing copy, Junior needed to understand the product from the inside.

Product Audit

Junior logged into BigAcademy, systematically explored every feature, and captured 48 UI screenshots. The audit covered:

Go Endless: Bloom's Taxonomy x Socratic method reading flow
Dotty AI Tutor: Conversational learning assistant
6-Trait Writing Coach: AI-powered essay feedback
Homework Tutor: Subject-specific help
Vocabulary System: 6,934 words across 7 test standards (SAT, ACT, GRE, TOEFL, etc.)

Competitive Analysis

Junior benchmarked BigAcademy against 4 major competitors:

Competitor	Valuation / Acquisition	Revenue	Key Weakness vs BigAcademy
Newsela	$1B valuation	$109M ARR	No AI tutoring, content-only
Achieve3000	Acquired by McGraw Hill	Undisclosed	Legacy platform, no Socratic method
Learning A-Z	Part of Cambium Learning	Undisclosed	No AI writing coach
Epic!	Acquired by BYJU'S for $500M	Undisclosed	Entertainment-focused, weak on assessment

The killer insight: BigAcademy's "Go Endless" feature, which combines Bloom's Taxonomy with the Socratic method, was a genuinely unique differentiator. That became the foundation of the entire narrative.

The Whitepaper

Junior produced "The 2 Sigma Breakthrough," a full evidence-based whitepaper for school administrators. This was not a marketing flyer. It cited NAEP 2024, PISA 2022, Bloom's original 1984 study, and MIT Media Lab 2025 research. Cormorant Garamond typography. A live bilingual web version at biglanding.apef.site/whitepaper1.

It reads like something a research team spent a month producing.

Phase 2: A Full Marketing Website, Built from Scratch

Junior handled the entire technical stack: Next.js application, nginx web server, SSL certificates, systemd services, SQLite databases, and REST APIs. All deployed on a Tencent Cloud server via SSH.

What Got Built

Page / Feature	Details	Live URL
Landing page	4 languages (EN/ZH/ES/JP), Google Ads ready, FAQ schema markup	biglanding.apef.site
Writing Coach page	Real student essay transformation demo (rough draft to 95/100)	Deep-dive conversion page
Reading Journey page	Go Endless feature walkthrough with real article data	Deep-dive conversion page
School proposals	Interactive pricing calculators, navy + gold design	/proposal/dehong
Lead capture forms	Teacher trial ($199 value) + Parent trial ($19.99 Plus card)	/trial
CRM system	Full-stack: SQLite + REST API + web UI, 4 pipeline tabs	/crm

Every page is live right now. You can visit the full showcase and click every link. Nothing is a mockup.

The CRM Deserves Its Own Mention

Junior didn't just capture leads. It built a complete CRM from scratch with:

Dual pipeline: teachers and parents tracked separately
Status flow: new → sent → replied → applied → converted
Traffic analytics with visitor logs
Email A/B test results embedded in the dashboard
REST API with authentication
Server-side search across 1,421 leads

Most teams spend weeks evaluating CRM software. Junior built one in days, customized exactly to BigAcademy's sales process.

Phase 3: Email Marketing at Scale

Junior built the full email infrastructure end-to-end: lead sourcing, visual HTML templates, 5-route A/B testing, bot-filtered tracking, and 9 campaign waves executed.

Reactivation Campaign

The first campaign targeted 200 existing users. Junior analyzed each user's actual learning data and segmented them into 4 tiers by Lexile score, AR score, reading count, and quiz completion rate.

Each email used the student's real data. Bilingual routing happened automatically by domain detection. 4 waves, 780 emails, every single one personalized.

Wave	Angle	Content
Wave 1	Learning report	Personalized progress stats per student
Wave 2	Peer comparison	Leaderboard ranking vs other students
Wave 3	AI critical thinking	How Socratic AI develops deeper skills
Wave 4	Conversion offer	$99/year PLUS plan upgrade

Lead Sourcing

Source	Leads	Type
Manual research (school directories, social media)	78 teachers + 95 parents	High quality, individually researched
Datazapp	1,248 parents	US homeschool mothers in California + Massachusetts
Instantly.ai prospecting	61,000+ available	Teacher pipeline for future waves
Total in CRM	1,421

A/B Testing Infrastructure

Junior set up 5 email routes per wave, 3 visual templates (dark navy, warm gold, teal), and custom subject lines for each route. A tracking pixel and redirect-based click links measured real engagement.

One critical discovery: 97% of "opens" were bots. Google, Yahoo, and Fastly scanners automatically trigger tracking pixels. Junior built an IP-based bot filter from scratch and switched to click-through rate as the primary metric.

The "Elite School" route led performance at 10% CTR. Live analytics dashboard at biglanding.apef.site/email-stats.

Phase 4: GEO Optimization for AI Search Engines

This is the part most marketing teams haven't thought about yet, and it might be the most forward-looking thing Junior built.

Perplexity, ChatGPT, and Gemini now answer questions directly, citing sources. If your product isn't in those citations, you're invisible to a growing segment of your audience.

The Gap

Junior tested 48 questions across 9 topic clusters. BigAcademy appeared in zero AI search results. Khan Academy, Epic!, and CommonLit dominated everything.

The Solution: 5 GEO Articles

Junior wrote 5 long-form, evidence-based articles specifically structured for AI engine citation:

Article	Target Cluster	Details
Bloom's 2 Sigma Problem	Education theory	40-year challenge, how AI changes it
BigAcademy vs Newsela vs CommonLit	Competitor comparison	Feature-by-feature for "Newsela alternative" searches
Is AI Making Your Child Lazy?	Parent concerns	Socratic AI vs answer-giving AI
Best AI Writing Coach for Kids	Product category	Writers vs checkers vs coaches
Homeschool Reading Curriculum	Buyer guide	Age-based guide with platform comparisons

Each article includes Article + FAQPage schema markup, 5 FAQ items, 9 to 11 minute read time, and honest competitor comparisons. All submitted to Bing via IndexNow. Authentic discussion posts were also prepared for 4 subreddits, 5 Quora threads, and 3 Medium articles.

Results: What 2 Weeks Produced

Category	Output
Content	1 whitepaper, 5 GEO articles, 3 email template sets, social media batches
Web pages	10+ live pages (landing, use cases, proposals, lead capture, CRM)
Email campaigns	2,000+ emails across 9 waves, 0 delivery failures
Lead generation	1,421 leads in CRM across 3 US regions
GEO optimization	5 schema-marked articles, FAQ + Organization schema, IndexNow
Analytics	Cookie tracking, UTM attribution, email pixel, click redirect, bot filtering
Infrastructure	Next.js + nginx + SSL + systemd + SQLite + REST APIs, all GitHub versioned

Everything is live. Everything is real. And everything can be expanded at any time by the same AI employee who built it, with full context intact.

How the Collaboration Worked

The entire 2-week collaboration happened in Slack DMs. John gave direction in plain language, often in Chinese. Junior executed, reported back, and asked when genuinely blocked.

John's role: Identify what matters, review key decisions, say "yes" or "change this." He brought product intuition and education expertise.

Junior's role: Everything else. Research, writing, coding, designing, deploying, sending emails, building analytics, tracking results.

Three things made this work:

The feedback loop was minutes, not days. When John said "this subject line feels too salesy," Junior rewrote 5 variants and resent within the hour. When John said "add a pricing calculator," it was live by the next morning.
No context was ever lost. Junior remembered every decision, every credential, every previous iteration. No re-explaining, no "let me get up to speed" conversations.
The work was visible to the whole team. Other BigAcademy team members could observe the Slack collaboration, creating organizational knowledge alongside the deliverables.

Best Practices from This Case Study

If you're considering using an AI employee for your go-to-market, here's what BigAcademy's experience teaches:

Start with a product audit, not marketing copy. Junior spent time understanding the product before writing anything. The killer feature ("Go Endless") became the narrative foundation. If Junior had skipped this step, the marketing would have been generic.
Let the AI handle the full stack. John didn't try to split work between Junior and other tools or freelancers. One AI employee owned everything from research to deployment. This eliminated handoff overhead and context loss.
Give direction, not specifications. John didn't write briefs or wireframes. He said things like "this feels too salesy" and "we need a pricing calculator." Junior translated that into execution. Your domain expertise is the input; the AI handles the how.
Build infrastructure, not just content. The CRM, A/B testing framework, bot-filtered analytics, and lead capture forms are permanent assets. They keep working and can be expanded. Content without infrastructure is a one-time effort.
Think about AI search engines early. GEO optimization is still new, and most competitors haven't started. That's exactly why it's valuable. The 5 articles Junior built are positioned to capture traffic that will only grow.

FAQ

What did BigAcademy accomplish with Junior in 2 weeks?

BigAcademy's co-founder used Junior to build a complete US go-to-market operation from scratch: a research-backed whitepaper, 10+ web pages, 2,000+ personalized emails across 9 campaign waves, a custom CRM with 1,421 leads, and 5 GEO-optimized articles. All directed through Slack conversations.

Do you need technical skills to use Junior for marketing?

No. John Wu is an education expert, not a marketer or engineer. He directed Junior entirely through plain-language Slack messages, often in Chinese. Junior handled the technical execution: code, deployment, email infrastructure, and analytics.

How does Junior compare to hiring a marketing agency?

Junior operates as a single AI employee with access to your full tool stack. Unlike an agency, there are no handoffs, no briefs, no waiting for deliverables. The feedback loop is measured in minutes. When John said "this subject line feels too salesy," Junior rewrote 5 variants and resent within the hour.

What tools did Junior use to build BigAcademy's marketing?

Junior used SSH access to deploy a Next.js site on Tencent Cloud, built SQLite databases for the CRM, set up SMTP for email campaigns, created A/B testing infrastructure with bot-filtered tracking, and used browser automation for product audits. All orchestrated through Slack.

Can Junior handle multilingual marketing campaigns?

Yes. Junior built BigAcademy's landing page in 4 languages (English, Chinese, Spanish, Japanese), wrote bilingual email campaigns that auto-routed by domain detection, and handled Slack conversations in Chinese while producing English marketing materials.

Ready to see what one AI employee can build for your team? Start with Junior.

See the full interactive showcase at biglanding.apef.site/showcase. Every URL is live, every page is real.