B2B SaaS Content Marketing: How Vyrill Turned Video Search Into a Commerce Engine

The $25 Million Problem Nobody Talks About

“BMW spends $25 million every year making video and the video ROI is very poor. It’s very low.”

That line, from Ajay — Founder & CEO of Vyrill (operating as Viral) — is the most expensive content marketing problem most B2B SaaS and e-commerce leaders have quietly accepted as normal. Brands pour capital into video production, distribute it across channels, and then watch 10–15% of site visitors actually consume it. The other 85–90% leave without ever engaging.

Ajay brings 18+ years of e-commerce experience to this problem, including a prior exit — selling an in-store shopping app company out of Boston. At Vyrill, he set out to solve what he calls the foundational gap in modern video commerce: 95% of the world’s video is not searchable. That single constraint kills personalization, destroys attribution, and caps ROI before a brand ever publishes a frame.

This episode is not about how to make better videos. It is about building the infrastructure that makes every video you already have — UGC, reviews, brand content — work as a revenue asset. The implications stretch across B2B SaaS content marketing strategy, e-commerce, and any growth team trying to tie content investment to pipeline.

Key Takeaways

• 95% of video content is unsearchable — which means it cannot be personalized, attributed, or optimized for conversion regardless of production quality.
• Video search is the prerequisite to hyperpersonalization — you cannot surface the right video to the right shopper without knowing what is inside each asset at a granular level.
• The real PMF was commerce, not insights — early enterprise customers repeatedly rejected a data-insights SaaS model; the pivot to revenue attribution unlocked actual sales conversations.
• Search is the stickiest SaaS integration — once embedded in an e-commerce store, video search becomes the entry point for shopper journeys and enables full purchase attribution.
• Human-generated video is appreciating in value as AI content floods the market — brands that build authentic UGC pipelines now have a compounding trust asset competitors cannot replicate with synthetic content.
• TikTok Shop’s $10B US revenue year signals the consumer behavior shift — video commerce is not a future trend; it is the current buying interface for a growing segment of consumers.
• Custom AI models beat LLMs for video analysis at scale — proprietary models are cheaper, faster, and more precise for domain-specific video intelligence tasks than general-purpose LLMs.

Deep Dive: How Video Search Becomes a B2B SaaS Growth Engine

Why 95% of Video Produces Zero ROI — and What That Actually Means

When Ajay says 95% of video is not searchable, he is not talking about SEO metadata or YouTube titles. He is pointing at a structural capability gap: the content inside the video — the spoken words, on-screen text, scenes, sentiment, faces, product attributes — is invisible to any downstream system.

Without that content layer, you cannot route the right video to the right user. You cannot moderate it for brand safety. You cannot analyze it for competitor intelligence. And you cannot connect a shopper’s intent signal to a relevant piece of content in real time.

“95% of world’s video today is not searchable. Hence, it’s not very useful.” — Ajay, Founder & CEO, Vyrill

The 10–15% video consumption rate on e-commerce sites is a downstream symptom of this problem, not a creative or distribution failure. Brands like BMW are not getting poor video ROI because the videos are bad. They are getting poor ROI because the videos are functionally invisible to both shoppers and the brand’s own systems.

This is the foundational insight that shapes every decision Viral makes as a platform — and it is the first principle any B2B SaaS content marketing team should internalize when auditing their own video spend.

The Four Cs Framework: Capture, Curate, Connect, Convert

Ajay’s operational model for video commerce runs through what he calls the Four Cs: Capture, Curate, Connect, and Convert. This is the architecture for turning video from a cost center into a revenue attribution system.

Capture means ingesting video at scale — from Instagram, YouTube, TikTok, live events (Viral collected 2,000 videos at a single Chicago Bulls game via QR code activation), and direct brand uploads. The volume problem is real: Porsche had 6,000 videos that would have taken 11 months to review manually. Viral’s AI processed them in one week.

Curate is where the proprietary AI runs. Viral analyzes each video across 18 dimensions — video frames, audio, on-screen text, images, and transcription — extracting demographic data, scene context, sentiment, topic classification, and brand safety signals. This is not an LLM wrapper. Ajay is explicit:

“We have built our own tools and our own AI. We have our own models not LLM. So we’re one cheaper than what you’d pay an LLM to analyze a video and do this at scale.”

That cost structure matters enormously when you are processing hundreds of thousands of videos for 500,000 TikTok Shop merchants.

Connect is the search and personalization layer — the step that turns analysis into shopper experience. Once you know what is in every video, you can match a shopper’s explicit query or inferred attributes (device, browsing history, size, gender, purchase intent) to the most relevant 1–2 videos instead of presenting a wall of 35 options.

Convert is full purchase attribution. Because video search is embedded into the site’s shopper journey, every click-through and subsequent purchase is traceable back to the specific video that initiated the session.

Why Video Search Is the Stickiest SaaS Integration

One of the most tactically useful observations in this conversation is about what makes video search uniquely defensible as a SaaS product — relevant to any B2B SaaS SEO strategy built around content assets.

“Search is a SaaS product, right? Once you integrate search into your e-commerce store, how do shoppers begin their shopper journey? They usually start with search. So search is extremely sticky and it works and you can do full attribution with search.”

This is a product positioning insight as much as a GTM one. Search is not a feature — it is the entry point for intent. When a brand embeds video search into their store, they are not adding a widget. They are rerouting the shopper’s first action through a system that can be measured, optimized, and attributed.

For B2B SaaS founders thinking about content marketing ROI, the parallel is direct: owned search (SEO-optimized content, video, tools) that captures intent at the first touch is the same sticky, attributable system. The brands and companies winning in 2026 are those who have built infrastructure — not just assets.

The PMF Pivot: From Insights to Commerce

One of the most instructive parts of this conversation is not about the product that works — it is about the product that almost worked but didn’t.

Viral’s original positioning was around video insights: delivering research-grade analysis of UGC and video reviews to enterprise brand teams. The pitch resonated. The meetings went well. And then, consistently, the same thing happened at the end of every enterprise conversation:

“They would leave the conversation and say, ‘But by the way, if you helped us at the end of the day, if you showed us how you can drive revenue and how you can drive customer acquisition and conversion, that’s all they care about.’”

Four or more enterprise customers said a version of the same thing. That is not noise — that is signal. The product was solving a real problem, but not the problem that controlled the budget. Brand insight teams were interested. Revenue teams controlled the spend. And revenue teams wanted attribution, not analysis.

The pivot Ajay describes follows a recognizable product-market fit discovery pattern:

1. Recognize the repeating objection across multiple conversations (not one outlier)
2. Identify whose budget you are actually competing for (insights team vs. commerce team)
3. Rebuild the product around the revenue motion (embedded commerce, not standalone reports)
4. Validate with a new ICP (Shopify merchants on a SaaS billing model vs. enterprise project contracts)
5. Expand use cases only after commerce is proven

For any B2B SaaS founder who has been collecting enthusiastic meetings that never close, this pattern is worth sitting with. The question is not whether people like your product. It is whether the person who likes it controls the budget.

Human-Generated Content Is Appreciating — Here Is Why That Changes Your Content Strategy

The flood of AI-generated content is producing an unexpected market effect: authentic human video is becoming a scarcer, more valuable signal. Ajay is direct about it:

“Human generated content is becoming even more valuable than AI generated content. The last thing you want is to see an AI generated for sure. Everybody’s a little sick of it already and it just started.”

This has immediate implications for content marketing leaders. Brands that have built UGC pipelines — structured ways to collect, moderate, and activate real customer video — have a compounding asset that AI-saturated competitors cannot replicate with synthetic content. TikTok Shop’s $10 billion in US revenue is not a social media story. It is evidence of a consumer behavior shift toward video-native, peer-validated purchasing.

The infrastructure play is collecting that UGC now, at scale, before the collection window closes. Viral’s QR code activation at live events (2,000 videos from a single Chicago Bulls game) is one mechanism. Systematized post-purchase video review requests are another. The brands that win the next phase of video commerce are building the library today.

ICP-Driven GTM: Two Completely Different Motions

Ajay runs two go-to-market motions simultaneously, and the contrast is instructive for any team managing multiple customer segments.

Enterprise GTM (Porsche, BMW-tier): Purely outbound. Investor introductions. Founder network leverage. White-glove onboarding. Project-based initial engagements with a longer path to SaaS billing. These customers have complex internal approval chains and need proof-of-value before committing to recurring spend.

SMB/Mid-Market GTM (Shopify merchants): LinkedIn lead generation as the primary inbound channel. Product-led components. Self-serve Shopify app integration. Immediate SaaS billing from day one.

“The early customers were purely outbound networking events and intros from our investors and from my network…we do a lot of LinkedIn lead generation which is turning out to be a great channel for us.”

The LinkedIn content motion for SMB is not separate from the product — it functions as a B2B SaaS content marketing engine that educates prospects on video commerce problems before they ever enter a sales conversation. That pre-education shortens the sales cycle and improves conversion because prospects arrive already convinced the problem is real.

About Ajay

Ajay is the Founder & CEO of Vyrill (Viral), a video commerce and AI video analysis platform. He brings 18+ years of e-commerce experience to the company, including a prior exit from an in-store shopping app company based in Boston. Vyrill powers video reviews for TikTok Shop’s 500,000-merchant ecosystem and serves enterprise brands and Shopify merchants with AI-driven video search, moderation, and personalization infrastructure. No company URL was available at time of publication.

Ready to Build a Content Marketing System That Actually Attributes Revenue?

The core lesson from Ajay’s journey is that content — including video — only creates ROI when it is searchable, attributable, and connected to the buyer’s journey at the moment of intent. The same principle applies to every B2B SaaS content marketing and B2B SaaS SEO strategy investment you are making right now. If your content assets are not wired to pipeline and revenue attribution, you are producing the equivalent of BMW’s $25 million video library: impressive to look at, functionally invisible to buyers.

At Rapid Product Growth, we help $2–5M ARR B2B tech companies build content and SEO systems that capture intent, convert qualified traffic, and tie directly to revenue — not vanity metrics. If you are ready to stop producing content that disappears and start building a system that compounds, let’s talk.

Talk to a Growth Strategist →

Frequently Asked Questions

How do you make video content searchable and indexable for e-commerce?

Video becomes searchable by analyzing its audio, text, images, and transcription across multiple dimensions using AI models. Vyrill analyzes videos across 18 dimensions, enabling keyword and semantic search across a brand’s entire video library — turning passive assets into a discoverable, shoppable content layer.

How can brands hyperpersonalize video recommendations to increase conversion rates?

Hyperpersonalization requires knowing what is inside each video first. Once content is analyzed — by topic, sentiment, demographics, and brand safety — you can match shopper attributes like gender, size, and purchase intent to the most relevant 1–2 videos, replacing a wall of 35 options with precision delivery.

How do you attribute video views to purchase behavior and ROI?

Search-integrated video platforms enable full attribution because shoppers begin their journey at the search bar. Every click, view, and subsequent purchase is tracked back to the specific video that initiated the session, giving brands direct evidence of which video content drives revenue — not just engagement.

What is the difference between a video insights platform and a video commerce platform?

A video insights platform delivers analysis and reporting to research teams. A video commerce platform embeds video search and personalization directly into the shopper journey, tying video consumption to purchase conversion. Enterprise buyers, as Ajay discovered across multiple sales cycles, fund commerce outcomes — not insights reports.

Why is human-generated video content becoming more valuable in 2026?

AI-generated content volume is increasing rapidly, making authentic human video a scarcer trust signal. Consumers are already fatigued by synthetic content. Brands with structured UGC pipelines — real customer video collected, moderated, and activated at scale — hold a compounding advantage that AI production cannot replicate.