The
Careervira
Story

The Careervira story is really three stories that had to work at the same time. First, a B2C learning marketplace that needed users it could not afford to buy. Then, an engagement problem: the users arrived but did not stay. Then, an enterprise pivot that meant building a second business, with a different buyer, a different product, and a different go-to-market motion, while the first one kept running.

What follows is organised the way the work actually happened: Phase 1 (acquisition), Phase 1B (engagement), Phase 2 (enterprise). Each phase includes the decisions behind the numbers, including the ones where we chose not to do something, and the ones we would sequence differently today.

The product was live. The courses were live. The users weren't coming. We had a marketplace with 200,000+ courses from 70+ global content partners, but no systematic way for search engines, or users, to discover it. Growth was flat, and CAC on paid channels was unsustainable for a company at our stage. The decision was to build an organic engine from first principles.

When I joined Careervira, it was an early-stage MVP: a learning marketplace with strong supply but weak demand discovery. Users who arrived couldn't easily find the right course for their specific goal. Users who didn't arrive had no reason to find us through search. Both problems needed solving simultaneously. One required a product strategy, the other a content architecture strategy, and we deliberately linked the two.

The governing insight was simple but operationally demanding: a 10:1 ratio of content pages to product pages. For every course listing or marketplace page, we needed roughly ten pieces of indexed, high-quality content serving adjacent user intents: role exploration, career path research, topic mastery planning, industry news. This was not content marketing in the traditional sense. It was product architecture designed for search distribution.

The core insight was that professionals searching for career guidance were underserved. They weren't just looking for courses; they were asking layered questions like "what skills do I need to become a Product Manager at a Series B startup in India?" We decided to build the product that answered those questions at scale, and let search do the distribution.

Working closely with the SEO team, we designed a programmatic content architecture across seven product verticals. The SEO team owned technical execution: keyword research, on-page optimisation, schema, indexing strategy. My role was to define the product scope, information architecture, and content depth for each vertical, and to make sure each page served genuine user intent rather than just search volume.

Every page was designed to answer a specific intent at a specific stage of a professional's career journey. The SEO team ensured discoverability; our job was to make sure that once a user arrived, the page was worth their time.

The results were not immediate. SEO compounding takes 6–12 months to show up in traffic analytics. But they were durable. After the architecture was built and indexed, traffic growth sustained with no additional production effort. That was by design: the goal was an asset, not a campaign.

If the content architecture was the slow-burn foundation, Google Web Stories were the accelerant. This was the highest-impact growth initiative of the first phase. Not because it was the obvious move, but because it was the right move at the right moment in the Google ecosystem cycle.

In 2022–2023, the Web Stories format was in active expansion within Google's ecosystem. The algorithm was promoting Story content to incentivise publisher adoption, which meant early movers received disproportionate distribution. We recognised that window early and moved decisively, in two deliberate steps.

Step one, the POC. Before committing a team, we tested a specific question: does career and role-focused story content, a non-traditional use of the format, get surfaced by the algorithm and generate material traffic beyond our existing pages? We built a small batch of stories across two intent categories (Roles and Learn Paths), measured impressions, clicks, and CTR through Search Console, and compared performance against baseline SEO pages on the same topics.

Step two, the scale decision. The POC exceeded expectations across impressions, CTR, and acquisition signals. We moved from test to dedicated function: a separate Web Stories team with defined production targets, an SEO-led distribution strategy, and story-specific content architecture. Roles stories were the strongest category, combining high search intent with a visually compelling format. Career Path stories drove the deepest engagement. Topic explainers earned high impressions but converted weakly, which told us they were awareness assets, not activation assets.

The result: Web Stories became a meaningful traffic driver within weeks of full deployment, far faster than standard SEO content compounds. The format's native placement in Google Discover also reached users who weren't actively searching but were in a discovery mindset, a qualitatively different acquisition vector. Three factors converged: algorithmic tailwind, content-format fit, and first-mover positioning. The window was real and finite, and we moved inside it.

The most significant signal from Phase 1 was not the peak traffic number but the durability of the asset. After the first 12–18 months, the team's attention shifted almost entirely to the B2B product. No new SEO campaigns. No new story production. Traffic held, because the engine had been built as an asset, not a campaign.

Getting users to the platform was Phase 1. Getting them to stay, and come back, was Phase 1B. Traffic without engagement is a vanity metric; the real signal of product-market fit is whether a single session delivers enough value to earn a return visit. We had solved discovery. We hadn't solved retention.

The diagnostic was clear. Users arrived with a learning or career intent, consumed the page they searched for, and left. Session depth was shallow, return rates were low. There was no personalised mechanism pulling a user toward their next relevant piece of content. The platform had information density but lacked an information architecture that served each individual user's journey.

Two products were built in sequence to fix this. Neither was a marketing initiative. They were product initiatives with a product marketing mandate: make the platform feel like it knows you, not like a library where you're on your own.

The Recommendations product solved the "what next" problem: the moment after a user finishes an action and has no obvious next step. Without personalised direction, users defaulted to leaving. With a relevant next step in front of them, they had a reason to continue.

The system drew on context signals (the current page), intent signals (browse history within the session), the role and topic tags attached to prior actions, collaborative filtering, and cohort popularity. It was designed to move four numbers: pages per session, course exploration depth, cross-vertical discovery, and return visit probability. The richness of the role, topic, and career path taxonomy we had built in Phase 1 turned out to be the engine's biggest advantage, because a recommendation system is only as good as the signals feeding it.

Recommendations solved session depth. The Profile with Goals product solved return visits. A user with no profile has no investment in the platform; every visit is transactional. A user with a stated goal, a progress state, and content anchored to that goal has a reason to come back.

Users set a career goal (target role, industry, geography, experience level), a learn goal (skills to master, with proficiency level and timeline), and their current state. That produced a personalised gap analysis, the distance between where the user is and where they want to be, which powered a curated, sequenced feed instead of a generic catalogue. Every recommendation after profile creation was anchored to the user's own stated goal. Progress states created a completion pull. Return prompts were tied to the user's intention, not to platform marketing.

The combined effect was a platform that felt directed rather than browse-only. Users who completed a profile and set a goal behaved differently: deeper sessions, more returns, more course exploration. This cohort was also the most commercially valuable, because a user with a specific goal is far closer to a course purchase decision than one browsing without intent.

B2C had been set in motion: traffic was compounding organically and the engagement products were working. But the commercial model of a B2C learning marketplace at this scale needed either a massive user base, which we were building slowly, or a parallel revenue stream with shorter cycles and clearer ROI visibility. Enterprise was that stream.

The demand signal emerged organically. Organisations that had seen Careervira's marketplace began asking whether the same discovery and recommendation capability could be deployed internally, as a managed learning environment for their employees. That was the wedge. We had the content infrastructure, the recommendation engine, and the course relationships. What we needed was an enterprise product layer: user management, admin controls, group learning workflows, compliance tracking, reporting, an L&D-facing interface. A different product, a different buyer, a different GTM motion entirely.

My role in this phase was dual: helping define product-market fit through customer discovery alongside the product team, while building the go-to-market infrastructure (positioning, messaging, sales enablement, demand generation) that made the product commercially viable. The two could not be decoupled. You cannot sell a product whose positioning hasn't been sharpened by customer conversations, and you cannot have those conversations without the assets that frame them.

Before positioning the B2B product, we had to understand who was buying, what problem they were solving, and what alternatives they compared us against. We spoke with HR Directors, L&D Heads, CHROs, and Talent Development leads at mid-to-large enterprises, plus the line managers and admins who would use the platform daily, and in some cases the employees who would consume the learning. Each persona carried a distinct problem and a distinct role in the decision.

What we heard, consistently: learning administration was too manual and too compliance-driven; there was no way to see what employees actually know versus what they need; existing LMS platforms had low completion and low voluntary usage; and nobody could connect L&D spend to business outcomes.

That discovery changed the product's identity. The initial positioning was "enterprise learning marketplace," a content delivery story. But enterprises were not buying content. They were buying skill intelligence, administrative efficiency, and proof of learning impact. The product was repositioned as an AI-native workforce learning and skill development platform, and every layer of messaging changed with it.

Enterprise L&D technology is one of the most crowded categories in HRTech. We had to be meaningfully distinct from players with years of brand equity, larger sales teams, and deeper relationships, while remaining credible enough to make enterprise shortlists.

The approach was to own a specific position rather than compete on feature parity. We were not the most feature-rich platform. We were the most intelligent one: the system that understood what each employee needed to learn, why, and in what sequence, and surfaced that intelligence in the interfaces HR, managers, and employees actually used.

Against the global LXPs (Degreed, 360Learning, Cornerstone, Docebo), three differentiators held:

1. AI-native, not AI-added. Incumbents bolted AI onto existing platforms. Careervira was built AI-native from the ground up; recommendations, skill mapping, and learning path generation were core architecture.
2. Content marketplace integration. Direct access to 200K+ courses from 70+ global partners, rather than a closed library requiring custom curation.
3. India-built, India-priced. Global platforms were expensive and over-engineered for mid-market Indian enterprises, a segment the global players largely ignored. Against Disprz, the closest India-built comparable, we differentiated on AI depth, marketplace breadth, and the credibility signal of the B2C platform itself.

Messaging then split by persona, because enterprise decisions involve multiple stakeholders with different purchase motivations. The CHRO buys strategic ROI. The HR manager buys operational efficiency. The line manager buys visibility into team capability. The employee buys relevance: a platform that seems to know what they need to learn for where they want to go.

Enterprise HRTech is a considered buying category: 3–9 month cycles, multi-stakeholder decisions, and a buyer who is rarely the user. Demand generation had to reflect that. The job was not volume; it was qualified pipeline at the right buyer stage.

Email became the primary pipeline generator. Targeted, personalised outreach to ICP-matched personas (CHRO, L&D Head, HR Director at 200+ employee companies in tech, BFSI, and manufacturing) with pain-led framing rather than product pitches. "Most L&D platforms track completions, not skill gaps" opened more conversations than any feature list. The discipline was list quality and message precision over volume.

Offline events became the highest-quality channel by unit economics. A single conversation with a CHRO at the right event carried more pipeline value than fifty email opens, because face-to-face compresses what would otherwise be three or four email exchanges, and HRTech buying is heavily relationship-mediated. We capped attendance at one event per month and selected events purely on whether the right buyers would be in the room.

Pipeline without conversion is just cost. The demand motion generated conversations; converting them required different assets and a different capability. For most of this period there was no dedicated sales team, so the product marketing function effectively doubled as the deal progression function.

Three systems carried that load. Layered pitch assets: an executive narrative deck for CHROs, a product demo deck for L&D Heads, and use-case-specific solution narratives, each built from customer discovery and updated quarterly against objection patterns. POC strategy: POCs were the pivotal moment where interest became evaluation, so I led the end-to-end design of each one: use case scoping, success criteria, stakeholder alignment, post-POC narrative. The goal was never to demo features; it was to prove the platform solved the specific pain the account had named. Objection response: the recurring objections ("we already have Cornerstone," "what's your HRMS integration story," "will employees actually use it," "how long is implementation") were documented, categorised, and answered with structured counter-narratives, which removed the deal stall that unanswered objections create in later stages.

The Careervira story is not a B2C story, and it is not a B2B story. It is the story of a platform that ran two distinct businesses simultaneously, each with its own GTM logic, its own user psychology, and its own definition of success, built from scratch by a small team operating across product and marketing at once.

On the B2C side, we proved you can build a 200K+ MAU platform with zero paid acquisition if you architect content as a distribution system rather than a marketing activity. The content products were not collateral. They had product specs, information hierarchies, and user intent mapping. The SEO team executed the distribution; the product and PMM function defined what to build and why it would perform. The results compounded and sustained long after the team moved on.

On the B2B side, we proved that in a considered, high-trust category, product marketing is the commercial function when there is no dedicated sales team. Positioning defined the category we competed in. Messaging determined whether we earned the second conversation. POC strategy determined whether evaluations became deals. Each of these was delivered by a function that also owned product strategy and discovery.