The Death of the “Seat License”
Why Your Per-User Software Model is Killing Your ROI For decades, enterprise software has been sold like office furniture: one chair for each person, one login for every employee. The per-user seat license model became so common that most organisations stopped questioning it. It just became part of doing business. But we are no longer in that era. Today, AI agents can resolve customer queries, qualify leads, update records, and trigger complex workflows without any human involvement. In this environment, paying for seat licenses is more than just an old habit; it acts as a tax on your growth. If you still measure your organisation’s digital maturity by the number of user logins you have bought, you are not leading a transformation. Instead, you are supporting an outdated system and paying your software vendor extra for it. This paper exhorts decision-makers who have the authority and responsibility, to think differently. The Ghost in the Machine: Paying for Idle Capacity Walk through the CRM dashboard of almost any mid-sized Indian enterprise, and you’ll find the same uncomfortable truth: a significant percentage of active licenses belong to what we call ‘Ghost Seats,’ viz. logins assigned to users who perform little more than basic data entry, occasional lookups, or routine ticket updates. They’re not power users, neither do they drive insights. They are, in the language of enterprise software procurement, ‘expensive passengers.’ Consider for example, a Pune-based regional insurance aggregator with a 200-seat CRM deployment. Of those 200 licenses, analysis reveal that fewer than 60 users are genuinely engaging with the system in ways that justify the per-seat premium. The remaining 140 seats exist because the process demands human touchpoints, not because those touchpoints create value. This is the perverse logic of the per-user model: it acts as a structural kickback to the software vendor that penalizes your internal process improvements. The more you optimize, the more you pay for empty capacity. The more people you hire to manage inefficiency, the more you pay your software vendor. Scaling your team to handle query volume becomes indistinguishable from scaling your software bill. The incentive to automate or genuinely fix the underlying process is structurally undermined by a pricing model that benefits from your inefficiency. True ROI demands a fundamental reframe: from paying for access to paying for outcomes. AI Doesn’t Need a Login, It Needs Agency The incumbent software vendors have a predictable response to the rise of AI: they wrap AI features around the existing seat-license model and present it as innovation. Every AI interaction, they argue, should still be tied to a human seat. The AI is the co-pilot. The human remains in the chair. Really? This perspective is not a product philosophy; it is a revenue protection strategy designed to keep seat counts artificially high even as your team’s workload is absorbed by an AI layer. It is an attempt to force an autonomous future into a legacy billing cage. The seat-license model is fundamentally incompatible with what we call the ‘Autonomous Enterprise,’ which is an organisation where AI agents handle high-volume, rules-based interactions end-to-end, without requiring a human to observe, approve, or close each transaction. The evidence is already visible in the Indian market. A leading NBFC in Mumbai deployed a conversational AI solution on WhatsApp to handle loan status enquiries, EMI reminders, and document collection requests. Within three months, over 75% of inbound customer interactions were being resolved without a human agent touching the conversation. Customer satisfaction scores improved. Average handling time collapsed. And yet, under a traditional seat-license model, this organisation would still be paying for every human agent whose workload had effectively been absorbed by the AI layer. The question is not philosophical; its financial. When your AI resolves 80% of interactions autonomously, why are you paying for a desk that sits empty? The future belongs to consumption-based or outcome-linked pricing models where you pay for the value created, not the number of humans watching a screen. Forward-thinking organisations are already demanding this shift in their vendor conversations. The Scaling Penalty: Why Growth Shouldn’t Cost This Much Here is the equation that no CFO should accept without scrutiny: in a legacy per-user model, your software costs scale linearly with your headcount. Double the team, double the bill. Triple the query volume, hire more agents, pay more licenses. This is what we call the Scaling Penalty; it is one of the most significant, yet least-discussed constraints on enterprise growth in India today. An AI-first architecture breaks this equation entirely. An AI-first architecture breaks this equation by enabling non-linear growth. While legacy models demand that a 3x surge in volume results in a 3x surge in costs, a sentient system allows your operational throughput to scale dramatically while your software bill remains flat-lined. A fast-growing D2C brand based in Bengaluru offers a useful illustration. Facing a 300% surge in customer queries during festive season sales, their traditional support model would have required proportional hiring, and proportional license costs to manage the volume. Instead, having invested in an AI-first CRM architecture and conversational automation layer, they absorbed the surge with no additional headcount and no emergency license purchases. Their cost-per-resolution fell sharply even as volume spiked. This is not a future scenario. It is happening now, in Indian businesses, across sectors. By decoupling business growth from license growth, organisations transform their CRM infrastructure from a fixed, headcount-driven overhead into a genuinely scalable profit engine. The technology becomes an asset that appreciates in value as it is used more, not a liability that grows more expensive as the business grows. From “User Adoption” to “System Architecture” There is a telling symptom of the seat-license era that most technology leaders will recognise immediately: the ‘user adoption programme.’ Organisations spend millions in consulting fees, change management initiatives, training budgets, and internal communications — trying to compel their people to log into and properly use complex CRM systems. Entire project workstreams are devoted to finding out whether employees are clicking the right buttons