In his 2011 essay “Why Software is Eating the World”, Marc Andreessen predicted that software would transform every part of the economy, one sector at a time. Fast forward 13 years, the same can now be said about generative AI.
The software-as-a-service (SaaS) sector is swamped with innumerable AI chatbots and co-pilots to streamline customer interactions and optimise workflow. The landscape is about to be disrupted once more—this time by AI agents..
AI agents go a step further, performing tasks, making decisions, and taking actions independently, without human intervention. Leveraging machine learning and natural language processing, these agents can engage with their environment, collect data, and apply it to execute tasks autonomously.
This marks a new era of AI tools that surpass the functionality of chatbots. AI agents are capable of managing more complex interactions, drawing insights from customer engagements, and responding to market dynamics. Their role is to automate processes and free up valuable resources for companies.
SaaS giants are already in on the game.
Salesforce , a leader in cloud-based customer relationship management (CRM) solutions, recently rolled out Agentforce to help organisations build and manage autonomous AI agents for everyday business operations across sales, customer service, and marketing. Oracle too is shipping 50 AI agents as part of its Fusion cloud business applications spanning HR, sales, customer service, and quality control.
Indian startups are surfing the same wave, transitioning from pure-play SaaS products to incorporating Gen AI into their offerings.
Intelligent co-workers
Take Ema, short for Enterprise Machine Assistant, which is deploying its AI agents across workplaces to automate mundane tasks. The San Francisco- and Bengaluru-based startup’s no-code AI agents, which it calls “universal AI employees”, can morph into taking any role–from customer support to legal and compliance.
“Building a Gen AI application is like cooking a Michelin star dish at home without prior training. The entire stack is complex and fragmented, with LLMs (large language models) being a piece of it,” Surojit Chatterjee, CEO of Ema tells
Bengaluru-based KOGO AI, which offers platform-as-a-service for enterprises and small businesses, has launched an ‘AI agent marketplace’ designed to automate business operations across sectors. It functions like an app store, housing over 100 AI agents.
“The AI store is a marketplace where businesses can easily deploy various agents for their needs. Businesses want to integrate AI but they don’t know how to do so,” explains Raj K Gopalakrishnan, Founder of KOGO, adding that the platform has garnered customer attraction in the Middle East and North America.
Such AI agents have emerged as the missing link in deploying LLMs for enterprise use, says Ramprakash Ramamoorthy, Director of AI research at Zoho and its division ManageEngine.
“Traditional SaaS applications often rely on static, predefined workflows, which limits their adaptability. AI agents, on the other hand, bring flexibility by interacting with multiple systems, adapting to real-time data and automating decision-making based on context, such as workload, skills, and priorities,” says Ramamoorthy.
AI venture Supervity , which offers an AI agent platform, recently launched SuperSales AI agent to automate processes in B2B sales. Chief Strategy Officer Omkar Pandharkame says the AI agent automates 60% of a salesperson’s workload by automating filling RFPs (defining project scope), updating CRM software, taking meeting notes, and sending personalised emails. He adds that such AI agents will most benefit B2B enterprises, financial services, healthcare, and manufacturing sectors.
Meanwhile, Infloso’s fully autonomous AI marketer, Molly, can independently handle entire marketing campaigns.
“Think of an agency with a team of marketers, analysts, researchers, copywriters, and strategists—all working 24×7 for a brand and they have access to 100X memory power. That’s what Molly can do alone,” explains Utkarsh Khandelwal, Founder of Infloso.
By transitioning to Gen AI and predictive AI solutions, the startup has improved operational efficiency and has become more cost-competitive.
“Previously, tasks such as generating relevant creatives, predicting outcomes of campaigns based on parameters, and others were executed by a combination of in-house trained models and human operators,” he explains.
Meanwhile, Mukesh Bansal’s Nurix AI, which recently raised $27.5 million from General Catalyst and Accel, aims to disrupt ecommerce with AI agents offering human-like voice and reasoning capabilities. Similarly, customer service automation startup Yellow.ai has launched VoiceX, an AI agent that can handle high volumes of customer queries while delivering natural, context-aware answers.
Hybrid pricing strategies
The lines between traditional SaaS and AI-enabled software are blurring rapidly. This has prompted SaaS firms, including Salesforce, to re-evaluate their pricing strategies and find new ways to measure outcomes. The SaaS leader is now pricing Agentforce at $2 per conversation for customer service and sales interactions.
“Instead of the number of social media posts generated, a new AI ROI metric will be a percentage improvement in social media engagement. This change in metrics will push for an automatic change in pricing—from subscription-based to outcome-based,” says Poorvi Vijay, Vice President, Elevation Capital.
For instance, American multinational publishing house Wiley adopted Agentforce, boosting self-service efficiency by 40% and freeing service reps to handle more complex cases. This led to a 213% ROI for the company, along with $230,000 in annual savings, Salesforce noted in its blog.
“Traditionally, SaaS companies have priced their solution on a per-seat-based/user-based fixed contract value. However, since the larger cost of AI agents can be attributed to the underlying compute utilisation, AI agents are increasingly using pay-per-use pricing models similar to database companies,” explains Vipul Patel, Partner, Seed Investing, IIMA Ventures.
Bengaluru-based provides agents for just Re 1 per minute following a pay-per-use model.
However, this new pricing model risks unpredictable costs, particularly if usage spikes unexpectedly, observes Manav Garg, Founding Partner, Together Fund. In contrast, traditional subscriptions offer cost predictability but can be inefficient if usage is lower than expected.
Platforms such as SuperSales offer flexible subscription pricing and an outcome-based model where businesses only pay for results—leads, deals, or tasks automated. A hybrid pricing strategy may offer a sustainable solution by balancing flexibility and predictable revenues, believes Venkat Vallabhaneni, Managing Partner, Inflexor Ventures.
KOGO offers two pricing models. Smaller businesses can choose monthly packages based on conversation volume and voice features. Alternatively, they can also opt for pricing based on specific outcomes and usage.
“In contrast, enterprise clients pay a licence fee for hosting the AI platform on-premises, plus additional fees based on agent usage from their own store,” says founder Gopalakrishnan.
News Source:
https://yourstory.com/2024/10/saas-startups-ai-agents-workflows-investors-generative-ai-salesforce