This article is part of Bane's Technology Report 2025
Over the past two years, Generator AI has taken the central stage with a commitment to increasing productivity by accelerating software development, streamlining marketing content, strengthening support solutions, and reducing management responsibilities. Despite their enthusiasm, most companies have not unlocked these profits on a large scale or have never seen any meaningful benefits for cost-effectiveness or revenue growth.
Currently, Agent AI is intervening with voluntary agents who can follow complex workflows, set goals, plan, execute and learn on the fly. Possibility? Smarter systems, faster results, and more room for people to focus on what's really important.
However, truly successful results remain rare. Many companies record small productivity improvements in several areas, such as software development, but only a few can measure success in double digits.
That's because most companies have not yet cracked their formulas about implementing large-scale AI. And sales represent a more challenging challenge than most activities for just a handful of reasons.
- One use case rarely moves the needle Because Seller's Day is fragmented across many tasks. As most companies have not returned to map end-to-end sales journeys, the effort remains fragmented.
- Bottom-up experiments don't work Because the purpose is essentially unknown.
- Applying AI to existing processes often results in slight productivity gains (Tree productivity) Because new bottlenecks are appearing. Without a process redesign, businesses will automate inefficiencies instead of removing them.
- AI requires large data context and cleanliness However, sales and market data are spread across many systems with little quality control or governance.
- Sales team grows and distractsand this is another tool in a long parade of technical promises. Unlike features such as engineering, where workflows are relatively standardized, sales processes vary widely from team to region, and individual to individual.
- Frontline teams are often reluctant to change Their actions. Creating quotas is considered “sufficient” and AI training is usually static.
However, the benefits are too promising to ignore. Sellers may only spend around 25% of the time they actually sell to customers. AI can double it by taking on much of the work surrounding sales, but it's not that much value. And that's only half of the picture. AI also helps teams improve conversion rates at every step of the sales funnel.
Mapping AI throughout the sales lifecycle
Sales teams looking at this possibility from AI will need to determine where AI can deliver the greatest profits and where to start. Bain's inter-business and inter-business technology and consumer companies deployed in sales identified 25 use cases across different steps in the sales lifecycle that leaders should explore to get the greatest benefits from AI deployment (see Figure 2). Some of these began as more traditional software automation and were enhanced by AI/machine learning. Many of them have been further enhanced by generator AI, and we can now see agent AI being deployed in several use cases.
Realize the possibilities of Agent AI
The deployment of Agent AI promises to further increase the value of your sales. The technology is moving rapidly, but most companies are still raw. Vendors could deliver more competent applications in the next six to 18 months, but target results will be seen at scale among companies using no-code workflows, for example (see Figure 3). The biggest hurdles involve data cleaning, standardizing the process, making difficult governance decisions and modifying work completion (which should include shutting down old ways of working and accessing old tools/data).
Identify where to start
Many companies have a hard time starting from considering a wide range of viable AI applications. The domain in Figure 2 illustrates often interdependent use cases, making it difficult to move forward without first addressing basic elements such as data architecture and business alignment.
Get lead generation and exploration. Without clean and connected data, sellers don't know why their accounts are hot, who engages, what to sell, or how to adjust their messages. Many companies jump on guided sales, but the rep needs first trustworthy, easy to act and genuinely new insights.
The most effective pilots focus on one or two domains on the front end of the sales lifecycle, and sellers need to be most useful in identifying, providing information and acting on leads. Large companies build from there and prioritize use cases based on business values and process preparation. That approach lays the foundation for the lasting profits of sales efficiency, enhanced customer engagement, and seller trust in AI tools.
Expand AI's potential in sales
In our work that helps businesses experiment with AI in sales, we have seen a consistent set of lessons emerge and pilots separate the expanding pilots.
- It employs an end-to-end view of the process. Generic or Agent AI may be a headline, but the real value lies in the combination of agent and Generated AI with traditional AI and automation, process redesign, data cleanup, top-down targeting, and execution focus.
- Rethink the process. Automating mediocre processes only accelerates mediocre results. Rethink your sales activities and develop best practice workflows.
- Narrow down the scope. Trying to do everything at once slows down the momentum. Start with highly impactful slices of the sales process (for example, one or two domains of the six in Figure 2) to build a roadmap that reflects the commercial movement.
- Focus on the data with a bias against perfect speed. Data is important, but perfection is not necessary. Focus on what's enough to move faster and what's needed to clean up your data to reach that point. The first step is to eliminate old, inaccurate or confusing data and content. It takes time and resources. Don't invest here.
- Testing, learning, repetition. Quick proof of concepts is the best tool to identify where value lies. They also build confidence about the vision and the steps to get there.
- C-level sponsorship and execution. Sturdy change management is table stakes. True AI conversion also requires a sustained focus from the executive suite. A dedicated implementation team with real-world features should be accountable to set goals and achieve them.
AI has great potential to transform sales, but most companies have yet to see meaningful results. To turn promises into performance, teams need to identify and prioritize high-value use cases, rethink critical processes, and clean up data. It all depends on a clear top-down commitment to deploying AI at scale. When done correctly, leaders can dramatically improve the lives of frontline sellers and build an edge that is more durable than their competitors.
Report details
Read Technology Report 2025
