A sales team can be busy without making progress. Enquiries accumulate, proposals take too long to prepare and forecasts depend on information that is incomplete or out of date. Meanwhile, time spent updating systems leaves less time for conversations with customers.
These are useful starting points for considering artificial intelligence. Its commercial value depends on whether it helps people address a specific problem and improves the quality or efficiency of their work. Achieving better sales results requires reliable information, clear responsibilities and a disciplined approach to implementation.
For sales leaders, the challenge is deciding where AI can make a worthwhile contribution and how to assess whether it is working.
Different forms of AI support different activities. Predictive models use patterns in data to estimate outcomes, such as the likelihood of an opportunity progressing. Generative AI produces content, including draft emails, meeting summaries and proposals. Systems commonly described as agentic AI can coordinate sequences of actions across connected tools, within defined permissions.
These capabilities should be assessed separately. A tool that drafts a convincing proposal does not necessarily provide a reliable sales forecast. A system that identifies promising prospects still needs relevant data and a way for salespeople to challenge its recommendations.
There are documented examples of practical benefits. McKinsey’s 2025 article, Unlocking profitable B2B growth through gen AI, describes a materials company that used AI to support meeting preparation and related activities, reporting that it freed up more than 10% of the target sales group’s time. That outcome illustrates the potential of a particular implementation; it is not a benchmark every organisation should expect to reproduce. McKinsey
The useful question is where similar capabilities could address a demonstrable problem in an organisation’s own sales process.
Sales strategy involves choices about which markets to pursue, which customers to serve and where to allocate limited resources. AI can assist by bringing together information that would otherwise be difficult to review consistently.
A distributor could, for example, examine purchasing histories alongside service requirements and account profitability to identify customers worth further attention. This is an illustrative use case: its value would depend on the quality of the information and whether the recommendations withstand commercial scrutiny.
Historical patterns also have limits. They can favour familiar customer segments and overlook emerging needs. Managers should ask what information a recommendation uses, what it excludes and whether recent developments have changed the assumptions behind it.
Performance measures should reflect these commercial choices. A larger pipeline may be encouraging, but its value depends on the quality of the opportunities, expected margins and the resources required to win and serve customers.
Lead scoring can help teams prioritise enquiries using relevant information in their customer relationship management (CRM) system. Website engagement and previous interactions may provide additional signals, where their collection and use are lawful.
Those signals need interpretation. Repeated visits to a service page might indicate a potential buyer, an existing customer seeking support or someone conducting research. A score can help guide attention, but it does not establish willingness or ability to buy.
Salespeople therefore need a way to correct the system. If apparently promising leads repeatedly prove unsuitable, those outcomes should trigger a review of the scoring approach. Otherwise, automation can reinforce an inaccurate view of the market.
Responsible targeting also requires attention to privacy. Organisations must identify an appropriate lawful basis for processing personal data and meet applicable transparency requirements. Individuals have a right to object to direct marketing, including related profiling. In the UK, PECR requirements must also be considered for relevant marketing channels; a data protection lawful basis alone does not settle whether a particular communication is permitted. ICO
These considerations should inform the design of sales processes from the outset.
For many teams, a manageable first application is assistance with preparation and administration. A draft follow-up email, a summary of approved account information or an initial proposal can provide a useful starting point.
Consider a hypothetical Cypriot technology company responding to an overseas enquiry. An AI assistant could organise the customer’s requirements and prepare a draft using approved product material. The salesperson would then check the specifications, pricing and commitments, and confirm that the proposal addresses the customer’s needs.
This review matters because generative AI can produce plausible but incorrect information. NIST identifies this risk in its Generative AI Profile, alongside the risk of people placing excessive confidence in automated outputs. nvlpubs.nist.gov
Organisations should establish which tools staff may use, what information they may enter and who approves material before it reaches a customer. Prices, translations, technical claims and contractual terms deserve particular care.
Where systems can take actions, permissions should be explicit. Preparing a quotation, changing a discount and sending a proposal involve different responsibilities. Each requires controls appropriate to its commercial consequences.
AI-supported analysis may help managers identify stalled opportunities or patterns that merit investigation. Its usefulness depends on accurate records and an understanding of the context behind the figures.
A salesperson with a modest short-term pipeline may be developing a strategically important account. Another may appear more productive because their records are more complete. Neither situation can be understood through a score alone.
The distinction between analysing a pipeline and evaluating employees also matters. Annex III of the EU AI Act identifies certain employment-related uses, including systems intended to monitor or evaluate workers’ performance and behaviour, as high-risk, subject to the Act’s classification rules. Organisations should assess the intended use and applicable requirements rather than assume that every sales dashboard falls into the same category. AI Act Service Desk
Managers should explain how information is used and provide meaningful opportunities to correct errors. Analysis becomes useful coaching when it leads to a discussion about customer needs, obstacles and development priorities.
A focused pilot gives an organisation a practical basis for investment. If slow quotations are the problem, begin there.
1. Establish current performance. Record turnaround time, staff effort, pricing errors and relevant commercial outcomes.
2. Define the task and controls. Specify the information the assistant may use, what it may produce and who approves the result.
3. Test on a manageable scale. Start with a limited team or product range, recording useful outputs and failures.
4. Assess the full cost and benefit. Include licences, integration, training and review time alongside improvements in speed, quality or sales outcomes.
5. Decide what to change. Expand, revise or discontinue the approach according to the evidence.
Comparisons need care. Differences in customers, deal complexity and market conditions can affect results. Faster quotations are valuable only if accuracy, pricing discipline and customer service remain sound.
Staff involvement is equally important. The people preparing proposals and handling objections are well placed to identify where assistance would help and where it would create extra work. Training should enable them to use the tool effectively, recognise its limits and report problems.
AI gives sales teams more options for organising information and supporting daily work. Turning those options into better performance requires clear commercial priorities and people who can evaluate recommendations, verify outputs and maintain customer trust.
A sensible starting point is one recurring problem, a defined process and an honest assessment of the results. The experience gained can then inform the next decision about where AI belongs in the organisation’s sales activity.
EIMF’s Accelerating Sales Performance: From Strategy to AI Driven Execution brings together sales planning, customer communication and the practical use of AI tools and CRM platforms. The programme supports participants in developing an actionable sales growth plan aligned with their organisational goals. eimf.eu
Further reading
– McKinsey & Company, Unlocking profitable B2B growth through gen AI.
– NIST, Artificial Intelligence Risk Management Framework: Generative Artificial Intelligence Profile.
– European Commission, AI Act — Annex III.
– Information Commissioner’s Office, Choosing your lawful basis for direct marketing.
Join us to gain insights from Alana Hill, learn practical strategies for turning conflict into opportunity, and discover how challenges can drive growth and stronger team performance.
On All Self-Paced eLearning CPD Courses in Financial Regulation
Hosted by the EIMF and the Chartered Governance Institute
Engage with 20+ leading experts and earn 6 CPD units in Financial Regulation.
Not sure if it’s right for you? Let’s talk.
Discount Coupon: NYNY10
Valid until 31 Jan 2025 23:59
Register now to receive a valuable educational resource each day and be automatically entered into our Grand Christmas Draw on 24th December – Don’t miss out!
On Self-Paced eLearning Courses
*complete your purchase before 21 April 2024