Key takeaways_

• AI in sales delivers real value only when data, process and CRM adoption are strong.
• Without foundations, AI often amplifies issues (poor forecasts, low trust, weak adoption) instead of fixing them.
• Focus on high-impact use cases (deal risk, prioritisation, coaching) and choose a platform that embeds AI in the flow of work.

 More and more areas of your business are experimenting with AI – and sales is no exception. Increasingly, sellers are using AI to draft email responses, summarise meetings and get suggestions on where to prioritise their efforts.

These are useful starting points, but they sit firmly at the surface. They help individual sellers move a little faster, yet they don’t fundamentally change how sales teams operate or the wins they get. The bigger gains people associate with AI in sales – things like reliable lead scoring, intelligent activity scheduling, clearer pipeline health and more accurate forecasting – depend on something deeper.

To move beyond isolated productivity wins, AI must be built into your processes. It needs consistent data, shared processes and a platform that connects activity, opportunities, customers and outcomes in one place.

With the right foundations in place, AI becomes far more meaningful. It shifts from helpful shortcuts to insight that genuinely supports better decision‑making across the sales organisation.

In this blog, we explore how to do AI in sales in right and drive greater results.

What sales leaders actually expect AI to deliver_

When sales leaders invest time in exploring AI, they’re not chasing abstract innovation. The expectations are shaped by real operational pressures: stretched teams, unpredictable pipelines, rising targets and increasing scrutiny from the board. In that context, AI is expected to help sales functions run better, not just faster.

Here is what leaders want in an ideal world:

• Reduced admin and friction. Teams spending more time with customers and less time updating systems. That means fewer manual CRM updates, less duplication across tools and better capture of activity without asking sellers to do more work.
• Clearer direction through next best actions. Sales environments are noisy and capacity is limited. But AI helps cut through the clutter. Informed prompts based on real activity and deal context uncover which opportunities are at risk, which accounts deserve attention now and where effort is likely to pay off.
• Sales leaders are under constant pressure to explain variance, justify confidence levels and spot problems earlier. AI is expected to surface risk signals humans miss, such as patterns of inactivity, stalled deals and pipeline that looks healthy on the surface but isn’t supported by behaviour. The aim isn’t a perfect forecast, but greater trust in the numbers being shared.
• Keeping sales in the hands of sellers. Importantly, leaders expect support for reps rather than replacement. This means better preparation, more insights and prioritisation support. In short, it takes out the legwork so sellers can focus on what they do best.

These expectations aren’t unrealistic. In fact, they closely align with what modern AI can deliver to sales teams. But they all rest on an often‑unspoken assumption: that the underlying sales environment is already structured enough to support it.

When those foundations aren’t in place, AI still produces outputs, but they don’t drive desired results.

Where AI in sales falls flat_

When AI in sales falls short of expectations, it’s tempting to assume the technology itself isn’t ready. In reality, disappointment far more often comes down to the environment AI is introduced into. This includes:

• Messy or incomplete data. Sales data is often inconsistent by default, such as outdated records or partially filled in fields. AI relies on patterns across this data to surface insight. When the underlying information is unreliable or out of date, the outputs are built on shaky ground.
• Fragmented CRM usage. Many organisations technically have a single CRM, but reps work outside the system because it feels quicker. Important context then lives in separated inboxes, calls or chat tools. In this world, AI doesn’t have a complete view of reality and the insights it produces reflect that.
• Absence of a shared sales process. Deal stages exist, but they don’t always mean the same thing to everyone. One manager pushes deals forward; another holds them back. AI depends on consistency: clear definitions, common signals and agreed moments that matter. Without it, output becomes inaccurate.
• Disconnected tools. Sales, service and finance often operate on separate systems, each optimised for its own needs. AI in sales, introduced in isolation, can’t see the full customer journey. That limits its ability to make meaningful recommendations and reinforces silos rather than breaking them down.

All this leads back to a simple but important truth AI can’t create insight; it scales whatever is already there. So, if the foundations are weak, the gaps become more obvious.

Value-adding use cases for AI (when the foundations are right)_

When data is reliable, processes are consistent and the platform connects how sales actually works, AI delivers real value. These are the most valuable use cases, once your foundations are right:

• Pipeline health and deal risk. Rather than relying solely on gut feel or end‑of‑month reviews, AI can surface early warning signals as deals progress or stall. Changes in activity, missed milestones, reduced engagement or unusual patterns across similar opportunities can highlight risk earlier, while there’s still time to intervene.
• Prioritisation. Sales teams are rarely short of activity, but they can be short of focus. When AI has a full view of accounts, opportunities and recent behaviour, it can help sellers and managers concentrate effort where it matters most. That might mean identifying which deals are most likely to close or which opportunities need attention now rather than later. Over time, this helps teams spend less energy spreading effort thinly and more time working on the right things.
• Sales–service alignment. When sales and service data is connected, AI can provide better visibility across the full customer lifecycle: improving handovers, reducing friction and giving sellers context. This shared view helps teams avoid dropped balls after the deal closes and supports more informed follow‑up, expansion and renewal conversations.
• Coaching and consistency. By analysing activity patterns, outcomes and progression, AI can surface what top performers do differently. This gives managers a stronger basis for coaching conversations and helps create more consistent execution across the team.

None of these use cases rely on grand claims or overnight transformation. The value of AI in sales shows up through better signals, clearer focus and more informed decisions. And these gains quickly add up when applied consistently across a sales organisation.

How to make AI really work in sales_

Ready to access AI that truly revolutionises the way your sales team works? Here’s what you need to do now.

Make CRM a system of work, not a reporting tool_

AI depends on the activity that happens during the selling process, not after. If CRM is only updated for pipeline reviews or forecasts, AI never sees the full picture, making output less accurate.

As AI learns from real activity (interactions, progression, momentum and outcomes), it’s crucial that this information is tracked in your CRM. You need a CRM that is fit for purpose, reducing resistance to admin, while automatically tracking selling activity as much as possible. Workflows should also be designed to mirror the real sales process.

When CRM reflects how sellers actually operate day‑to‑day, AI insights become far more relevant and credible.

Actionable tips to do now_

• Remove or de‑prioritise fields that are only used for management reporting and add friction for sellers
• Use CRM in deal reviews and one‑to‑ones so it becomes part of how conversations happen, not just where data lives
• Design entry and exit criteria for stages that rely on behaviour and evidence, not subjective judgement

Strengthen data foundations by simplifying them_

AI in sales relies on consistent signals. When data models become bloated or inconsistently maintained, AI becomes based on patterns that don’t reflect reality.

Strong data foundations are less about collecting more information and collecting it deliberately. Sellers need to know which fields actually matter, when they should be updated and how that information is used. Clear definitions and disciplined usage matter far more than exhaustive data capture.

Simplifying your data model reduces friction for sellers and increases confidence in AI outputs. When the same data points are completed consistently, AI can spot meaningful trends in deal progression, risk and behaviour rather than noise.

Actionable tips to do now_

• Identify the 10–15 data fields that genuinely influence sales decisions, forecasting and prioritisation – and remove the rest
• Agree and document clear definitions for key fields and stages so they mean the same thing across the team
• Make data quality part of regular sales leadership discussions, not just something reviewed during clean‑up exercises

Be explicit about accountability for adoption and trust_

When AI insights appear without clear expectations around how they’re used, they quickly become optional background noise rather than decision‑making inputs.

Sales teams take cues from leadership behaviour. If AI recommendations aren’t referenced in pipeline reviews, coaching conversations or prioritisation decisions, adoption stalls. Similarly, if no one owns the accuracy of the data feeding AI, trust erodes quickly.

Clear accountability ensures AI in sales becomes embedded in how the function runs, not something teams experiment with on the side.

Actionable tips to do now_

• Assign clear ownership for CRM adoption, data quality and how AI insights should be used in reviews, forecasting and coaching
• Decide where AI insights should show up (e.g. weekly pipeline reviews, deal risk discussions, account planning sessions)
• Encourage constructive challenge of AI outputs and use that feedback to improve foundations rather than dismiss the technology

Choose a platform designed to evolve, not be retrofitted_

Many organisations attempt to layer AI onto fragmented systems, stitching together CRM, email, spreadsheets and reporting tools after the fact. While this can surface insights, it often results in outputs that are hard to explain, govern or trust.

Platforms designed with AI in mind embed intelligence directly into day‑to‑day workflows, using shared data models and consistent security controls. This makes AI insights contextual, explainable and far more likely to influence behaviour.

Actionable tips to do now_

• Review where AI insights currently surface and whether they sit inside or outside core sales workflows
• Assess how easy it is to explain why an AI recommendation was made — opacity undermines trust
• Prioritise platforms that allow AI capabilities to expand over time without rebuilding data models or integrations

A platform for AI sales excellence: Dynamics 365_

AI in sales only delivers meaningful value when it’s grounded in a CRM that reflects how selling actually happens. Dynamics 365 is well suited to this because AI isn’t treated as an add‑on or experiment. It’s embedded into the platform, the data model and the day‑to‑day workflows sales teams already use.

At its core, Dynamics 365 brings together sales activity, opportunity data, customer history and outcomes within a single, consistent environment. That unified view gives AI the context it needs to move beyond surface‑level assistance and support more advanced use cases such as deal risk, prioritisation and forecasting. Because activity is captured within the same system that drives pipeline and reporting, AI insights are based on real behaviour.

Crucially, Dynamics 365 is designed for governance and trust. AI insights inherit the same security, permissions and role‑based access controls as the rest of the platform, which matters when recommendations start influencing forecasts and executive decisions. Sellers, managers and leaders all see insights relevant to them, in the flow of work.

Dynamics 365 also supports extension over reinvention. Instead of retrofitting AI onto fragmented tools, organisations can start with focused use cases and expand over time without rebuilding their CRM or data foundations. This makes AI adoption more sustainable – and more powerful.

How to approach AI in sales sensibly (before you scale it)_

Before expanding AI across forecasting, prioritisation and coaching, the most effective sales leaders take a measured approach. This checklist helps separate productive progress from rushed deployment.

Fix one sales problem first_

• Clearly name one sales problem you want to improve (e.g. late‑stage deal slippage, weak prioritisation, low forecast confidence).
• Ask sales leadership and frontline managers whether this problem genuinely affects day‑to‑day decisions.
• Write down how this problem shows up today without AI (where judgement breaks down, where surprises happen).

If you can’t clearly explain the problem, AI can’t clarify it for you.

Standardise the data and process around that problem_

• Align on what good looks like for the stages, activities or signals involved in this part of the sales process.
• Reduce data capture to what’s essential for understanding progress and risk – remove anything that doesn’t support decision‑making.
• Make sure this data is captured as part of normal selling activity, not added retrospectively for reporting.

This step creates the foundation AI will learn from.

Introduce AI to support the process, not replace it_

• Decide where AI insight should appear in existing workflows (e.g. before pipeline reviews, during deal coaching, at weekly planning).
• Make sure sellers and managers can see why an insight or recommendation has been generated.
• Encourage teams to challenge AI outputs and compare them with human judgement – especially early on.

AI should inform conversations, as opposed to shutting them down.

Measure behavioural change before revenue impact_

• Look for earlier interventions in deals that might otherwise have slipped.
• Track whether prioritisation decisions are becoming clearer and more consistent.
• Use qualitative feedback from managers and sellers to understand whether AI insights are influencing how they work.

At this stage, confidence and consistency matter more than closed‑won figures.

Expand only once trust is established_

• Pay attention to whether teams reference AI insights unprompted in forecast calls and reviews.
• Check whether leaders trust AI‑supported views enough to use them in planning and reporting.
• Only extend AI to new use cases once it’s actively relied on, not just available.

Scaling AI before trust exists usually slows adoption rather than accelerating it.

AI in sales works best when it’s built into a smarter CRM_

AI in sales delivers real value when it removes friction rather than adding complexity. That means less manual admin, fewer disconnected tools and clearer visibility across sales and service.

When AI is embedded into a modern CRM that reflects how teams operate, it quietly handles the heavy lifting in the background: capturing interactions, surfacing relevant insights and suggesting next steps at the right moments. Sales teams move faster without feeling monitored. Service teams gain visibility without chasing information. Leaders get a clearer view of what’s happening across the customer lifecycle, from first conversation through to long‑term retention. That’s the difference between experimenting with AI and building AI in sales that genuinely improves results.

If your current CRM feels more like a blocker than a growth engine (adding admin, causing handover issues or sitting outside how your teams really work), we’re exploring exactly this in our upcoming session: Time for a smarter CRM?

In just 25 minutes, we’ll show how modern teams are using a smarter CRM to reduce manual effort, improve visibility across sales and service and let AI handle the background work so teams can focus on what actually drives outcomes.

Register for your space while they’re available.