In sales trainings and conversations, people are asking about the future of the sales profession and the impact of AI on sales.
It’s a question that comes up – sometimes with impatience, others with fear – when we’re discussing sales processes and skills during training sessions or webinars.
When it comes to AI, every profession experiences a mix of FOMO and fear. We don’t want to miss out on the chance to make the most of it, but we fear it may bring changes we won’t be able to adapt to, that it might replace us or render us irrelevant. Rather than predicting the future, i think what we’re really asking for is guidance for the present.
The question takes the form:
‘Can AI do this?’
Or, in its most optimistic version,
‘Can I do this with AI?’
I believe that sales is a particularly interesting sector for asking this question, because—much like the arts—it combines intellectual, creative and interpersonal tasks. AI’s effectiveness varies across these three areas. And, most importantly, so does the public’s willingness to trust automation.
At present, i don’t think Artificial Intelligence has the capacity to replace a sales team in its mission to build trust and reach agreements. But it can take on a large part of the pre-sales operational tasks involved in tactical marketing, and also to improve sales management processes.
This is probably not the preferred and expected answer for many people. Marketing is all the rage; sales, not so much. Talking about branding, multi-channel strategies, content and segmentation seems to appeal more to a new generation of professionals than picking up the phone, arranging an appointment and having a conversation. If that’s the case for you, I’ve got bad news: what you’re avoiding is probably the only thing the machine can’t do better than you.
To understand this, let’s take a brief look at the sales process. Broadly speaking, we can see it as a journey that begins with an idea and ends with an order.
First steps: strategic marketing
To have a product you need an idea that solves an existing problem. Listening to the market to discover which problems are asking to be solved is the first step towards success. This listening involves sifting through sources of information, gathering data, identifying patterns… tasks at which AI excels. There are highly efficient survey and data analytics platforms (e.g. AYTM) or business intelligence platforms that analyse all the information on the competition for us, monitoring relevant changes (i.e. Crayon). But the existing platforms are not as powerful as the options that can be custom-built for the company using no-code tools and agents. This is happening once the door has been opened for AI to programme for us based on instructions in human language.
Another task involving a certain degree of intellectual complexity is pricing strategy. Beyond simply calculating costs and adding a margin, what is the right price that maximises profit, ensures customer satisfaction, and positions us in the right niche? Dynamic pricing models are a well-known example of automating these decisions based on large amounts of real-time data on costs, supply and demand, and AI is powering and accelerating these efforts.
Promotional marketing
When the time comes for the launch, it’s also time for coordinated creativity. Spectacular images? Striking copy? Perfect alignment of the brand image with all the content? Generative AI can take care of much of that. Where existing tools fall short, you can build your own using vibe coding tools such as Lovable.
When it comes to generating visual or text content, we’ve experienced a certain degree of excessive euphoria. The market is probably starting to demand a degree of elegance and restraint. It’s not about doing everything the tool is capable of, like a child with a new toy: ugliness and excess also undermine trust. That’s why human artistic oversight is essential. But AI will do the manual part.
In search of the customer: lead generation
When all that marketing effort begins to generate interest in the market, it is time to identify leads. Whether we do this by sending out content (outbound) or by capturing attention on our own website or blog (inbound), we need tools to capture data, organise it, clean it up, and manage it.
We need to identify stakeholders and their contact details by searching the internet and LinkedIn, compare them with our ideal customer profile to assess possible business scenarios and evaluate their potential. This is pre-sales work that AI is significantly accelerating.
Poorly selected leads are just noise for the sales team. Quality leads are a treasure. Lead scoring is needed to ensure quality and to categorise leads according to their ‘temperature’ and reliability. And yes, AI can do this too. Leading CRMs such as Salesforce or HubSpot have AI applications designed for this purpose.
Good lead scoring, combined with effective territory allocation and clear priorities, set the basis on which a salesperson can their activities. To do this, they must start each morning with a clear and up-to-date overview of the business in their CRM.
Process discipline
Keeping a CRM up to date and organised has been a nightmare for sales teams for too long. As salespeople, we don’t like entering reports and correcting data. When you can update the CRM by voice using dictation assistants, or directly from notes taken during a meeting by a virtual assistant, there are no more excuses for poorly updated CRMs.
We sales managers like to say that a CRM is only as good as the information you put into it… But let’s face it: good data is of no use if we can’t extract it easily, in an organised and clear manner. Can your CRM provide you with the insights you need to make decisions in just a couple of clicks? Personally, I’m a very analytical manager. I like to see data that makes sense. I must admit that in the past I’ve spent countless hours in Excel converting commas to full stops, reordering columns, VLOOKUPing… or negotiating with the IT department to get them to do it for me. Back then – not so long ago – I would have been blown away by a glimpse of the fully customised graphical dashboards that some small businesses are now achieving through vibe coding.
The moment of truth
It’s eight o’clock in the morning. You’ve got a well-designed product, a market you know well, a strong brand reputation, a list of leads, a flawless CRM and solid account information. You even have an assistant like Gong, Avoma or Fireflies, ready to take notes during your next sales meeting and log them in the CRM or turn them into next steps.
Now is your moment. Because there’s one thing that Artificial Intelligence currently can’t do better than you: build trust.
People buy from other people because they trust them. That’s been the case since the invention of trade, and it hasn’t changed with AI. Not yet, anyway.
A lot happens during a sales conversation:
- An opening where defensive barriers are lowered and parts agree to listen to each other.
- An exploration of reality through questions, active listening, mutual respect and growing mutual curiosity.
- Gradually, curiosity turns into trust, which is needed to move on to the difficult questions or candid explanations, until the decision to buy—which belongs to the customer—becomes a collaborative endeavour.
Influencing without pressuring, supporting the decision without pushing the customer, requires an understanding of how people make decisions. No decision is purely rational. A good salesperson needs to use empathy to gain permission to connect with the customer’s emotional side.
Can Artificial Intelligence develop the ability to build relationships that foster trust?
Modern voice agents can qualify leads, manage appointments, follow up and even address objections. But they are far from mastering the subtle and precise interpersonal skills needed to build rapport and trust in complex situations.
That is why, although there is enormous scope for delegation in everything to do with managing information, interpersonal relationships cannot be delegated.
At the risk of exaggerating slightly, one could say that AI can do much of the work of the marketing director, the head of sales and the director of sales operations, but it cannot do the work of the salesperson. (1)
This has immediate implications for the sales department. When it comes to marketing and sales operations, think like a leader – define what you want – and let AI execute it. But if you’re a salesperson, do our part. Stop staring at data and writing emails. Pick up the phone, or knock on that door, and have a conversation.
That is the only thing AI doesn’t do better than you.
For now.
The near future
Perhaps in the near future AI can be trained to understand gestures, read between the lines, interpret silences, extract information from a pause or a smile, and adapt its tone… It may well be that empathy can be programmed. But this leads us to the following question. Do we want that?
At a time when almost anything is possible, the question ‘what can be done’ is becoming less and less interesting, whilst the question ‘what do we want to do’ is becoming increasingly necessary.
Are we inclined to trust a system that is not human?
Voice interfaces, virtual assistants, companion robots… continue to provoke considerable resistance amongst a large section of the public. Many people are reluctant to communicate with technology as they would with a person. I don’t think this is merely resistance to change, but something deeper. Conversing, having a name, looking each other in the eye, trusting one another… these are the things that give our lives meaning. If we remove the human factor from business and leave only optimised data, we may end up with highly profitable companies, but nobody will want to work for them. The fact that this is not just about computational possibilities but also about acceptance is what makes me say that we are a long way from automating sales. We still need you, sales person.
(1) I use the word ‘salesperson’ in a general sense, as I believe that a key account manager, a field application specialist, a business development director and a customer success specialist—if they do not feel comfortable referring to themselves as ‘salespeople’—have not understood their job.