The Highest Form of Selling Is Making Ten Thousand Customers Each Feel Like the Only One You Serve

A while back, I had dinner with a few friends who work in sales.

The dishes had barely started arriving when one of them began to sigh.

"It's strange," he said. "More customers than before. More tools than before. A bigger team than before. So why do I feel more and more exhausted?"

Another friend picked up the thread: that's exactly where the exhaustion comes from — every customer wants to feel that "you get me."

Ten customers? You can get them all. A hundred? Grit your teeth. A thousand?

Ten thousand?

Which brings us to today's topic: personalization at scale.

What is personalization at scale?

It means that when your customers grow from one hundred to ten thousand, you still remember what stage each person has reached, where they're stuck, and what you should say to them next.

The warmth never downgrades.

The Breakfast Shop Owner's Homework

Let me start with an observation from close to home.

There's a breakfast shop downstairs from my home, and the owner has an astonishing memory. I don't eat cilantro — she remembers. The uncle next door wants extra chili — she remembers. The girl in the school uniform is always in a hurry — her steamed buns are packed and ready first, every time.

On that "remembering" alone, the little shop does a booming business.

Think about it: what if the shop grows into a chain, with a million customers?

Who remembers them all?

No one. So the chain can only blast out text messages: "Dear customer, storewide spend-and-save event."

Whoever receives it feels nothing at all.

The essence of personalization isn't some profound technology. It's the feeling of being remembered.

What customers have always wanted was never your text message. It's "you remember me."

But no human brain can remember a million people. So what do you do?

Where the Warmth Leaks Out

For many companies, the CRM in their hands is essentially a notebook. The customer's name, what was discussed last time, when the order was placed — it's all in there.

And after it's written down?

That's the end of it. It doesn't tell you what to do next.

So salespeople can only wait for the customer to make a move, then take one step in response. If the customer stays silent, the salesperson doesn't think of them either. That's called being passive. And passive salespeople are forever fighting fires.

Where do the fires come from? Let me break it down for you: the warmth usually leaks out in three places.

1. Before meeting a customer, the homework never quite gets done. The customer's history, preferences, sticking points — all scattered across emails, chat logs, and meeting notes, and the salesperson has to piece them together bit by bit. They piece until midnight, and the next day they still miss the one detail that mattered most.
2. After the deal closes, the person disappears. Follow-up runs hot and cold, and the records that should be updated in the system never are. The customer feels "used and tossed aside," and the chances for renewal and expanded purchases quietly slip away.
3. More channels, more faces. One story in email, another in text messages, yet another in the livestream. The customer feels like they're dealing with three different companies.

So what's the fix?

Switch to something different. Stop using a "notebook" — find a "strategist."

Let the tool go from "recording what happened" to "reminding you what to do."

Which brings us to AI.

What Does AI Actually Do for Sales

What AI does in sales boils down to one sentence: it turns "reading reports after the fact" into "getting advice before the fact."

It watches customer behavior in real time: which pages they've viewed, which emails they've answered, how long it's been since they last responded.

Then it tells you: this one — time to schedule a call. That one — time to send the proposal. And another one's renewal is coming up — go have a chat.

The salesperson doesn't have to guess anymore. Just do what it says.

Microsoft's Dynamics 365 and Copilot take exactly this path.

Talk is cheap. Let me tell you three stories.

The first: Investec, a financial institution. They use Microsoft 365 Copilot for Sales to help their sellers get the full picture before a customer meeting: what's been discussed, what the customer cares about, where things are stuck — all at a glance. How much time saved? About 200 hours a year.

What does 200 hours mean? At 8 hours a day, that's 25 working days.

The equivalent of 5 extra working weeks. Every year.

What do you spend those 5 weeks on? The work only humans can do: talking with customers, thinking on their behalf.

So worth it.

The second: Lynk & Co, a car brand. Their playbook is a fairly new one: you can buy a car, you can borrow one, or you can subscribe to one. Once the model gets flexible, the processes get messy — everyone uses cars differently, so how do you manage it? With Dynamics 365's customizable tools, they quickly built a set of processes of their own, matching a different experience to each type of customer.

The cars are still the same cars. The experience became one per person.

The third: Zurich Insurance Group. What gives insurance companies the biggest headache is customer data — too much of it, too messy. They brought in Dynamics 365 Customer Insights and used it to find customers and design customer journeys. The result? Lead quality improved by more than 40%.

In other words, the lists in the sales team's hands are now 40% finer gold.

Those are the three stories. Notice something? None of it is AI "replacing sales."

It carries away the chores — digging up information, tracking schedules, reconciling channels — and hands the saved time back to the salesperson, so the salesperson can be the one "who remembers the customers."

Four Things You Can Put to Use Right Away

Within Dynamics 365 and Copilot, a few capabilities strike me as especially practical.

First, ask your data in plain language. No formulas to write, no reports to wait for — just ask: "Which customers are coming up on renewal?" The answer comes back instantly, and you can act the moment you ask.

Second, segment customers in plain language. Say, "Pull out the customers who viewed the pricing page in the past three months but didn't buy," and Query Assist circles out that group for you. Circle them right, and you'll hit the mark.

Third, build journeys in plain language. State your goal clearly, and Copilot spans email, text messages, and offline channels to assemble a complete customer journey. Who is at which stage, receiving what content — all laid out, plain as day.

Fourth, let it write the first draft for you. Emails, follow-up scripts — give it the tone and the key points, it puts out a first version, and you rework it into your own flavor. You reply to customers faster, and the flavor isn't lost.

The results? Microsoft has cited three numbers: a 15% revenue lift per customer journey, 75% less time to develop a journey, and 50% lower offline marketing spend.

Your numbers will depend on your situation, but the direction is clear: every minute saved is handed back to the customer.

Back to the Breakfast Shop

Writing this far, I think again of that breakfast shop downstairs.

The owner keeps fifty regulars in her head — enough, because the shop is only so big. But if your business means to grow, if your customers will swell past ten thousand, past a hundred thousand, the brain runs out.

When it runs out, don't force it.

Let AI be your add-on brain and remember every person for you. Your only job is to hand over the words "I get you" — with your own hands.

Scale answers "how many." Personalization answers "who." Do both together, and you have a business.

May every one of your customers feel that you serve them — and them alone.