What's going to make a difference between a successful sales organization and a non successful sales org is going to be the human capital and the management and the leadership.
Tanguy CailletCRO, XenetaA well-written, personalized email used to show that a seller had done the homework. Buyers now assume a model wrote it, so the effort goes unrewarded before anyone reads the message. Every revenue team can buy the same tooling, which means the edge sits in knowing which account is ready for a conversation and in leaders who turn pipeline data into coaching. Pilots that leave reps out of the loop and dashboards that never reach a coaching conversation waste the spend.
Tanguy Caillet, Chief Revenue Officer at Xeneta, leads sales, marketing and customer success at the ocean and air freight rate benchmarking platform. He spent more than two decades in supply chain consulting and technology before moving into revenue leadership, and he's now piloting a way to give SDRs one view of every account. He argues that technology speeds up the work while leaders decide the results.
“What's going to make a difference between a successful sales organization and a non successful sales org is going to be the human capital and the management and the leadership,” Caillet says. That puts the weight on decisions a revenue leader owns: which accounts get a rep's time, how the number gets forecast and how the data gets used to coach. He starts with the outreach buyers have learned to discount.
Buyers want a rep with a real point of view
Five or six years ago, Caillet says, a strong SDR was someone quick enough to process a lot of information and draft a very good email that earned a reply, and he saw craft in that. Today so much AI-written email is in circulation that a well-aimed message earns no extra credit. “It feels like anyway there is an AI behind it,” he says, which makes the work harder for the people still doing it with care.
That has pushed outreach toward channels where a real person is visibly on the other end. LinkedIn carries a similar risk, he says, but someone with a real account still has to reach out, and from there teams return to cold calls and events. “When customers have issues, they look for solutions and they want to speak to people who can actually bring them a point of view,” Caillet says.
Showing that point of view takes demonstrated competence and expertise in front of the customer, and AI keeps its place in the process. “AI and automation and technology help us to outreach faster, clearly better, more efficiently. That's not going away,” he says.
One view of every account shows SDRs when a prospect is ready
Caillet's first attempt at an AI SDR followed the vendor's playbook. The pitch was that the tool was too complex for SDRs, so revenue operations designed the campaigns and the prompts. “There was no SDR in the loop unless there was a client ready to talk to or to make a cold call,” he says. He cut it.
What he wanted instead was a full view of each account across inbound, outbound, events and webinars, so one SDR can judge timing. Each channel lives in its own platform, and he calls connecting those signals the most difficult part. He wants an SDR to be able to answer a simple set of questions about every account. “At a point in time, is that the right time to go after this customer? Is it warm enough? Are they ready?” he says.
He's now testing a tool built around that idea in a pilot with two SDRs, kept separate from the running of the business. “I could isolate them so that they can ramp up and do it properly,” he says. The plan is to bring what works to the rest of the organization, and marketing's notes from customer roundtables are headed to SDRs as hooks for their outreach.
Forecasting still rests on pipeline math, and quarterly tests keep it honest
Caillet doesn't see AI rewriting forecasting. Managing a pipeline properly, converting it stage by stage and watching the speed and success rate, “has not changed for 20, 30 years,” he says. What technology changes is how fast he reaches the number. He now knows his conversion rate from one quarter to the next and which deals at which stage will slip into the following quarter. Customers still can't sign on time, legal gets delayed and budgets disappear, so he keeps more pipeline than the number needs.
That makes forecasting a question of what to commit. “The art of forecasting is more the art of properly committing a number you know you're going to reach and then having some upside,” he says.
Budget season is when it shows. Caillet builds next year's forecast from drivers and the initiatives behind them, such as stronger SDR support, tooling and pre-sales, and then goes quarter by quarter. “You put hypothesis and then every quarter you test the hypothesis,” he says. His teams use the quarterly business review to say what worked, what didn't and how to fine-tune before the next reforecast.
Data changes results only when leaders use it to coach
Call analysis is where Caillet sees AI's clearest help. He can review every call his AEs take, and a call-intelligence tool ranks them against a qualification framework, sometimes filling in what an AE would otherwise have to record. The limit is context. “There is context that the AI doesn't get,” he says, so some calls get interpreted the wrong way.
The harder question is what happens next. A RevOps team can build a polished dashboard, Caillet says, but “it's still a human management capability that does or does not take it and use it for changing the behaviors of the people.” His AEs can read the same report he does, and without coaching on what it means they won't improve. A VP of Sales or a CRO who can't turn the data into coaching leaves it doing very little.
He allows that technology changes how people work in theory, and then points to what every team shares. “They all have access to the same technology,” he says. The account pilot, the quarterly forecast and the call reviews all come back to who decides what to do with the data. “At the end it's still the responsibility of a management team to make the difference,” he says.





