The strain to maneuver quick on AI is actual. So is the strain to point out outcomes. At Have interaction Boston 2026, Gretchen Keefner, SVP of enterprise gross sales at Bullhorn, sat down with Marc Menschel, head of gross sales at Slack, Salesforce, for a candid dialog about what AI adoption really seems to be like throughout industries exterior of staffing. Menschel works every day with a number of the most revolutionary corporations on this planet, and the patterns he sees in what works and what doesn’t translate on to the choices staffing leaders are making proper now.
Right here’s what they lined.
What separates the corporations seeing outcomes from those that aren’t
Menschel was direct from the beginning: no single trade is getting this proper constantly. What he does see is a set of behaviors that separate organizations making actual progress from these caught in experimentation.
The corporations seeing outcomes begin by understanding what they’re really making an attempt to perform earlier than they deploy something. They run pilots tied to particular outcomes, not broad mandates to make use of AI. They choose trusted companions reasonably than constructing the whole lot themselves. They usually query what precisely do AI instruments substitute and what they’ll unlock from it.
Actual examples of ROI from exterior staffing
Salesforce’s personal journey with AI is instructive, partly as a result of it consists of some early failures. After they first rolled out AI to assist.salesforce.com, the expertise wasn’t what clients wished. They mounted it, and by the tip of final yr 68% of holiday makers resolved their situation with out ever talking to a human, with buyer satisfaction practically doubling within the course of. The lesson Menschel drew from it: assumptions about how clients need to work together are sometimes fallacious, and asking is extra helpful than guessing.
The second instance was extra rapid. Salesforce had roughly 53,000 leads thought-about utterly chilly, contacts who had enquired however by no means been adopted up with. Relatively than assign human SDRs to work leads that didn’t justify the fee, they deployed an AI SDR agent. In underneath one week it generated 25 closed offers and $2.5 million in income.
Southwest Airways took an identical strategy to customer support. Through the use of AI to route and triage incoming calls extra successfully, they introduced common wait instances down from 4 minutes to 45 seconds, and measured the impression in income from clients who might now guide reasonably than wait on maintain.
The widespread thread throughout all three examples is similar. Every began with a transparent course of downside, reimagined how that course of ought to work, after which utilized AI to the reimagined model reasonably than the prevailing one.
Why DIY is failing and what to do as an alternative
One of the vital constant patterns Menschel sees throughout the enterprise is the failure of organizations making an attempt to construct their very own AI options. This isn’t restricted to small corporations. It’s occurring at a number of the largest expertise corporations on this planet. They’re nonetheless shopping for from platforms reasonably than constructing from scratch, as a result of the time and sources required to construct one thing a trusted companion has already constructed don’t make sense.
For staffing corporations, the implication is direct. The info is already in Bullhorn. The compliance and safety layers are already in place. The belief is already established. Constructing one thing adjoining to that from scratch isn’t innovation. Partnering with a platform that has already accomplished the work and continues to iterate is.
A examine from MIT discovered that 90% of AI pilots fail. Menschel’s learn on that quantity was measured: failure is a part of the method, and studying from it’s how organizations get higher. What issues is that pilots are tied to actual enterprise outcomes and measured towards the fitting issues. Hours saved solely matter if that point goes into one thing that generates worth. The query to ask at the beginning of any pilot is easy: what’s the precise impression on the enterprise if this works?
The expertise case for AI
One level Menschel returned to a number of instances was the impact of AI on worker expertise, and it’s one that always will get misplaced within the dialog about productiveness and price.
When organizations take away repetitive, low-value duties from workers’ days, satisfaction goes up. A big cellular retailer with annual turnover above 90% began investigating why individuals had been leaving. Pay and advantages had been components, however so was one thing easier: frustration with repetitive duties that felt like a waste of time. After they began addressing these duties via automation, retention improved.
Staffing corporations are seeing the identical dynamic. A couple of agency at Have interaction shared that worker turnover has moved in a optimistic path since giving their groups AI instruments. The recruiters doing extra useful work are staying longer.
What to observe and the place to begin
On the query of whether or not corporations that really feel behind can catch up, Menschel was clear. Every single day of delay makes it more durable, however the window continues to be open.
His sensible recommendation was in step with what has come up all through Have interaction. Begin with one use case, be certain that the underlying course of is sound, and measure towards income impression reasonably than exercise metrics. Choosing companions who’re maintaining with compliance and regulation issues too, as a result of the tempo of change makes it inconceivable to do this alone.
Prepared to begin your AI journey? Click on right here to get began at this time.
