If you were forwarded this newsletter, join 2,083 weekly readers—some of the smartest GTM founders and operators—by subscribing here:
Happy Friday, y’all!
My friend, Danny Martinez did a spotlight on me this week, and I wanted to share here as well.
It’s a prediction (always dangerous).
But, more importantly, it’s a way of thinking about staffing GTM teams in the future. Which, in this new (AI) era we’re entering, is worth thinking about—or, rethinking—from first principles.
Here’s Danny’s original post. I have slightly altered it, below. Thank you for featuring me, Danny! Go check out his substack, Hiring Humans.
Let’s get into it.
How many SDRs will exist in 5 years? ~80% fewer…
How do you prepare for the future as an SDR? How about as an SDR leader or revenue leader? We’re in the most significant change to how we work we’ve ever witnessed. AI is tearing up the rulebook across every function in white-collar work.
This is especially pertinent for SDRs. It’s one of the roles that—when done effectively—requires a lot of creativity, empathy and outside-of-the-box thinking. But, there are also parts of the SDR role that are redundant, monotonous, and repetitive. So will AI replace the SDR role entirely, or not at all? It’s not that simple.
We’ve all had bad SDRs clunkily sending us emails/DMs. After scanning these messages, you’re left scratching your head, trying to figure out how you even became a target. In many cases, it’s simply “spray and pray”, where the SDR hasn’t taken the time to identify an ideal customer profile (ICP), much less look through yours.
The opposite is also true. We’ve all received an outbound message that’s felt like magic. It doesn’t feel like sales. The best sellers act as sherpas or guides that add value by showing prospects how to do their job better. Sometimes, with the product or service they’re selling, other times by referring them elsewhere.
The clunky SDR will be eliminated. In a spray-and-pray approach, the wider the “spray”, the better. If the total addressable market (TAM) is large enough and the annual contract value (ACV) is low enough, the unit economics do not justify putting a human to work. This is where jobs will automated. An AI agent makes more sense and effectively reaches more prospects. See the chart below for illustrational purposes.
What about the more sophisticated SDRs? Should they pack up and go home, too? Let’s break down the job of a “sophisticated SDR”. Broadly, there are 3 key stages.
Research: finding prospects that match an ICP, identifying where to find them and understanding their most pressing pain points. At this stage, the objective is to build a target list and gather useful information that will help in the next stage.
Outreach: involves writing an outreach message that gets a prospect's attention. It involves talking to the prospect about a pain point in their daily lives. Done effectively, this should not feel like selling.
Qualification: once an SDR has understood the prospect well enough, they are ready to pass them on as a qualified lead to an account exec or sales rep responsible for progressing them through the funnel.
In theory, AI could perform all of these tasks, but doing them well requires creativity and thoughtfulness, which AI doesn’t have (at least not yet!).
For sophisticated SDRs, rather than considering where an AI agent can replace them, it’s more helpful to think about where an AI agent can help to scale them. To find more prospects and qualify more customers.
Let’s explore each stage through this lens:
Research
This is where AI can help the most. Anyone who’s spent time using tools like perplexity will know how much of a step change it is for research. With the proper context, inputs, and tools, SDRs can increase their efficiency massively at this stage.
If you’ve spent a few hours playing with these models (OpenAI, Claude, Perplexity, and Gemini) and delegating “SDR tasks” to them, you know how powerful and magical they are. They’re not perfect, but they’re incredibly impressive - and this is the worst they’ll ever be. It’s a glimpse into the future that we’re inevitably heading.
Outreach
It’s easy for AI to generate an outbound message. Generating one that stands out is much harder and will become even harder as AI agents are further adopted and inboxes are flooded. Human empathy is most valuable at this stage. Figuring out what a prospect cares about and how to provide value is not easy for humans, let alone AI.
That’s not to say AI can’t help with brainstorming, iterating on copy and executing on actually sending the message. However, human judgement must prevail at this stage to determine what a prospect will care about the most and how to get their attention.
Qualification
If the initial outbound resonates enough, a prospect will provide an SDR information that an AI agent would struggle to find. The SDR should be in the driver’s seat now. If a prospect thinks they’re talking to an AI, they will lose interest. As much as tools improve, human psychology stays the same: we prefer speaking to other humans, at least the ones we like.
AI can still help here, but it should be less customer-facing. It should help the SDR avoid wasting time on questions an AI agent can answer. Instead, the SDR should spend time gaining new information about the prospect’s current context.
Closing thoughts
Unsophisticated SDRs will need to upskill, and sophisticated SDRs will need to scale. For the latter, figuring out where to use AI and where a human touch is valuable will be the skill that differentiates them from the crowd and ensures they are not replaced by an SDR using AI. Broadly, the “customer-facing” touchpoints will need to feel more authentic and human. Everything else is up for discussion.
Personally, I cannot be more bullish on AI. This is the year go-to-market teams deploy AI/agents into production. We’re still early here (imho).
I’ve said it before, and I’ll say it a hundred times again:
AI won’t replace SDRs, but SDRs using AI will.
This new breed of SDRs will have (AI) superpowers. And they’ll be 10x more productive than the old way of SDR’ing.
Don’t fight AI. It’s here to help us, not replace us. Lean in.
•••
As always, thank you for your continued attention and trust.
I do not take it for granted.
Have a great weekend honoring the legend, MLK.
And we’ll see you next time,
Brendan 🫡
My thoughts here is the AI will get SDRs 90% of the way there (research, outreach, qualification) as the AI still requires a human touch to essentially QA whatever is going out (at this current moment).
With their feedback on the quality of the messaging, the AI prompting will be refined over time by GTM engineers - to the point where that part of their job is also redundant (the QAing role might also be taken over by AI).
In my mind, it will be normalized that SDRs become cold calling machines or we see the return of the much maligned 360 AE who manage the sales cycle once a prospect has become “warm enough”. This is a human action that can’t be automated by AI (yet).
When I speak with SDRs in recent times, I find myself always giving this advice: get on the blower and learn to close! These are the people who will always be valued in sales.
Great to have collaborated, Brendan 🤝