AI SDR

“And unless AI SDRs are able to do that, I think it might be a temporary dead end for now until we get an order or two of magnitude of improvements in LLMs, which could happen very quickly.”

“Tomas Tungas came and said he did this survey and no one could get them to work. I didn't have an opinion, but I did it on our LinkedIn poll. We had 3000 folks. Only two percent folks said they get the AISD work.”

“In just two weeks, little after, Became the highest response rate on the platform that we use for our AISDR. We opened pipelines, we booked meetings, we touched LAPS accounts. It both seems fast and slow, but it's better than historical human averages we had …”

“But if you don't have two weeks to onboard your AI, just like you would spend two weeks onboarding a human being, then don't do it. Like you literally need to spend the same amount of time, maybe more, on training the AI that you would a human being.”

“At the end of the day, if your AI is adding value, it will not care it's an AI.”

“Again, it's not a magic wand. It's not a genie. Don't say I'm going to turn on the AISDR and I could dump everything else I'm doing in marketing or fire my whole marketing or fire all my STRs. It doesn't work that way.”

“And we started to hyper segment more and actually like rip out some of the data from this to make it more hyper segmented and actually a little bit more simple. And that is part of why it's also working so well for us.”

“You might want to spend the next, even like six weeks of data might be enough for it to get up and running and going with something things. So it has some historical context to help your open rates. I think otherwise it will start to default to being too gener…”

“We replaced our human SDR team with an AI. Yes. We had to train it for a month. Yes. We have to train it for every day, but once we do it works. 1206 leads to Sastra from this from 505,216 messages. That's pretty good. 136 responses. So that's a pretty high di…”

“This is, you will get more benefits, but you do not get time back. You lose time. You lose time with these agents.”

“The reason we got started on this Saster journey, and I think people are missing is Amelia and I didn't want to use AI to make Saster better. We wanted to use AI because people kept quitting that didn't want to do the work. We, the reason we replaced our human…”

“You should never just unleash an AI agent on your entire, entire database. Like, Do not do that. It will not be customized. It's not going to work like this. Like, you do need to train either sub-agents, depending on what platform you use, or do, like, sub-cam…”

“The real bar is better than the average human with 24, seven consistency. And that's which, that's what you're going to see in these emails, like pretty good. With not a lot of errors. That's good enough to crush it.”

“And I really think, especially in 20, 26, we're all going to work in these heterogeneous worlds where we mix AIs and humans and imagining that I somehow mind talking to a great AI is backwards. A great AI is often going to add more value than a mediocre human.”

“And a majority of these target accounts are receiving, you know, highly customized messages and reply on the first send. So that's something I think that is also a big learning, is if you do this right and train the agents the right way, not only is it differe…”

“If you can't sell it yourself, the AI can't sell it for you. It's that simple.”

“These guys are blowing up because the products unlock huge TAM expansion by replacing a lot of humans, a lot of SDR. And none of these guys would be remotely as successful if they had to charge like traditional SaaS companies.”

“There are so many, there are many reasons we had the failures of the AISDR failures in 20, 24, raw LLM based in my experience. It didn't work. The product said he worked before, before Claude four, they didn't work. It was slop.”

“Pretty good. I would say is the bar. It does not need to be brilliant. It does not need to be the best email on planet earth or even an email that you might write yourself. It just needs to be pretty good to get this person down the path you want them to go.”

“So what we found is if, and it might be hard to quantify that, maybe it isn't, but if you get above a certain line, if you invest the time to train your agent, if you read the email exceptions every day, if you upload all the documents, you're going to find th…”

“Read everything. I would say in the first 30 days and then read almost everything. Every day after that. So if you're rolling out your first AISDR agent, you are going to want to read everything. And maybe you're going to want to read everything before you hit…”

“Some specifically to with an outbound agent, they can take a little bit longer, especially if it's warming up anything like specific IPs or domains or dedicated evil addresses. Sometimes that process just takes two to three weeks in and of itself.”

“I, if you have less than, let's say, 10 to 20,000 visitors a month, it's probably not worth it.”

“And what we've found for the last year is that most people still prefer to chat with an AISDR. Over chat or over email versus either phone, voice, or video.”

“But for the most part, 99% of people are happy to talk to Amelia because we're busy. Like, I can't come answer your questions about tickets or Where are the hotel room block is? Or when did, when does the ticket code expire or when do prices go up and what's t…”

“Now we have, you know, more than 20 agents In our full stack, we've deployed more than one AISDR.”

“You need at least one, but ideally two essential humans to run all this and to deploy your AI SDR. you can't have zero people.”

“Nothing we have found is instant. So I would be wary of anything that you see is like an instant AISDR. Oh, roll it out today. I'd be wary of that for a lot of reasons.”

“Most of what you will need to do slash want to do with your agents, you'll probably be fine using just one vendor. Versus using the four or five that we have for outbound.”

“If you haven't gotten any outbound to work, Buying an AI to do it isn't, it's not going to help you. You have to prove it. Any of the agents we use, any of the dozens of agents we use, we are taking a playbook that at least works a little bit.”

“And the problem with self-serve is there's no way to catch for that. Self-serve is click some buttons, load in a list, use some AI-generated copy, probably from Claude or Chat, GBT, and let the agent run, right? So like, How will you ever catch in a self-serve…”

“For this moment in time, specialization is still the right answer because of the four you just talked about and how we're using each of the four, that is playing to their strengths.”

“Our, listen, you may have the world's best SDR and outbound team out there, but our AISDRs have done better than any single outbound person has ever worked at Sastor. Ever. They are better, but they weren't good in 2024, ok?”

“There's no way any AISDR tool is going to have that level of context. That's it. In theory, they could, you know, here's an interesting thing. In theory, of course they could, right? This is just LLMs and some data. But in practice, they're never going to be a…”