AI Agents are Here

Steve Simonson.m4a [ 00:00:00 ]Hey everybody, it's me, Steve Simonson, and I'm at the Market Masters Think Tank brought to you by BDSS. And I'm here today to talk about a really exciting subject, AI agents. They're coming for you, everybody. I mean, they're coming to you. This is good news because AI agents - this is not the Skynet that we're worried about. They're here to help you. They're here to make your life better. And I intend to illustrate to you over the course of the next few minutes exactly how you can take action today for any business and put AI agents to work for you right away. So, actionable, ready to go. Let's do it. Okay, priority number one, this is very important, is come up with the idea of a brand bot. [ 00:00:41 ] Every single company on planet Earth needs a brand bot. If you have a website and you don't want to do customer service yourself, a brand bot will do text messaging in a very insightful and useful way for each of your customers. Obvious benefits are on the screen. It's easy to go, hey, they're 24-7. I got a lower cost, efficient handling, etc. But I'll give you a quick analogy of a company called Klarna. You can search this online. They rerouted 66% of their customer service inquiries in one month. That was 2.3 million inquiries. The AI scored as well as humans for satisfaction, went from 11 minutes of human time to 2 minutes of computer time, and had higher first-call resolution. [ 00:01:27 ] It was the equivalent of 700 full-time agents, which means that was an incredibly successful illustration of a brand bot working on a website. Everybody who has a website, you can upload and you can train, and you can get it to be a perfect, I would say as smart as any customer service agent can be, and accessible 24-7. Brand bots are the first way to go, and you can lower your cost and increase customer satisfaction today. Okay, the next one. And you can try this number out on the screen. Just call the number, and it's a bot that will answer the call. It's an AI voice, and it will go into a little script, but you can interrupt it, and you can change it at any time. This technology is from Bland.ai. [ 00:02:11 ] It is a very effective illustration of how you can use, for 9 cents a minute from this particular company, a bot to answer your phones now 24/7. Again, all that training that I just mentioned for the brand bot would apply here for the telephone agent. You know, all of your care and maintenance, all your installation instructions, warranties, and if you really tie in with your databases, you can do order tracking and other advanced collaboration with your system as well. So priority two, think about telephone agents. Okay, priority three is your internal chat bot. Now, this is not for everybody, but for enterprises who are growing, you need a central database. You need the glue that holds the whole system together. And that. Is something like ERP-type of system. [ 00:02:58 ] When you have that, it's great. A lot of data, not a lot of information, but with an internal database chat bot, you can now just talk to that database. Hey, what was my number one seller yesterday? Hey, what are the top three items that are declining in profit over the last three months versus last year? You don't need any custom reports. The database chatbot will be able to do all of that for you. And it's all on your own data. It's 100 percent private. Nobody gets to share it. So, the LLM doesn't get it. And the way you orchestrate that gives you incredible data enrichment, accessibility and on-point questions. And for any manager, this is an essential thing coming for any enterprise doing more than five million. [ 00:03:40 ] You will absolutely want to have this within a short period of time. Now, one of the next ones, one of the most interesting and I would say exciting parts is the bot army. Right. So this is, again, not Skynet. Just pump the brakes. You don't have to be scared. But it is the kind of bot army that you've always dreamed you could have. What if you could have a group of agents who work together either asynchronously or in a hierarchical strategy to go and execute an objective? So imagine you have a social media department and you say, hey, I'm going to make a bot that's the manager bot, right? You train it to be the manager bot. Hey, you're really good at marketing, blah, blah, blah. Then you say, hey, you're the hashtag researcher. [ 00:04:21 ] Go, hashtag, you know, research on these subjects. You give another one. Hey, go do SEO research on titles and subjects and whatever. Another one says, all right, now take the hashtag stuff and take the research stuff. Make social content for each of these channels. Maybe you have a Twitter bot or you got a Facebook bot, a TikTok bot. It doesn't matter. All of this can be done. And this becomes an entire marketing department. Your social media department could be 10 bots working every single day, day and night, constantly churning out high quality content. You could even have quality. Bots and you can have scheduling bots and so many others. So I want you guys to really come up and get friendly with this idea of multiple bots acting in concert with one another as its own little department or crew or army, however you want to think of it. [ 00:05:11 ] But it's a group of bots working together in individual trained specialties to come out to a common goal. Very, very important. OK, this is kind of how some of these things work. You don't have to memorize how, you know, all of these different aspects of a of a chat bot might work, but it's actually quite simple for anybody who's really advanced and wants to check this out. You can go to n8n.io. So the letter N, the number eight and another letter N.io. You can do visual bots yourself today, right now, and you can connect to many different LLMs. It's highly, highly effective and incredibly exciting because you can attach memory. You can do things with laying chain or pine cone vector databases. [ 00:05:55 ] I'm not going to jargon you to death here, but all of that can be highly extensible to the average user if you can handle visual interfaces. It's not for everybody, but for the advanced folks. All right. So last couple examples. I just mentioned the multi-agent taskbots. There's a company called crewai.com, and they talk about this process. Now, that one is development and code driven, so it's not for everybody. But the concept I want you guys to get familiar with is what if you could have a marketing team? What if you could have a marketing? What if you had a finance department? What if you had a product ideation department? And it was just a bunch of bots doing the jobs that need to be done on a regular basis, and the cost is quite low, and the output is incrementally much, much better for your business than you could possibly imagine. [ 00:06:42 ] All right. So I'm from Parsimony, parsimony.com. We are an ERP company that creates AI extensions. So, you know, you want to make a bunch of blogs. You say to the thing, 'Hey, I need 50. I need 50 blog article ideas.' And it goes, 'Hey, here's 50 ideas.' You pick 25, and you go, 'Hey, go write me a blog, insert images, give me the keywords, headlines, you know, cutouts for the, you know, snippets and things like that. All of that can be done directly. The key with Parsimony or any business operating system is it's the glue that holds the whole thing together. What your objective is in life is to be able to go, 'Where's my single source of truth? Where do I go to find?' The answers. [ 00:07:26 ] And if you put all of your data, every piece of data you can into one structured database, you can chat with it, you can talk to it, you can get reports, your life becomes a lot better. My secret to success over the years, and I've taken many companies from zero to 50 to 100 to 200 and well beyond, that success is based on systems. Business operating software systems, we call it BOSS, is the way a company goes from 5 to 20 to 50 and so on and so on. So. Anybody who needs help on that, you can reach out to me. I'm Steve at Parsimony.com. I don't usually give out my email and I may not reply to everybody, but I can tell you this. Your objective should be to take action. [ 00:08:08 ] Stop reading the thing, right? Just go take action. Every kind of bot I just mentioned is absolutely critical to your success. If you don't have a brand bot in 12 months, and all of your competitors do, they're delivering higher customer satisfaction, higher levels of service, higher levels of access probably better quality customer service. If you don't have a brand bot or excuse me, a telephone bot in again, 12 months roughly, you're falling behind the competition. You don't yet understand that your competition will have all this stuff. And I have one piece of good news and one piece of bad news. If you operate outside of a native language, so let's say you're selling from China in the United States, this is huge, incredible good news for you because new now you have access to a lot of good stuff. [ 00:08:55 ] You can get access to culture and English and precision content creation without having to worry about translation. The bad news is for all the native guys, right? The good news, if you're native, as you could sell in other channels, whether it's France, Germany, Japan, whatever, and get the same access. Those brand bots can can type back to the person in their native language. So your site may be in English. Your catalog is in English. Everything’s in English, but they can have conversations in German or Spanish or Mandarin or Urdu. Whatever. Whatever you want. My whole point is these things are coming. Bots are here. You don’t have the luxury of sitting on the sidelines and going, this AI thing is a fad. It's going to blow over. It’s happening right now. I encourage you to take action. Let’s go out there and get it done. Again, my name's Steve Simonson. You can find me online, LinkedIn or other various social channels, or you could just go to parsimony.com and check out what we do there, both with data and structured AI bots and analysis. Thanks, everybody.