The Future of AI Automation Agencies | Humans Talking Agents Episode 2

Key Takeaways

all of a sudden um the entire Market of building these uh really impressive agents opened up to like the SB and the mid market then a lot of people came in and said I don't know how to do this nor do I have the time to do this well the biggest one right now that I'm seeing being automated is like meeting booking for an AI agency who's listening to this like how can they future prooof their business or like where is the business going yeah I think there's a big opportunity for AI automation agencies to start having um vertical focuses hey you are in BWB SAS in this industry welcome back to humans talking agents today we're talking about the rise and evolution of AI automation agencies I'm Braden I'm Daniel let's talk about it so Daniel we've seen a big Trend over the past two years now of the rise of AI automation agencies you are really connected to the space what are they yeah so this has been um like a pretty crazy change that's happened over the past I think two years so we saw this um Mid 2023 so after GPT came out chat GPT came out um what happened was that um you know this this quality of an agent that was really limited to a large Enterprise company before like you would need 40 people to build this thing could now all of a sudden be built by like two people in a week and so all of a sudden um the entire Market of building these really impressive agents opened up to like the SB in the mid market for me at least like the Assumption the initial assumption was is that a lot of the companies themselves are going to be building agents and you know we we've actually seen that for for a long of the time with a lot of companies building on tools like make n8n Etc but it also opened up a wave of agencies to build more complex um automations or complex um agents and the drive really came from businesses uh saying Hey I want something on my website I want somebody that can interact with my customers on WhatsApp like chat text whatever it is plus I want all these internal automations and so then what happened is you had um two waves you had new people getting into the agency world and so these are a lot of times we see them as like former developers um that are like oh I could actually build some of these or kind of former innovators tinkerers people who love to build um seeing this as an opportunity to actually go in and build a business and on the other side we see a whole wave of existing agencies so these are digital marketing agencies these are web development agencies these are like you know Salesforce integrators whoever it might be seeing this as an opportunity to actually add on a whole new skew that their customers are asking for so that's really the trend started and really over the past like one and a half years it's it's completely blown up I mean I think the easy way to look at this is um so it's been reported we have no knowledge you know uh confidential about this but um apparently cost the Bank of America $100 million to build their erot right and this was like pre- chat gbt pre-large language models now with Enterprises there's always a lot of moving Parts there who knows if the figure's 10 times smaller regardless it probably cost them a lot of money right and we actually know this from our own experience working with Enterprises like pre-chat gbt you know vo has been around for a while um it it it was really expensive to actually go ahead and build AI Automation and now what you're finding is large language models have enabled you to build that same High Fidelity experience at orders of magnitude less and this has opened up like the SMB Market before it didn't really make sense like there wasn't really like an Roi statement for an SMB or a startup or a midmarket to build their own custom agent right you would just often use like a point solution like an intercom or like there's tons of great apps that are very specific to one use case whether it was like you know re cons ilation payments or customer support or drift with lead generation you had these like very specific Solutions but building a custom agent didn't really make sense well I think like even going further than that right like going out of the scope of like like a chatot basically what we're talking about um from my mind at least and and kind of in the mind of every customer that we talk to like AI is really at the Forefront and I think that when people think of AI obviously they think of a lot of different stuff but at the end of the day they think about like how much uh what things that I a couldn't could do now that I couldn't do before or B what things I was doing before with a lot of people that could be done way more efficiently with AI right and so that's where I think the the the Crux of the push is coming from and I think that comes out in many different forms and um yeah so all the things you were saying were were totally right but I think it also expands category into like different types of automation different types of Agents right you have like the internal side when it comes to uh efficiency and that can be from an individual doing something um like a marketer going and creating an ad campaign or copy um or it can be looking from like a more kind of like um structural lens so for example like we there's an agency building on voice slow that builds for um energy companies and he actually specifically focuses on like internal compliance uh with a lot of energy companies like uh employees have to fill out a lot of information a lot of form when they're doing things see this is what I mean though is a lot of those use cases were not viable to build before because if it if it cost $100,000 to build an agent yeah and you know the use case was only going to provide maybe $50,000 in value that like the ROI just didn't make sense yeah but with large language models making it now so easy and you know the advance of new tools like voice flow to make it so you know little pluged there uh to make it so much easier to build these agents suddenly there's Roi and all these longtail use cases that just didn't didn't exist before yeah and a lot of the companies don't have the internal uh skill set today right like they don't have prompt engineering expertise agent Builders who are able to put these things together and so you're seeing this like Services layer emerge to fill that gap between the companies who now realize there's an Roi statement to be made there and um the actual tools with the capabilities yeah I think there's also like a perspective shift from a lot of businesses like to be frank like a lot of businesses before like when you when you said AI like it was just as buzzword um but I think once people started talking to chat CP everyday life I think the um understanding of like what the quality of experience could be like is a lot higher and so I think in areas where actually they're you know they could have automated better before um now there's a bit of like you know sexiness around automating it and also just like generally a smoother experience because like for the most part right when you when I look at like 100 quote unquote agents that are out there like 80 of those are like just straight up automations right they're automations with a little bit of large language models in them to make that experience bit smoother but it it makes it feel like it's much more magic than it used to be then you've got like you know the 15 or the 10% of experiences that are now getting a bit more complicated there's more logic involved um and you're kind of getting closer towards talking to like end customers um either from like a you know really Niche standpoint um for something like you know internal like ticket management where your end customers are actually your employees are doing things like sub tickets to your actual end customers when you're dealing with people in the millions now right so there's like a scale there of like difficulty as well um so I think that's the other part it's like you know businesses themselves are now like okay I need to use this my business for me to survive and then yeah you're totally right like the use cases themselves have actually opened up and like I think the quality of an experience that you can build in a short amount of time on the long tail has now gotten like exponentially easier well that's yeah there there's a surgent demand now that is I think coming from sort of two points one is that r y point made before there's a lot of use cases that have been unlocked that before just weren't accessible didn't make sense I think the other is the technology uh that the foundation models are putting out like chat gbt advanced voice mode has caused an explosion in voice demand and it's not necessarily net new there's certainly a lot of net new there's also a lot of businesses that have Bots but they're Legacy right and suddenly the expectations from the consumer have just 10x right because they've now experienced chat gbtv streaming right they've experienced uh really good voice uh automation before and so if your business has an old bot you know two years ago every bot was horrible that was fine right like if your Bot was horrible that's okay cuz that was par for the course yeah now suddenly if your Bot is horrible it's like 10 years behind in the course of two years right so it's it's creating this like huge surge of demand and again these companies don't have the expertise in order to do a lot of these implementations themselves I mean we see this a lot at voice flow where we're working with customers where there is a demand from an exec to implement AI right we need to use AI in our business and then that demand gets kind of like shuffled down a few ranks and then it gets down to the person who's supposed to do it and they're like how do I do this right and that's where agencies are able to fill the Gap and say hey we have the expertise we've done these use cases before yeah let's go ahead let's run a PC really quickly we can get something up and running yeah and I think that's it's kind of interesting too cuz there's like there's there's two levels right like one of them is like there's teams that actually build a lot of this stuff in house and then there's the other side where it's like you don't have a team in house and so you're going to an agency um so it is it is pretty interesting and I also think it depends on like how sensitive the use case is as well um because I I think from like an agency perspective right like I see one gigantic use case all the time and that is like the customer facing agents especially for um I would say like smaller businesses like businesses that don't have enough of a team in house um that's where they're going directly to agencies to be able to build yeah like on the other side when it comes to a lot of like internal automation use cases like right now there's a lot of agencies doing this but like the the floor of how difficult it is to create is is dropping so um like I'm starting to see more tools pop up that make it incredibly easy to do like internal automations so I think from an agency perspective uh we're probably going to see more business on the customer facing side because it's much more high risk so unless you have the team dedicated internally to be able to build it um it's much easier to be able to work with an agency on that I I will say on that whole the dedicated team Point yeah we saw this you know when voice got started like kind of like 5 years ago it was all nlu we were just starting to see the creation of In-House teams for nlu driven Bots right conversation designer is a title that started to really emerge as essentially the ux leg of an internal team dedicated to building Bots and you know now as we've shifted into um more of an agentic world with prompt engineering prompt chaining ackn knowledge of how to being able to put agents together um we haven't seen that shift to internal teams yet but I'd imagine similar to what we saw with nlu driven Bots and similar to what you also saw with web and mobile where things started in agencies and then over time they start to move capabilities inous I think we'll probably see the same thing with AI agents and so agencies are going to have to get really specific on what they're good at what are they uniquely focused on and sort of you know able to provide a 10x experience over a big company building that capability in house right yeah and I think that's actually a better focus of this conversation too right like for an AI agency who's listening to this like how can they future prooof their business or like where is the business going yeah um and so yeah I think that's I think it's pretty interesting like when it comes to a lot of like automation agencies I'm starting to see a lot of them focus in on specific vertical so um there are like real estate is actually like real estate and e-commerce are actually the two biggest ones I've seen like e-commerce I thought was a big why do you think that is yeah I actually it's it's kind of interesting so um e-commerce was a bit more predictable right like you already have um a lot of customer experiences there um there's already a lot of agencies involved in that space right so we have great agencies like Parkfield Commerce that do um like Shopify plus builds um that are also now integrating and building out agents for their customers and so you're seeing things like uh you know being able to provide a much better experience for like where's my order processing returns doing product recommendations um answering questions about products almost like an expert like the way you go to a store and talk to someone so that's coming there I think that was actually a bit more predictable um but the surprising thing for me is that even that industry even e-commerce is I think still pretty early in the adoption curve um a lot of the stuff that we see or the the more kind of innovative experiences that are being implemented are actually with smaller companies um like smaller e-commerce businesses and um the bigger ones are just starting to kind of now pay attention and that's where we're starting to see some of these larger Ecom companies come in use whistle and build experiences to you know starting off with like where is my order um they eventually want to get to product recommendations and kind of being a lot more of a smart agent um but right now it's just just like let's just stick with the basics we know people are asking provide a better experience and so so if if I had to guess yeah as to why it's e-commerce and real estate today that are really seeing big booms it's probably because the there's kind of a few things that need to be true in order for automation to make sense one is there needs to be dollars to be saved or made right you have to have enough customers in order for automation to make sense if you only have one customer it doesn't really make sense to build a bot right so there has to be enough volume when you think with e-commerce it really fits that need of you have a very wide base there are millions of e-commerce stores right the margins are thinner but they have a very large base of customers right so automation makes a ton of sense there now with real estate I think what's really interesting is you're seeing a mix of um lead gen but also a little bit of support as well and I think it's because there are again similar to e-commerce millions of real estate agents real estate firms um and they have a ton of customer a ton of clients and also on the lead gen side they also want to generate business too right and so they both actually have similar properties in that sense and I think it's why you're not seeing um AI agencies oh last thing actually is there's also less regulatory concern like the information that they're they're working with is not as serious as maybe like a bank or a healthcare use case right but I think for you know maybe AI agency folks watching this to start thinking about what other Industries match that characteristic of Real Estate and Ecommerce where there's a very wide base of of folks right because you don't want to go after an industry that only has five incumbent right um you want to go after a wide base and each of those uh companies is dealing with a lot of folks so they have the need for automation especially if they're thin in margin as well and they don't need to have like as bespoke of uh you know human to Human Experience yeah and I think the other part that's interesting in there as well right is both these industries have um traditionally been been very like people oriented so e-commerce before obviously the website right like you go into a store and it' be a very much more personal experience where you have questions about the product especially if it's like a larger ticket item um and I think that's actually where we're starting to see some of the some of the more interesting use cases on the e-commerce side are with these items that are like medium to high ticket because people a lot of questions um and they want to be able to understand that before they can actually go and purchase the product it's the same thing with real estate right you have a lot of questions about the house but there's also um a lot of kind of like lead capture and Lead nurturing in real estate where um rather than having an agent talk to to 100 people can you have or like a real estate agent can you have like an AI agent talk to 100 people right like understand their questions understand their concerns gather that initial information uh and then be able to serve results that can best match those and then do that on a repeating basis so yeah it is pretty interesting and like the other one I see too is a lot of um there is a lot more like customer facing Automation in businesses that uh like talk to people on the front end so like voice is a whole another skew that's kind of come in right so um like the biggest one right now that I'm seeing being on automated is like meeting booking um and that's really looking at like office receptionists right every SMB has or a lot of SB especially with appointment based ones have someone whose job is sitting there U well they have a lot of things on their job but one aspect of their job is picking up their phone booking appointments and so now that voice has gone to a point where this is actually um it's pretty pretty passible um that's where I'm seeing the the first kind of uh point be automated and I start I think I start to see a lot more uh going from that as well like in areas where um you know Healthcare is a bit uh easier to work with um I see a lot of uh people working with clinics um doing the same thing right appointment booking and then it's like answering questions like finding doctors like other parts like that that aren't medical care but are like the administration side of it so we're talking a lot about inbound agents right now right inbound support Etc um I think there's going to be a huge explosion in outbound calling especially for startups startups have um especially Venture back startups so like series A Plus have cash in the bank to make investments and the biggest thing they care about is not cost savings it's usually Topline Revenue growth right yeah um and so I think there's a big opportunity for AI automation agencies to start having um vertical focuses hey you in B2B SAS in this industry I have a template for an incredible outbound caller right you give me the value props of your product and I'm going to go ahead and hit your entire Target list I'm going to have the Integrations built with your CRM with Salesforce or HubSpot or whatever CRM you use and I'm just going to hit your your Target list I mean we heard from um I don't know if there was an agency but you you were speaking with a customer this week who was doing a million calls yeah right and there was actually really interesting implementation as well yeah I think like yeah the way that they implemented it that I thought was really cool was so they're doing a ton of these outbound calls and the first message they have is a record a message so obviously it's smoother and then basically they'll activate the agent when the person actually um asks question right so they actually want to engage so why would why would they do that just for um just for like saving costs and so able to that you're sending a million messages like that's a lot of messages so being able to cash that first one is better obviously create a higher experience and for the most part to like and like I have a lot of skepticism when it comes to bun call I'll kind of say that off the bat but I think right now it's a kind do you have skepticism I think I think it's a great use case I think it's the same thing as automated like outbound on the email side on the email side because it started off as like a really hot use case for AI right everyone wanted to do it but like I have a bunch of friends who work at companies that are doing this and then most people tried it and the reality is that at the beginning it worked really well cuz people weren't expecting to receive a really cool personalized email that was like a high lift thing before and now it's the norm no so this is my point I think there's a moment in time right now yeah where I have not gotten a good outbound Ai call and my phone number is all over the Internet unfortunately um but I I you know I when I hear a voice that sounds robotic yeah um I've gotten outbound calls before but they're often they're not like true agents it's like a press one press two or it's just like a very static message I would love to get like a couple of really good AI calls like I'd try to break the prompt I'd try to chat with it but at least you're getting the engagement and I'm going to continue to hear about the brand if the uh bot's Built Well now once everyone starts doing that you know you got to move on to another Channel and there's a thing right I think it's like it's like I feel like that's going to that that that moment of window is like pretty small and I think there right I think it's right now I think there's also a lot of Regulation with out on calling that makes it tough that's true so where I think the opportunity is actually there's more of an opportunity is on doing like um more like doing up on calls like more bottom of funnel and I think that's where there it becomes a lot more interesting what what do you mean by bottom ofun so for example right like and there's a couple like e-commerce companies that are using voice FL and kind of building different experiences but the common threat I've actually heard from a couple of them is like here this is what we want to get to is um you know let's say a customer has added a repeat customer has maybe added something to the cart or a new customers add some in a card that's like a high tiet item and they haven't checked out yet right now you know the common thing is just kind of send them three emails right they be like please your card here's a discount coup Etc um a lot of the times too like they're thinking laying on like what are other ways that we can reach up to the customer and rather than it being a one-sided conversation actually get some of those questions that are blocking them before actually checking out with stuff like that right like let's say I add something a cart I have an account um being able to receive an outbound call that's a lot more tailored to me saying hey Daniel you know thank you for being a customer here like I saw you just P you just put this in your cart you weren't looking to check out yet I I would love to hear why um I think starts to create like a much more uh engaging experience that someone's more willing to interact act with um I still think there's a lot of room on the outbound side I don't think outbound will ever go away um but I think that like this area of like more how do you make like higher touch like bot more bottom of funnel um calls in a way that's like scalable uh is kind of an untapped opportunity to say can we increase conversion at like a certain point in the funnel by layering on this new channel so we've talked about what are AI agencies where they came from kind of the why now moment that so many of them are popping up how are you seeing AI agencies actually get leads it's one thing to have an AI yeah but like how do you drive your own pipe yeah that's actually the other part um that I would say is the biggest challenge to the agencies right is like one of them is niching down a lot of the agencies are kind of just like being like I will automate whatever um I think you got to pick an industry first of all and I think you got to narrow down to a really specific point in the industry it's like any business really but how do you come up with that thesis though right like I actually um I had attempted to start my own AI automation agency called weekend automation which is a great name by the way don't don't take that you do not have time for um and I was going to do it on weekends and my target was uh B2B startups who were using zenes for Ticket automation yeah right but you know it's I still struggle to get leads right like I wanted to to try to build my own automation a agency try to build empathy and like really understand the problem and the first problem it wasn't a product thing it wasn't like a branding thing it was just strictly like how do I get customers in the door yeah I um I won't pretend speak like I'm an expert on this but the thing that I have seen work for people um especially agencies that are using voice a lot um has been content and being able to create valuable content and put it out especially stuff that really shows like the building experience we're getting in the weeds um what I've found today is that especially with businesses like a lot of people are like very interested in Ai and you know there's two types of businesses there businesses that are like yeah I need this and then there are businesses that are like I need this to a point where like it's kind of more important for me and I I let me go see if I like how to do this yeah right and those folks are there's like this entire Builder Persona where they're going in and they're looking at tutorials and they're like okay I want to get started in this because I think it's going to be valuable for my business and um those folks are usually like the highest intent and that's where I have seen success where you know they will watch a video they'll reach out to an agency they'll reach out to community whatever it is to be like hey you know I've started this I think this could be really valuable um I want to bring someone in who who could know this so that's like one method I'm sure there's other methods that are out there that are really effective but that one specifically um I found has been really good for new agencies entering the space some push back on the content piece if the content you're creating is a tutorial on how to use a tool what you're doing is you're capturing um the subset of the market where those folks know what tool they want they generally know the use case they just don't think they can do it themselves yeah you're not necessarily generating New Leads right so for example if you create a a tutorial that says how to create an uh outbound caller for real estate agents using xtool right the market that you're addressing that content to which is good it's niche but you're addressing it to the people who have tried that tool and failed for that use case it'd be interesting um to see folks start creating content that yes has tutorials right because that is a viable subset of the market we've also seen it's it's super successful for some of the agency owners doing that but you're not necessarily addressing the people who just have the problem but they don't know what the solution is yet right like how to automate real estate leads or how to automate uh you know XY Z how to get more leads like starting to speak more in business value versus you tried to do the thing you couldn't do the thing here's how you do the thing hire me to do it yeah right yeah yeah that's also Fair um honestly I think the best way to be to just ask agenes themselves like what's been working and so um even what we're doing on our side is over doing a lot more of agency Focus events just having agencies who have been able to have success um share what they've been learning cuz I think the reality and I've seen this problem across uh existing agencies like agencies that um you know already have clients that are adding on AI agents as a skew um and then brand new agencies is all of them are figuring out like ay how do I get more customers in this bucket and then also B then how do I like um you know communicate scope show value of these new tools especially when it comes to that mismatch and expectations right I think there's also some responsibility for the platforms themselves like if you look at shathi when Drop Shipping in e-commerce was really starting to First Take off agencies was a huge part of that ecosystem and Shopify did a really good job um collect basically Shopify was marketing the ability to create an online store so that created the Demand right then a lot of people came in and said I don't know how to do this nor do I have the time to do this well and then they were able to link it to the folks who had the expertise the time the agencies right and so there's also um some imperative for the platforms to kind of put in the work on creating the market demand right so in the case of voice FL AI agents it's like how can we help the industry by just blanketing the need for AI agents and as you pull people in if you don't know how to do it yourself if you can't use the tool you don't have the time whatever it is connect those folks with the uh the actual agency owners you know what be interesting actually is um seeing if we can like chat with uh like agencies who were early in like web ecosystem and shopa ecosystem and all these ecosystems that like grew over time just to be like what was it like in the early days because I do think there's a lot of similarities here where it's like new functionality has been unlocked for customers right for like for Shopify e-commerce it was like now all of a sudden it was easier than ever to start an online store you didn't have to be a big company to do it you could be a small business and do it yeah I think the the parallel is like very similar here even with like like web blow right I think like that Al Another scenario where it's like you don't have to be a big company now to be able to do this or put up a lot of money um you know it's easier for anyone to have this high quality type of website so I'd be kind of curious as here from those folks who have been early to say like okay what was it like in the early days and like where do they think similarities are um for something like AI agents so well AI agencies in five years yeah what's that look like oh I think it looks like um Wa Would from like I so I think from a like pop populist standpoint like it looks like um a lot of the other ecosystems we see around like web development um around like uh like ads right like I think that becomes its own area but then the other part of me is also like hm I wonder if like AI automation agencies remain as an isolated fact or they just become a subset of like a core agency right like because the other aspect we we talked a lot about AI agencies is there's a lot of existing agencies that are adding on AI as a service yeah right so I see a lot of web development agencies web flow agencies specifically adding on AI as a service and building in more AI native websites yeah you havei plus and like e-commerce Focus agencies that are building more AI specific focuses for their Niche right so I wonder if like 5 years from now this term of AI automation agency goes away and it's an agency that does like a specific vertical like a business or whatever like a you know website like mobile whatever it is um but AI is the core part of how they do it right because the other aspect too is like there's a really good agency called appen Adam so they're they also use voice flow um and they're a mobile app development shop right they'll build mobile apps for businesses that already have Brands and now they're adding on U the AI layer to it right they're building in a lot of like inapp assistance so they're doing it for their existing clients who already have apps and then they're also kind of pitching this is a new service to new clients to be like we'll build you an AI native mobile app and so I think that's like the other aspect of where things will go um because I'm starting to see it already today but again it's an adoption curve like right now it starts off as like an isolated thing and then as adoption picks up it becomes a core part of the way that people do their everyday business yeah I think it's a hot take but I think in a couple years you start seeing successful AI automation agencies get acquired by companies to bring them in house yeah because actually happened a lot in websites it's happened in a lot of previous shifts mobile especially where big companies once they deem that this is this is going to be important for us right we've got five 10 100 use Cas we I mean we've seen some big companies we work with the number of use cases are endless they have so much scale they have 500,000 employees right they say this is a priority we don't have the time or the ability to go recruit some of the best people who are really interested in this and who have all the expertise they've done the real work before let's just acquire that agency right we work with them they've done some work with us we know the people I think you'll start to see that I mean Shopify early going to say like scooping up design agencies scooping them up right and so I I think you'll actually start to see that as well where big companies will just start buying these agencies I think you also see the same thing from Big agencies as well because like a lot of the larger agencies right like a lot of their clients are asking for it but they don't have the competency in house to be able to do this and so like I've seen scenarios where there are like a automation agencies and they're actual the primary deal flows from other agencies that are subcontracting them and so I think that'll be the other thing like in the in the same way that today you have like large agency and conglomerations conglomerations scoop up these smaller ones um I think you'll see the same thing too so that's that's a good take I I want to bring one last point I i' I saw this debate on YouTube um I forget there there was a Creator who's talking about what should your strategy be from a tooling perspective for AI agencies and you know there's there's sort of two approaches you can either learn one tool extremely well or you can know plethora of tools right I actually don't think that there's necessarily I think in the early days when you're just trying to get leads it's actually useful to know met like several tools uh because what that allows you to do is be able to cover a wider set of use cases right um some tools are just good at different things right I think as you start to Niche down and you start to choose okay this is going to be my vertical I think that's when it makes sense to just go all in on like one tool where you know the skill set really well but early on does make it a little bit more sense to like kind of keep your options open you're testing out different use cases and you at least know about all the different tools and the different strengths yeah yeah I think it's like a it's like developing a core competency right and in this case the core competency is like I I think it's actually twofold one of them is obviously like understanding like AI systems and automation the other one is like honestly a little bit development you got to be a bit of both um especially in today's world to be able to get like the most value out of it and so I think if you're building up that competency and you're being able to like what I see a lot is like graduating up your school up your tool set like a lot of people will start when it comes to AI automation quote they'll start off a very Bas level with like GPT and C GPS then they'll move to like a zapier right and they'll be able to connect different tools then they'll move to like a make.com because they need more flexibility and need more custom code then they'll move to like and N is blow up a lot you're talking about like on the integration side on the integration side specifically like that's the that's that's a graduation path and even like with the use cases right people will start with these tools that are just doing kind of like more um specific automations and then they'll actually on like a tool like voice flow that works with that tool to be able to create more like complex experiences and the best people we see that have like graduated up this thing to like okay started with zapier started with make started with NN then you know using make and voice flow using NN and voice flow um then what they start to do is they started to actually create their own like microservices in the back where instead of using a tool and products we've also seen a lot of people start to layer on like their own products well they layer on products on the front end and the back end right so then like the best way to use some of these tools just becomes like orchestration um and you know you can do orchestration in code or you can do orchestration in a tool that like voice that makes just a lot that a lot faster and handles the state and other aspects but that's like the evolution that we see and so I think part of it right now is mastering that Evolution so you can be um like very dangerous when it comes to being able to implement any sort of AI automation solu because you know how to connect all the tools together you know how to be able to manage State you know to be able to handle like incoming and outcoming data and so it's like you know the advice would just be like master that skill set and the best way to master It Is by working on real projects right so whether that's real projects with the business or stuff that you're facing and use that as a course still that's always the tough thing when picking up you know like when I was first learning the code it's like finding you know it's one thing to just learn the fundamentals you know do code academy and these other things um you often don't learn until you're actually doing a project you're really passionate about and actually building something because like otherwise you're like what direction should I be taking my learning right um and that's again it kind of gets back to that chicken or egg though of like well I got to get a client first I mean who knows maybe the best thing for AI automation agencies who haven't got a at is focus on your own lead gen Bots you know what I mean like focus on like building hey I'm going to build an outbound caller I'm going to build an outbound emailer like whatever it is to like go get me clients and like that's going to be the project you actually invest your time in to be able to learn a tool because this is one point I saw on that like YouTube thread I was talking about is most people do not understand the power of the tool that they're using yeah they scratch the surface you look at like a zapier or a make or an in like uh naden there's there's actually a ton of like integration platforms and like on that side most people are just scratching the surface and might think that they are all similar hey zapier looks like it's a little bit more expensive a little bit easier but they don't even understand the power of like an 8m same thing in the bot building world right like a custom GPT we see this whole time what's the difference between like building like a just a custom GPT straight up on open AI versus like a voice flow and it's like you haven't really gone 100% into the tool sets yet right like and I think I think what's tough is if you don't have a use case that you're working on yeah you don't even know how to like find the path to complexity and advanced building right you're just using the basic stuff cuz you're just kind of like messing around with it right um and I think that's where often people get a little stuck switching between like five or six different tools and I know this is kind of contrary to the advice I gave before but like you really only get to the advanced stuff once you spend a lot of time at one tool right and you have a use case that actually guides you down that path to the advanced yeah but I think we're going need a little off track from the fol like hour so let's let's wrap it there um I think that was a good good take um I think there's something interesting 10 bits in there but there's a lot for us to go on in I think future episodes great thanks everyone [Music]