Subscribe
Share
Share
Embed
IIIMPACT is a Product UX Design and Development Strategy Consulting Agency.
We emphasize strategic planning, intuitive UX design, and better collaboration between business, design to development. By integrating best practices with our clients, we not only speed up market entry but also enhance the overall quality of software products. We help our clients launch better products, faster.
We explore topics about product, strategy, design and development. Hear stories and learnings on how our experienced team has helped launch 100s of software products in almost every industry vertical.
With these automation tools at hand, we can now sort of get to the foundational phase, I would say, almost within weeks and not months.
Speaker 2:It'll be curious to see if that actually helps people actually implement better processes because it it just takes 3 things. It takes the people on your team, the process in your organization, and the product that you're actually making to actually be successful.
Speaker 3:Can you
Speaker 4:run a business if you never actually knew the business processes that were happening beforehand? Like, you come in brand new to a business, and you're just talking to a bot now and kind of manage everything for you in a way that you didn't really actually knew, like, there were actual intricacies of what was really going on behind the scenes.
Speaker 3:It doesn't necessarily take away the sort of creative space because you always had, you know, website templates things. But it does allow you to just shift through all the the time consuming work, as you're saying, and and get to, you know, get to design a lot quicker and start, you know, working with base designs and wireframes, which is really useful.
Speaker 2:Hello, everybody. Welcome back to another episode of make an impact. I'm your host, Makoto Kern, and I've got my usual suspects and a and a new one today. Guys, introduce yourselves, please.
Speaker 3:Hey. I'm Brittany Evans. I'm, one of the UX strategists at Impact. Joe, you would
Speaker 4:There we go. Go for it. Yeah. Yeah. I'm Joe, software architect, developer, everything tech.
Speaker 4:I'm into it. So, yeah, good to be here again.
Speaker 2:And we've got a an special guest today, Aldo. He's been on the impact design team for a while and, got some, great content today, Aldo.
Speaker 1:What's up, everyone?
Speaker 3:Well, tell us Yeah. We're looking forward
Speaker 2:to this. About your
Speaker 3:exciting lineup.
Speaker 2:Cool. Great. So we've got a lot of good topics today, mainly, around AI tools and automation and modern product design. So I think although, he's got some really good content around this and and recently gave a gave a talk around on this very subject, so I think it'd be beneficial to our audience to, listen to see what he says and, have a good conversation about this.
Speaker 1:Yeah. Sounds great. Let me kick it off for us.
Speaker 2:Sounds good.
Speaker 1:Alright. So, obviously, as we know, AI has taken over not just a product design space, but also literally any business space that we're moving in, you'll see a sort of form of AI. It's not necessarily a new thing, but it's it's now a big thing that's obviously taken the world by storm. And especially in the product design space, we do a lot of things manually, and it it has always been a very manual process. If we think about sort of user research that we need to do, the physical designs that we need to do, design system set up, having to to do handovers with the devs.
Speaker 1:And there's always been this sort of gap between design and development and trying to bridge that sort of let's call it that the language gap from design to to development. And now with the AI tools that we that we have at hand and the new advancements with it, you can now simply upload a photo of your design, and AI can now quite accurately design an interactive prototype of your sort of design that you've just designed. So it's a very useful tool, in all honesty, to utilize within the product design space to sort of bridge that gap. And we can look at tools from a starting standpoint, tools like Relume, where it's mostly used for websites. But in all honesty, it's an excellent tool if you just want to set up information architecture and get that drop down list and just get the foundation going.
Speaker 1:Now I'm not saying, hey, I should replace everything that you're doing from a product design space, but I definitely see it as a way to obviously save time and perhaps set up a good foundation. Because I've picked up myself that AI usually thinks of something a little bit differently than I do. Especially if you're the only designer on a team, it's like having a teammate that you can sort of bounce ideas off of or someone that basically just does the grunt work and you you can go in and take the shine for it.
Speaker 2:Thanks, Aldo. We don't need you anymore. We'll use AI for now.
Speaker 3:That's fine. These are kind of the these are the tools that we've dreamt about, really, I think, you know, years ago, just, you know, knowing that this, as you say, so much grunt work. That you know, how how efficient can you be as a designer? And I think these tools don't necessarily take away the you know, we've we've had to you look at something like the is it Renemay or, I don't know how you pronounce it, but I was playing around with that as well. It doesn't necessarily take away the sort of creative space because you always had, you know, website templates and things, but it does allow you to just shift through all the the time consuming work, as you're saying, and and get to, know, get to design a lot quicker and start, you know, working with base designs and wireframes, which is really useful.
Speaker 1:Yeah. Yeah. 100%. And user generated content in general is obviously making, like, a massive leap as well. From wireframes, you can now also obviously go into high fidelity very quickly.
Speaker 1:But not only that is you now have the capabilities where you can actually prompt the images that you're looking for. Because I don't know about you guys, but always in, like, high fidelity, especially in website design, you always start, obviously, with the first draft or wait for content from the client or something in that sense. And then you just put it like a stock image there. Now you can actually sort of prompt something very, very cool or prompt exactly what you need, and that can almost be your your final results. So it can move you very quickly through the draft as well, which is quite cool.
Speaker 1:And then, obviously, touching onto that from going from the drafts to, as I mentioned previously, bridging the gap to development, is we now have tools like cursor and v zero. Those are tools that I use. I know there's a there's a lot more in that space. But where you can simply sort of go in, upload your image, and you can prompt the interactions that needs to take place. You can prompt certain animations.
Speaker 1:You can even on cursor, you can bring in a design system JSON file. So you can bring in your whole file, and it will code what you have. And you can basically give that to a developer as foundational item. Because it's mainly used, obviously, to support devs, not replace them. So you'll still need someone with that skill, especially for very complex systems.
Speaker 1:And I think that touching onto that, we now have the capability with the AI tools to go from 0 to 1 very quickly. And the big thing connected to that subject is that, especially in the startup space and sort of your pitch taking space where you go and pitch to investors, the pitch deck itself is now almost a sub to your MVP because we're now moving into an MVP state of product development. Because everyone now has the capabilities to go and test their ideas, sort of get some sort of visualization, and then get someone on board to do that. So it helps with getting your foundation up very quickly and then obviously getting the right team involved with with the funding that you're receiving. And, I mean, that's for that's for a start up solution.
Speaker 1:If you go in sort of your very big commercial spaces, it allows you to obviously, you now have a massive connection of sort of design system set up. You've got code that's already existing that you can bring in. So you can literally pull up agents that can help you do product validation and sort of the feature validation as well and just get it out there as quickly as possible and do AB testing like crazy within a few days. So that's also a big landscape. That's basically opening up not just for us as product designers, but in general in the product space.
Speaker 1:So that's that's quite exciting to see what's what's going to happen there.
Speaker 2:Yeah. Hopefully, there'll be, you know, you lower the barrier to entry for sure, and there's gonna be probably a more flood of products out there. But hopefully, there will be I don't know if the percentage, you know, they always say 9 out of 10 startups fail, and then 9 out of 10 of those barely make it to year 3 or make profit. So and it seems like there's a myriad of, of course, reasons for that, but not getting user feedback, not testing properly, just kinda building what you think is right for the customer by features instead of being user centric. It always seems to be, like, the kind of the base problems of why these things fail.
Speaker 2:It'll be curious to see if that actually helps people actually implement better processes because it just takes 3 things. It takes the people on your team, the process in your organization, and the product that you're actually making to actually be successful. So hopefully, if you capture, you know, a few of those other p's, you get the the process down correctly to actually enhance your product better or figure out faster that it's not gonna work, so you can pivot.
Speaker 1:Yes. Yes. And I also, saw a post recently about this exact topic where it the market will be flooded with a bunch of new sort of sauce products and new products that's just popping up. But the differentiator would be then the good product from the bad products. Usually, it was Yeah.
Speaker 1:Get your product out there, test it. Now it's like, get your product as good as possible, as fast as possible as well.
Speaker 2:It's like the website boom. Yeah. Everybody when everybody had a dotcom, it you know, oh my god. It's a multibillion dollar company just because you had a website. But now we see that it's really you know, the websites have have really evolved into something much better, and some of them evolved into just they all look the same as well.
Speaker 2:So because there's a there's a general template that works.
Speaker 1:Yeah. Yeah.
Speaker 2:That seemed to work.
Speaker 1:That's very true. And you touched on a on a very interesting topic, the processes. Obviously, we've got 2 massive tools out there. There's another one in 8 in. We've got make.com, and we've got Zapier.
Speaker 1:So sorry. That's 3 big ones where you can actually go in and define those processes and automate a bunch of, sort of, manual processes that would generally take you hours. You can now most likely complete in minutes. And to bring it back to us as product designers or people working in the product space, as we know, sort of that foundational setup of a project usually takes months to do. Setting up the design system, doing documentation, getting those workshops going.
Speaker 1:With these automation tools at hand, we can now sort of get to the foundational phase, I would say, almost within weeks and not months. You can do a lot of the design system itself from physically just prompting a platform like cursor or some other, sort of dev platform. And they would essentially give you those artifacts, which you can then basically just pull into Figma. And you can automate that whole process and run everything from a Slack channel. And the the main thing of, of sort of a communication platform like Slack or WhatsApp or or any platform that you're moving in teams It's that you can use that specific sort of chat interface to drive a lot of the action that you're doing.
Speaker 1:Like, if you're in a meeting, you have platforms like, Copilot, you've got Otter, You've got Fathom. All those sort of transcribers that can pull in the information that you're chatting about. And you can actually within a chat that you're having now, I can go in and say, okay. So here's specific tasks that I want to complete for the end of this month. And after that, I can basically list out everything that I need to do or assign things by just talking to people and the transcribing software would sort of summarize that.
Speaker 1:And you can automate that to bring that into your Slack channel. You can automate the communication that basically say, okay. Cool. Please assign these to these people and you can link that to ClickUp Notion. All of that.
Speaker 1:And basically, update your Notion pages, your ClickUp documentation. Wherever you're doing your documentation, you can do everything through that. And basically, just have one interface and everything basically spreads out to that. So it brings you to sort of one chatbot that you need to interact with to essentially drive a whole business in in all honesty.
Speaker 3:Yeah. It's it's amazing. I mean, tell you that we go ahead, Joe. So something we touched on before was was the agents and, you know, just having this sort of, like, you talk about the central, you know, AI and then, you know, having all these sub agents that, you know, sort out all the different sort of specialties. So whether it's task organization or doing something with code or automation.
Speaker 3:And that's really is the future, you know, more and more just, you know, linkage and speeding up workflow. It's quite amazing. And then we've got to look at the flip side, which you touched on before, where, you know, you have a lot of the security considerations and, you know, data privacy and, you know, how you can try and not stifle this amazing innovation that's happening, you know, by sort of working through those at the same time and and kind of, you know, getting around those over those hurdles.
Speaker 2:Yeah. I've seen some people where they were talking about I can't recall the tools, but, you know, they generate code for you very quickly. But somebody was able to hack to generate a code within just minutes or seconds, especially when it deals with payments. So it's something that you still need to have foundational, like, really good people to make sure there isn't a way to, you know, to break it. And so, yeah, it's just it was just very interesting where somebody was selling a product where you can actually set up some kind of payment system and all that, and they said, oh, within really 2 minutes, they were able to have to I can't remember if it was HTML or whatever, but to basically say they pay for the product and got it for free.
Speaker 2:So it's pretty interesting.
Speaker 4:Mhmm. Yeah. I guess that's I always said it's Bryn.
Speaker 2:It's Brynjolf.
Speaker 4:As Aldo saying, yeah, Brynjolf probably gonna say say the same thing. These tools are great for getting you, like, 80% of the way there. But, yeah, the rest of that 20%, you do still need to have the understanding of what's going on. And I was gonna say earlier, with a lot of these tools, it is quite interesting because we're kind of power users at the moment the way we're using them because we can understand in the background what's happening. So if you're talking to these tools, we can visualize, okay, in the background has created these objects and these tasks, and it's updated the Figma board.
Speaker 4:And we can kind of steer the AI in the right direction, right, because we know where we're leading it to, and we can we can manage it that way because you know what it's actually doing in the background. But it's going to be interesting to see as time goes on if new users can so, you know, people who don't really understand that background, what's actually happening behind the scenes, if they'll be able to interact with these sort of chatbots too and have that natural language sort of conversation without really understanding the technical side of it and still get that same results. Right? So can you run a business if you never actually knew the business processes that were happening beforehand? Like, you come in brand new to a business, you're just talking to a bot now and kind of manage everything for you in a way that you didn't really actually knew, like, know the actual intricacies of what was really going on behind the scenes.
Speaker 4:Just going to be interesting to see how that goes longer term again because we're just looking at this from a very high level detail perspective. We'll see what it's like for just general users wanting to start using these tools and, yeah, be pretty interesting to see what that sort of uptake looks like going forward.
Speaker 2:Yeah. That's an interesting point because if you have, a certain level of, you know, power user or a certain level of a user, does that mean if you're not that good, oh, well, I can have this designer or product owner or somebody or this developer do x. And then you're like, well, actually, I'll just ask, you know, the AI to do it, and so I don't need you because I know just enough to know where I need to get it to. So it's like you have to up your skill set as a, like, independent creator to be able to utilize that. But then now do you take middle management out?
Speaker 2:Do you take upper management out? Do are are they needed? You know? So it's gonna be a probably a battle between the doers and the makers. So
Speaker 4:Yeah. It
Speaker 3:is interesting to see where the dust and where the the debt dust actually gets a a chance to settle with the pace that everything's moving at, because we can we can argue now, well, the tools aren't that good. As, Joe, you were saying, and something I've I've found when you're working with code and you're utilizing, you know, whether it's cursor or something like that, it'll get you, as you're saying, a lot of the way there, but you'll have to identify what's doing wrong. Say, no. No. You're actually this is wrong.
Speaker 3:Rather do this. And it helps you Yeah. You know, speeds up everything. And you could project out. You could say, well, program is starting today, jumping ahead, you know, with these tools.
Speaker 3:Are they going to have that problem solving? And what what are the long term implications of that? But then with the way that it's it's moving, the tools are probably so different when they get to that. The quality of code is going to improve. The products that we're seeing now as somewhat a scramble of trying to assimilate everything together using these tools.
Speaker 3:So although all the the tools you are mentioning, it is sort of bridging certain gaps between different software. You know, they're they're tying systems together from an automation perspective or they're, you know, linking design to development. But we're probably gonna see even more consolidation of tools where, you know, talking about the task management as well, where it becomes a suite that you know a lot less about, and they're just all these outputs from it. And I think we've touched on this before as well, but you look at a UI now, and there's so much that needs to go into some of these more complex so you look at your kind of work your task management systems. You need to be able to order and drag across and have different states.
Speaker 3:That's all to solve, you know, us being able to, you know, deal with it and manage that. But when AI is doing it, you don't need to. You literally see a very simple interface of easier task today. This is what you need to be working with and ask it, well, can you show me some other you know, order it differently or show me the upcoming year? That interface is is somewhat simplified.
Speaker 3:So it is fascinating to see where we're at at the moment, and, you know, we can project out a lot of different scenarios, but, actually, the way that the tools will probably move won't actually, you know we won't see the eventualities of the developers, not new developers, you know, falling into bad habits because either they're going to be replaced completely or, you know, there's going to be just a completely different toolset that we're going to interact with.
Speaker 1:Yeah. Yeah. No. That's that's a very good point. And and I think you touched on something there where we, as consumers, would need a lot less of information upfront.
Speaker 1:So I think even, like, when we get to websites or product itself, you can basically be faced with a card, and this is what we want you to see first. But after that, you basically prompt what you need of the website. What do you want to know of the website? Why are you here? And you can prompt it, and it can give you an answer with some visualization of it.
Speaker 1:I can see it going in that direction, but I think we, as people, are always visually appealed by things that we see. So I do think there's still space for us to be creative around using that AI, but bringing it into sort of a hyper focused view and then obviously splitting off from that. And I think that carries me on to next conversation or to sort of, I would say, summarizing topic is obviously real world examples and sort of how to set up these quick wins. And AI is great, but you need to know when to use AI and when it actually does make sense to use AI, not just from a a usage perspective, but always from a cost perspective as well. Because setting up an AI automation system is great, but we must keep in mind that it does take a lot of upfront sort of implementation to set up a automation system because you usually do, like, a service audit.
Speaker 1:You, need to identify the gaps. You need to either learn specific platforms that need to integrate with each other, and that takes you time. And that's usually time that you have to take away from your own business to get that done just so that you can reap the benefits after that. And also, you need to validate if it's actually saving you x amount of time or if it's actually just saving you, like, 2 hours a month. That's not worth automating, in my opinion.
Speaker 1:So there's obviously sort of usages for that to get things done. And I think that's also one of the reasons why we don't yet have a design system power automation tool out there that's used as a platform because, yes, the design system is sort of ever growing, but the setup of a design system is usually sort of a one off. And the automation of that is sort of the bigger thing that you want to look at. The documentation of it is a different story, and that subject we did touch on. So to summarize, basically, when working with AI, you need to know the goal that you need to achieve with it.
Speaker 1:If there's repetitive tasks that's happening, can you set up an AI agent that we mentioned previously to sort of have as a sound board or as a partner in crime or or something like that that costs less than to sort of get in on a whole platform, like a SaaS platform or or something like that. Because even our SaaS platforms are getting more expensive now because they need to sort of compete with our automation tools that's out there that can do the job, but everything's sort of external from a specific platform and you can now drive it from Slack. That's sort of the main things that we need to look at when we look at AI as a whole from product itself. So, yes, we can get to drop version quicker, but is it the best version that we can produce at the end? And, yeah, I think that's what this topic was was all about today.
Speaker 2:Yeah. I think it's I think it's, oh, go ahead, Bradley.
Speaker 3:No. No. No. Just coming back to to an earlier point where, Makoto, you're talking about the process and and also, Aldo, you mentioned kind of investment. We think, you know, does this make it more difficult for investors now to, you know, have all of this put in front of them where, previously, someone wouldn't have had the the technical know how to do that?
Speaker 3:And it becomes how well they can execute after that. I guess it's your your track record with startups or you know, because suddenly there's so many more ideas that are coming to the forefront of being presented to investors. And, you know, it becomes a a question of, well, how well can you actually run with this if you're given investment?
Speaker 2:Yeah. What I what I see is if you're creating a product, especially now, there's so many companies that are like whether it's a chat g p t wrapper or whatever, they're being obsoleted within months or within a new release. It's, scary to, like, you put a lot of resources and time into something, and then OpenAI decides to throw, you know they're getting 100 of 1,000,000,000 of dollars in investment, so they decide to throw a 2 person team on something, and you're all of a sudden obsolete, whether it's your entire industry, a big portion of it, or the product you just spent a $1,000,000 on and and got investments for is now you know, nobody's gonna use it. So, again, it's if you kinda go back to the dotcom paradigm where everybody was building some kind of website, some kind of Craigslist slash Ebay slash something, you're copying, so you're you're trying to build something that's new. But then, you know, as a person, do I have my own personal AI assistant?
Speaker 2:Hey. I wanna go go find me a, you know, brand new outfit, and then, okay, it comes back and shows you, like, 3 things from 3 different okay. Show me the one that has the best price, you know, with the best discounts. You know, it's it's you've got that good just going out there and finding maybe it's not even looking at websites. Maybe it's just searching for SKUs from, you know, different companies, and then it'll just bring up the picture of what it could look like.
Speaker 2:So, you know, it should be really interesting of where ecommerce b to b or b to c will go to because of that. Yeah. Yeah. I mean,
Speaker 1:I've got a AI of myself or every minor task that you can think of.
Speaker 2:You're talking to it right now, aren't we?
Speaker 1:Yeah. I'm actually not real. I'm actually on another call at the moment.
Speaker 3:So I guess the the question comes out if you're a company listening to this. You know, there are a lot of exciting options that are there, and we've looked at them, you know, covering, you know, a multitude of different areas for a tech business. What would our advice be, you know, for someone that wants to, you know, start adopting AI? You know, we've really we've highlighted well, the risk is you could you know, your your software could be replaced, or your tool could be replaced in the near future with something a lot better. You know, what are the things that companies should be concentrating on?
Speaker 3:And I think from looking at our own, you know, at at Impact clients, you know, what we've seen is things like, you know, knowledge and starting to align with the possibilities of AI, you know, establishing what are things that can change within your organization that aren't necessarily external services, but that AI can be plugged into. So you're always, you know, spending money on whether it's refining your knowledge or internal documentation or creating more APIs to pass knowledge and data over to, you know, even if it's internal AI models? What are the things that, you know, like that that are, I guess, really valuable for people to be focusing on. Because especially if you're hesitant, you know, there are, I think, areas like that that you could say, you know, start the change now, and you're going to be in a much better position to adopt any of these new technologies as they come out. 1st going, well, let's throw all our eggs in one basket.
Speaker 3:Let's jump on this specific platform, and that platform changes, or there are a lot of hidden issues with it that, you know, looking at some of the risk factors of, you know, some of these automated tools, you know, you've got things like just looking at, like, the workflow automation, so things like make.com and Zapier. There's such a benefit from being able to do this sort of 0 code or low code setup, but then you also got to wonder, well, once you've set up, if you've got third party platforms that are changing their APIs and you're looking at it as a once off effort. That's great. Once we're integrated, we don't need to maintain that anymore. I think you kind of need to do that risk analysis upfront and go, look.
Speaker 3:We're gonna save a lot with this, but we may need to spend more on, you know, monitoring or anything like that. So it's kind of I would always recommend, you know, anyone that's interested in it, you know, can reach out to, you know, to someone like like Impact and, you know, get a a risk assessment and, you know, a a sort of cost versus benefit assessment as well on, you know, what technologies will work and what the the hidden risks are and and considerations around those.
Speaker 2:Definitely. Well, do you guys have anything else to say before we I think this is a good probably time to to end this podcast.
Speaker 3:I think there is so much to say. We could just keep talking.
Speaker 2:I think we're gonna do another
Speaker 3:one of just, you know Oh, yeah. Where where everything's going. I think that's the most fascinating sort of discussion point is where are we gonna see things like search engines in 10 years? But that's another podcast. So I think
Speaker 2:Oh, yeah.
Speaker 3:We'll pick that up another time.
Speaker 2:Alright. Well, Gray, we'll wrap it up we'll wrap it up here. Thanks again for tuning in. Aldo, thanks for the,
Speaker 3:Yeah. Thanks, Aldo. It was really interesting. Yeah.
Speaker 2:Yeah. For sure. Good content. And, yeah, tune in again. Till next time.
Speaker 2:Like and subscribe, and, we'll see you soon. Take care, everybody.
Speaker 3:Catch you then, Ed.
Speaker 1:Cheers, bro. Bye.