AI to the Rescue: Google Bard and ChatGPT Help Plan Your Software Project

AI to the Rescue: Google Bard and ChatGPT Help Plan Your Software Project

Show Video

hello everyone welcome to today's episode of the Manila show hi I'm Matt I'm Jason and we're going to do a fun little workshop today should be interesting should be fun that sounds fun Hands-On Hands-On Workshop yeah we've just been talking at you now we're going to talk at you with the computer in front of us right that's right over here kind of in front of me not really yeah yeah I'm gonna have to squint to see because I'm blind so we are going to implement uh so the the main or or the really the most important phase of any software development project is the Discovery phase the analysis the research Gathering the feasibility testing um and we're we're going to combine that do a little high level exercise here with with AI and how can AI help us with that if we're you know if we have a small team or we want to or maybe we're a single entrepreneur solo Opera entrepreneur and we want to figure out how can we you know get past or get through this discovery phase in an impactful way uh without having to spend a whole lot of time you know tracking people down you know doing interviews with people um and so on yeah or even just like giving you prompts not like chatpt prompts but prompts for where to where your mind should go where the things you should think of are like little ideas and I don't know sometimes you might have a great idea for an app but you don't really have a great idea for I don't know user personas maybe you don't want to Define people you don't really like have the breadth of knowledge of it right right especially if you're a small team now you've got like there's a computer that can help you do it instead of having to get together a team for that if you're just like you said solopreneur of or a duo or whatever the case may be yep yep so like we mentioned before the the discovery phase is the most important phase of of the software development project um a lot of things are are you know a lot of aspects about the projectors are taken care of during the discovery phase really no designer development is really done during the discovery phase view potentially put together some prototypes I mentioned last week like work just isn't really getting done like work work isn't getting done it's all planning of thought processes like uh scaling things out on whiteboards and stuff like that yep yep so some some aspects of the discovery phase right stakeholder interviews and workshops so um if there are really if you're if you're a small team or you're delivering an app for another another uh company or a client you'll want to interview the people that are really wanting this app who really you know who are really looking to implement this app could be for an internal Workforce so you want to definitely make sure you're interviewing the employees or using it but most importantly the stakeholders you want to sit them down whiteboard with them really pick their brain and really understand what it is they want to get out of it um during the discovery phases we're in market research and competitive analysis also gets done right so if you're a small team or you know maybe you want to you know you know get your hands on some data or like a Gartner or look at some other you know Small Business Association type data models that'll give you some some information but AI could potentially help you out there which we'll show here here in a little while during the discovery page you also you know kind of list out the project goals like what is this thing supposed to achieve um some other things like developing user personas which we mentioned earlier right really defining who it is that is going to be using this app if it's going to be an employee right it's it's it's Jennifer uh in it uh she's you know 27 years old she's got a family at home uh she's going to secondary schooling right she's got goals in her life she you know she shops at Target she you know really loves Starbucks I really get to know this user Persona they're not necessarily like an actual real person right but they're kind of like a Persona of what like a market yeah just a model like an example of of a real person that you might might find so Hobbies interests people who would or would not be interested in those types like the thing that you're building right and if you're if the thing that you're building or the app or what you're looking to release there's already kind of like a market out there for that is this user Persona using one of those apps already so just understanding marking down whether or not they're using it user Journey mapping right so if you know at the point of what your app does get designed or developed and you know deployed and people are using it how are they using it you know are they going from uh you know the home screen to another screen to another screen to another screen are they seeing this overlay are they seeing this drawer are they seeing what is what is the map that they're you know from point A to point B or all the way from A to Z what are they seeing in all the different aspects of of the app so uh we talked about requirement Gathering and Analysis right so it's really like defining what are the requirements like this thing needs to do this right user accounts are a must in this app right so people need to be able to reset their passwords people need to be able to get an email to gain access to their account people need to create account people need to change their account information stuff that's like very important but could be overlooked like you know each kind of take that for granted when you're when you're in an app right you just you just assume right you assume that all the apps are going to have login capability and that's just a thing but if you don't plan for it and design for it and think about it beforehand you could end up like not thinking about the time that it's going to take to do those things just assuming that oh this is an app shouldn't just have that right no you have to you have to think about those things you got to put that that plan into place right oh you actually want to change your email address on your on your user account into that clients bring that up before you change it then the previous email will need to get a email saying hey your email address has changed and then the new one gets the email do you expect think about it right and then uh you know during Discovery we also talk about technical feasibility studies right so I actually put out some type of study we talked about earlier about the prototypes right you might sketch on a piece of paper or maybe draw something or actually use software no code tool to build kind of a prototype of what your app is going to be so that you can get that out there to you know a small focus group and you know hey is this something that would be feasible with your you know did you have any issues with this or do you see any problems with this what did you think right so those studies are done during this during this process as well and then a couple other things like risk assessment right during this process you're gonna unearth some things that are I didn't really think of that oh that might be risky okay let's write that down in the risk column right and at the end of the day once you've done all of your Discovery right before you jump into design you've done some type of go no-go like is this a go or is it not a go and you likely have some items in the risk column especially if you've done like if you built like some type of a road map you'll understand why this thing's probably going to take a lot longer than I expected or it's probably going to be a lot more difficult than I realized in the very beginning which is okay it's fine but it'll just get listed in the risk column and maybe the benefits and the advantages might far outweigh any risk that you have in that column but it's always important to build some type of list of risks list of risks listings risks yeah you did well good job and then of course you know project documentation and estimation right so as you're going through everything you're documenting going through each stage of the process you've got some type of maybe it's a Google doc that you're writing up or something yeah there's an ocean yeah notion's good for that um but yeah keeping track of everything and then at the end of the day you're going to have some type of an estimation right like we mentioned earlier about the roadmap you're going to know this thing's going to take could take three weeks could take 18 weeks could take a few months you never know it depends on your roadmap but at least you know at least you know and you could potentially use some AI to really understand what this you know what this potentially could cost you can talk to some local people or give us a call and well yeah we'll let you know right right right so that's that's Discovery in a nutshell really at a high level yeah um just really understanding all the different aspects of what goes into building an app before you actually get in and design and spend all the time designing and or maybe even jump past design and just jump right into the code which can benefit sometimes but you know you definitely want to make sure you at least do some variation of the discovery certainly the danger grows the bigger the project is going to be so small like one page website probably don't need like you know seven weeks of Discovery for that but you know your your big giant web app that competes with Airbnb probably might want to think about that a little bit longer yep yeah so why don't we you know you'd mentioned Ai and how a AI can help let's kind of talk a little bit about what AI is doing like a real small like overview I mean obviously it doesn't really need introduction to those who are in the kind of the tech space like it's the big buzzword that's the big thing that everybody is kind of like focused on right now is that what I saw on CNN last night probably okay yeah or maybe it wasn't because CNN's been a little kooky lately we don't get too into it but yikes um but yeah yeah so like it's the new it's the new like web3 in the tech World everybody's really not on AI so you basically have like your your there are many more but you're two big competitors right now for the space right you've got open AI is GPT model their model four is the one that's out recently and and Google's barred which just came out like two months ago to the public it's been making advances they just made it um available for everybody no more wait list type of a thing so um so that's pretty cool GPT four was the the latest thing from from open AI uh with it Advanced reasoning it's got a high higher ability to kind of like arrange context and read from different contexts and and pull that in and give you like an overview type of thing basically like thinking about the context of different documents at the same time and just combining those things together um it does have internet capability if you're one of the lucky users that has plugins available they did say that they're they're bringing that out to more people soon or now I am not one of them so thank you open AI for that um I've been wanting plugins this whole time but anyway plugins allow internet accessibility which um you know accessible web accessibility right it doesn't have for all users right now certainly Microsoft Bing has chat not chat EBT but gpt4 kind of built into its search capability so it has a chat function which we'll we'll kind of talk about in a sec um Bart on the other hand they recently upgraded their language model uh so they did like a 3.5 to 4 if you're doing a comparison between open Ai and Google it's supposed to have better logic and reasoning and that that whole thing it does have an internet connection web accessibility not web accessibility but website access web access I guess builds into it and it is no longer under a waitlist so everybody can just go to you know bard.google.com and just start using it um so that's kind of where we're at with the race uh both of them are probably working on newer things Bard is public with their newer things I don't know besides the um the plug-in stuff I don't think gpt4 is really talk or open AIS really talked about too big a like huge things you're coming out with right um certainly they probably got something going on underneath the hood but we'll we'll see about that um so yeah but the idea is that you you you give these AI uh AI tools these websites which we'll show you in a minute you give it some context or a prompt and it will based on its language model give you a response particularly an answer for what you were looking for and what you're looking for might not be just a simple yes or no answer or or a one-word answer it might it might be you know give me a recipe I have these I have these ingredients what types of different recipes could I make and it could crank out a handful of different recipes with cook time and measurements of things and prop time and or give me statistics based on this demographic for this country for this state for this age group and what their favorite food is and their favorite color and and that sort of thing so just like stuff you would normally have to do research for that maybe somebody's already done on the internet and it can just go off and find that information or have that information in its brain and just kind of contextualize it for you in milliseconds yeah and like very very quickly which you'll see here so what we're going to do is we're going to we're going to model this right we're going to we're going to basically plan out or start to plan out an app um so yeah I think Matt you're gonna yeah so take it over from here so we're gonna arbitrarily choose a company um and then we're gonna walk through how that company might go through some of the initial Discovery phases that the initial Discovery phase how that might go through some of the things that are inside of that Discovery base and use AI to help generate some content to get the creative juices flowing yeah to get the thoughts growing to start you know to start building up documentation to start to analyze things of course you don't want to take what it gives you and use that as verbatim you want to use that as your stepping stone into your next okay wow this is great information let's dive deeper into that let me go look at the sources let me go read some articles let me go see does this make sense how how powerful is this data that it's showing it's really interesting to note that number one all of the companies have said that their models can be wrong and give give information that isn't 100 correct in some cases depending on the prompts and depending on the situations they the the the models can give you answers and just make those statistics up make those facts up make a source up right so um definitely don't want to be like you know using this for your business and then not do the backup research to make sure that what it told you was was actually like a factual thing because that can be a pretty bad way to base your business right off of something right so so let's get into it so what we're going to do we're going to research to build a mobile app it's going to be for a tech forward restaurant nice um and that that restaurant wants to keep people or its users or people come in there to eat under their own branded experience so they don't want to send people to a third-party website to go view a menu they don't they don't want to send someone to another third party web app to go order food yeah they want to keep their visitors their the people in the restaurant inside of the Branded experience that they offer so this this restaurant will serve an American Cuisine for the zelenial palette located in southeast Michigan this mobile app should handle viewing the menus um it'll handle user accounts that'll handle uh ordering both on location and for pickup very high level and then it should offer in-app it should provide NF offers and including coupons specials promotions and then it should also have a point reward system because we learned in a previous the Manila show uh podcast episode that a very very important benefit or a key to the success of a mobile app when it comes to retention is gamification oh yeah we we did talk about that in the in a previous episode that's actually one of our that was episode six I think yeah because this is the the last of our first run yep of episodes that we did old guy with the beard was the picture old guy with the beard yeah that's right oh yeah he was taking the selfie yeah yeah yeah yeah awesome so we're going to get into it here we're going to be uh sharing our screen those of you listening on the podcast we're going to talk this through it's going to make complete sense um so sure hope so we'll make sure so we've got our our prompt that we're going to take which is a lot of things that we just talked about yeah and we also want to point out that if you're hoping to get a lot of good information out of the AI tools it's important to give really good information into the AI tools yeah so we're doing very basic prompts right now um would be worth it for you to do some prompt research of your own just to you know see what might work better for you what other people have said we are definitely not like Pro prompt Engineers or anything anything like that you know we're users ourselves so do your own research for that so our first prompt is we want to understand what our total addressable Market is going to be yeah so in order to get that information out of the AI tools we need to prompt the tool with what exactly are we looking to get the total addressable Market of right so we are asking it for you know the restaurant Will Serve American Cuisine focused on those Millennial palette located in southeast Michigan we're building a mobile app for a tech forward restaurant that wants to keep its users in its brand new experience the mobile app should handle viewing the menus user accounts and ordering both location on location and pickup it should offer in-app it should provide in-app offers including coupons specials promotions they should also have a blank rewards system that offers free food in exchange for earned points what is our total addressable market so we gave it a lot of information there it knows we're a restaurant it knows it's an American Cuisine it knows it's the selenial palette it knows the physical location right and it knows what we want the app to do so we're starting off in Google bard and we're going to provide that and we're going to see what Google bard provides us so while it's kind of thinking about that what is what is total addressable market like what should we be expecting from this prompt that just showed up on the screen so we should so really it's it's what is what is the size or what is the the slice of the population that you could generate Revenue off of got it right so it's good that we provided the physical location so it can at least run numbers and understand okay your total addressable Market it can start with something here got it got it which is what it did right so it gave us a total address so market for your mobile app is the number of people in southeast Michigan who are interested in this Millennial Cuisine who are likely to use the mobile app to order food according to a recent study there are over 2 million people in southeast Michigan who are between the ages of 18 and 24 so there's your zillennial Market yeah right of these over 70 are interested in the zillennial cuisine additionally over 90 of people in this age group use a mobile app to order food and at least at least once a month that is really good detail yeah and it does like the math and stuff for you like it understands how it like it's not just that it took those numbers and did some things it it knew what to do with those numbers and what to find and it says here based on these figures we can estimate that your total addressable Market is over 1.4 million people this is a significant Market opportunity and your mobile app has the potential to reach a large number of potential customers that's good information very definitely would think about um making sure that that initial number starts correctly like is that the real population of Southeast Michigan but once you have that you could even just extrapolate or maybe even ask Bard for the equation to like what if I have a different population what how do I pull that in to get together oh yeah what was the formula yeah exactly yeah so maybe it's not two million people maybe it's three million people what would the result be yep interesting so yeah so now we're going to ask uh bing the exact same the exact same question we're going to give it all the exact same detail we're going to let it know all the information about the about the restaurant about the Boma lab about the millennial palette about where we're physically located yeah um and that we want the total addressable Market yeah and kind of a note about being obviously with Chachi pt4 or tbt4 built in it's got access to the internet that sort of thing so it's um kind of different than chat gbt4 in that in that in that respect yep so we should expect newer numbers or at least like newer dates in theory depending on what it finds right might might find something that has kind of a better click-through rate or more uh that's what I'm looking for like uh um value prominence uh uh markets share what's what there's a word for it I don't know it's it's a it's um Authority Market Authority like yeah it has more something has more Authority maybe being is going to give you that instead of like the most recent statistic so that's something so we got our response out of Bing you want to hear it I kind of do all right because I can't read it according to a report by Morgan Stanley the total addressable market for online delivery is set to grow from 260 billion to 217 um into 2017 to 325 billion in 2020. and possibly 470 billion by 2025. so it

automatically went down the wrong path thinking that we wanted total addressable Market um to be of a food delivery app yeah and it's just giving us the the value of food delivery apps over the past five years because we didn't even say anything about delivery we specifically even said yeah in restaurants and pick up on location and pick up we did the same thing about delivery so it kind of went off right so here's where you're going to want to understand wait a minute this isn't what I asked for um so you would need to refine things to get it back on track of what you're actually asking for um I wouldn't even say it even got close no to what we're looking for it didn't even mention like the local area anything about Michigan the millennial palette yeah no no no real like I don't know right in detail and it just gave us you know numbers and those numbers look to be more like USA as a whole numbers not even like Michigan itself or right some particular area interesting so now we'll plug the same exact prompt into uh open ai's gpt4 so chat gpt4 yeah and then let's see what we get here what do you what do you think we're expecting to see different because its knowledge is only up to like 2019 21 something like that I don't remember but yeah so it should should be different um I'd imagine that it'll be the value is probably going to be less than what Bard gave us you think so just because of the the knowledge it has all right I don't know so let's see why don't you want to see what we got here yeah it's going through it's it's actually thinking through and it's presenting out to the screen and here's an interesting bit gpt4 is generating it much slower than we got in barred so that's that's something interesting yeah Bard took a few seconds and this is this is still presenting its response there is a difference between the three different um the three different tools by the way too because chat gbt is you type in your prompt it'll immediately start to respond but it might be slow in finishing its response to where Bard kind of waited a second or two and then it provided its whole response right and then uh Bing chat it mentioned hey I'm gonna go find your answer now and have the little dot dot dot loading loading and then it just provided so all three are kind of different I mean chat chat GPT obviously like really takes chat into effects and it mimics like as someone is typing out that response to you whereas the other ones are much more like browser like uh search based type of things like a yeah a web search I needed that so here's the response we got on a chat gpt4 ready same exact Prime yeah to estimate the total addressable market for your mobile app there are a few key factors we need to consider the population size of Southeast Michigan nice percentage of the population that falls within the zillennial age range the percentage of those Millennials that are likely to be interested in any Tech forward restaurant and the percentage of those interested who have access to a smartphone to use the mobile app like it knows it grouped our prompt into separate uh groups of data right yeah so and it made it kind of similar to Bard yeah I mean obviously you didn't like specify that up front it it Bard kind of just said this and the number that and the number this and the number as it continues it says as of my knowledge cut off in 2021 I don't have exact figures but I can give you rough estimates based on available data population of Southeast Michigan in 2020 was around 5 million the zolanial demographics also known as gen Z um population United States was about 68 million or roughly 20 percent of the total population we assume the same proportion in southeast Michigan that's about a 1 million potential customers Tech forward interests um if we assume around 50 percent of Millennials are interested in Tech forward restaurant experience that's about 500 000 potential customers that sounds like a rough guess and then smartphone access so roughly 95 of Americans own a cell phone of some kind and about 85 percent of those are smartphones we take that into account we can we get a potential customer base of approximately 425 000 zillennials now of course it goes into you know these are very high level rough estimates please check into them but as you remember from Bard there was kind of almost like a formula it was doing yeah it was almost thinking through step by step by step so Chachi BT kind of did the same thing here so if you can actually go back and validate those numbers and really go back and say make some make some more educated assumptions weighted heavier more towards one side you might get yourself to a number that's a little bit more right right a little bit more accurate and with with Bard the starting number of that population of Southeast Michigan was much much different I think it was double what chat GPT gave us yeah um so obviously that changes things but if you do that research and plug in that correct number then both of them would give you back it seemed pretty accurate like they were they were pretty pretty close to each other so there's a way to use AI to understand your total addressable Market pretty pretty straightforward yeah fact check some of the numbers right and build yourself a backward formula to understand where it's getting its numbers from and how those numbers affect the decision but there you go yeah so now as we jump over our total addressable Market let's build some user personas let's say Okay of that total addressable Market let's build some user personas so our next prompt now is we don't have to remind these AI tools that we're talking about a restaurant and that we're at Southeast Michigan or that we're building a mobile app right it's already there now you just continue to build on to this conversation that you're having yeah so now it's going to take the responses that it has given us and add that to the context of what it knows plus the new questions that were that we're asking them yep so here we're going to say of that total addressable Market please generate free user personas of varying genders and provide demographic as well as psychographic details place a fictional name to each user Persona as well as a job title and goals they may have and this is going to be interesting buyer is thinking we haven't used Bard to build user personas yet yeah yeah and here we go sure here are three user personas for your mobile app exactly what we were looking for user Persona one Emily 22 years old female Detroit Michigan she's a student she has goals to eat healthy affordable affordable food that is also delicious she fits in the 18 to 24 age range she's you know gender she's got income between 25 000 and 50 000 she's a college student she lives in southeast Michigan wow she's pretty rich for a college student yeah right psychographic details that she's interested in food and fashion and music uh she values Health sustainability and affordability and her goals are to eat healthy affordable affordable food that is also delicious so it's the same goals that were that were set up above now we got our second user Persona right Ben age 23. hey Ben from uh Ann Arbor Michigan software engineer goal is to try new food and support local businesses demographics fits in the millennial so 18 to 24 mail income 50 to 75 000 education a college graduate lives in southeast Michigan psychographics is technology Sports and music values Community Innovation sustainability I mean these are these are amazing yeah then you got user Persona number three Maria age 21 female lives in Royal Oak Michigan occupational retail manager she's got goals to save time and money on food income 25 000 to 50 000 education college student lives in southeast Michigan psychographic details she's interested in shopping travel and music values convenience affordability and efficiency very cool right Maria you could probably be making more as a retail manager you should you should you should search a competitor for yeah she needs indeed yeah she needs indeed she does I I've worked in in retail spaces before uh and and know what those retail managers have made and that's not the the price range for that so so we are going to xnay uh uh Bing chat because Bing chat is still trying to figure out what it is we're trying to build here yeah so we're going to stick with Bard and open ad Chad gpt4 so we're putting the exact same prompt into chat gbt4 we want three user personas absolutely here are three user personas for your Tech forward restaurant last names on these ones yeah I don't know if that's relevant but it does give us a little more like demographic understanding because the first one it didn't say anything about um um you know ethnicity or anything like that right and that's that's certainly something that you want to kind of understand a little bit so here's a user Persona uh Josh Martinez 23 years old male 1 Junior software developer uh demographic so just recently graduated from college moved to Southeast Michigan for his first job at a tech company who lives and shares an apartment downtown with a couple of roommates psychographic details uh Tech enthusiasts always likes to stay updated with the latest Technologies and Trends he enjoys eating out and trying new places he's social butterfly and frequently dines with friends or co-workers a little Junior software developer social butterfly doing good now we got Melissa Thompson this is interesting Melissa Thompson age 28 kind of like on that higher end of the of the zillennial right digital marketing specialist uh Melissa is single lives in a trendy Loft in up and coming neighborhood she has a comfortable salary and enjoys spending her disposable income on new experiences uh some psychographic details Melissa is a foodie foodie and loves trying out new and unique dining experiences she's always on the lookout for places that offer twists and traditional dishes interesting she precedes convenience and values businesses that offer digital Solutions right and then Melissa has a goal of seamless and personalized dining experiences she appreciates appreciates a restaurant that knows her preference and offers customized recommendation recommendations she's also interested in earning points and rewards for her loyalty pretty cool huh yeah and then we have the third one Alex Lee 21 non-binary college student uh demographics Alex is a full-time student studying computer science at a local University they live in campus and eat uh you know due to their busy schedule uh some psychographic details other Tech forward experiences appreciates when businesses use technology in innovative ways they like to support local businesses and are always on the lookout for good good deals due to their tight budget and they got a goal you know they have a goal here set for them offers a fun and interactive dining experience they're interested in an app that makes ordering food easy especially for pickup orders between study sessions uh they would love to tackle they would love to take advantage of any student discounts or rewards nice how how great is like so impactful these user personas yeah and you could even like take these even deeper right like there's no competitive information in here either are they currently using what types of apps they're currently using you certainly could take these user personas document them separately and maybe even dive deeper into each one separately maybe even find these people that might find people on Facebook or social or in your in your circle of friends that might fit in these and maybe interview them and go deeper into cycle psychographic details and maybe um goals or competitive stuff just speaking on the two responses that we received obviously gpt4 is more descriptive with what they're getting like it's it's paragraphs it's actual like writing whereas The Bard is just very and maybe this is you know part of what Bard is doing is we've given it very basic detail and questions like we didn't put any flower or fluff into the questions that we asked or the description that we did we just did like basic facts and maybe that's why it gave us back like just just answers uh whereas gpt4 like obviously we got flowery stuff that we could take and we could we could put that text on a website somewhere right right yeah so now taking it a step further now in the discovery phase we want to start to build a feature roadmap for this app and really understand what are the different features that we're going to need and where are those going to fall on the roadmap should some be done first or should should some be done later so the prompt that we're giving to the different AI tools is for the mobile app provide a feature roadmap of only the features we would need for a minimum viable product and the minimum viable product is the product that you release out to the public that's minimum in value uh minimum and viability meaning it doesn't have everything in it just yet it has just the minimum amount of things that would need to generate revenue or to put a smile on your customers faces or to at least begin to start making money now start making money now right and to begin to learn how people are using it right yeah um so that's what we're saying so for the mobile app provide a feature roadmap of only the features that we need for a minimum viable product please provide an estimated timeline for each feature in the roadmap right so we're asking this to Bard this is a pretty technical thing that we're asking it to yep ooh interesting and and it came back some with some high level stuff so sure here's a feature roadmap and of the minimum viable product for your mobile app and really it just listed out view menus user accounts ordering in-app Point reward system one month one month two months two months three months so kind of high level I would have liked for it to break down for the different menus need to be able to view the menu and see different food items that are in that menu um so it really didn't break it up too much it's interesting that it put one month at view menus and user accounts and knew that ordering is probably more complex and would take longer than that it would take a full two months to do that right so you might take this and you might you know ask it to Dig Down Deeper or maybe document view menus and then you you ask it okay build me a road map of just the view menu part all right so let's ask the same exact thing to um check ubt4 and see what we get sure here's a simple road map for a minimum viable product in parentheses MVP of your mobile app along with estimated timelines starts with user and login First user registration two weeks two weeks right wow that's a tight that's a tight uh manager there but it says this feature allows users to create an account using their email or social media they can they can then log in log out of the app it would include password reset functionality so it does go into a little bit of detail yeah menu viewing three weeks a digital easy to navigate menu that allows users to see the food drink options are available this should include clear images descriptions and prices of all items so that's kind of cool we never mentioned drinks so it just knew right that we're yeah right took that context and understood we're talking about restaurants yep and then it went into in-app ordering four weeks went into Rewards program three weeks special offers and promotions two weeks user user profile and order history three weeks nice and some of this stuff like is really arbitrary like special offers and promotions two weeks you could really build like a a very sophisticated and complex reward or offer promotion system that sends out push notifications of specific things that only allows A specific group of people to see us you know some promotions so that two weeks is really arbitrary because there really could be a lot more than that or maybe it could be you know coupon codes and yeah a table to look them up right so here we've got a road map ish right something that we can build on on both of these yeah I wouldn't necessarily say like these are full-on you know huge road maps but that's certainly something that you could build on yeah and maybe maybe this means you know the prompt should have been more specific on what you wanted out of sure of that right so like Define what the roadmap is what it should look like you can definitely do that in these right in in that prompt right oh yeah you could say for for the feature viewing menus please dig deep into view maps of each feature of viewing a menu viewing a food item right seeing the price seeing the details for sure so now as we go through our Discovery phase we're going to do a little bit of competitive review so we're going to ask we're going to give these AI tools The Prompt of provide a competitive review uh provide a competitor a competitive review of all the possible restaurants in the area that we may compete with based on our user personas do we want a competitive review or another kind of reviewer do you look for provide a list of competitors yeah list of competitors it was a real-time real-time changes here yeah so provide a list of competitors of all the possible restaurants so even when you have these prompts all ready to go it's always you always want to read them back and see okay we've you know refined this in any way so we're asking us to borrow it and we'll see what Bard comes back with oh wow okay so it kind of did a high level sure here are some possible competitors for your restaurant in the area really it broke it up into fast food restaurants casual dining fine dining and delivery services um it's certainly listed a handful of examples um so you potentially could put those into a document or some type of Google sheet this fact check does Morton's what is it Morton's the steakhouse does that exist I think it does Morton's The Steakhouse it does it definitely exists that's really a thing it is the best steak anywhere I guess so that's what Morton says it's got a nice little video planned yep this is interesting because we were doing this type of research for a client recently yep that certainly could have been helpful for you to find those those examples yeah but now we know about Morton's The Steakhouse yeah yeah they're not the steakhouse called Morton's right right all right so we're gonna ask the same exact prompt to chat gbt for about a list of competitors and let's see what we get we'll have to put Mortons in the show notes we will because they've mentioned them free publicity to Morton that's right so it's now responding back and saying as an AI developed by openai which I'm not sure I've seen that before yeah I don't have real time or location-based data access thus I can provide a list of specific restaurants in southeast Michigan that would be competitors as of this current date however I can give an idea the types of competitors you may encounter based on your user personas and your restaurant Concepts so Tech forward restaurants it gives an example well it talks about them but doesn't give actually it just defines them basically which Bard did too but then Bard also provided right name examples and it's interesting because certainly there are restaurants out there that existed in 2021 that will continue to exist this entire time like McDonald's right right very I mean I don't know if they went through and listed no American Cuisine restaurants McDonald's is certainly an American restaurant right no it didn't even it gave yeah like you said it's just defining it yeah I mean food delivery services and apps like I don't know maybe it's specifically it's told because I don't know if these things will exist when somebody is looking this up maybe I shouldn't include them in there so it's kind of a win for Bard right so then to end off you know this discovery Workshop right of course this would be a lot more we we would go into a lot more detail you should obviously go into a lot more detail fact check this stuff way more to it than this um fact check the stuff you're getting out of out of the AI tools but let's now like I've been doing yeah let's ask it now let's do a little bit of fun stuff right like as you get ready so you've gone through this process and potentially you're okay this is a go like let's move forward to maybe help out design or to maybe give your designer or creative team like a little bit of something to go off of you can ask these AI tools to generate some type of a color palette or variations of a color palette based on yeah those Millennial details based on some psychographic details it has been given so we're going to ask both these AI tools provide us with three color palettes with primary secondary tertiary colors with hex codes because we're we want to be smart we want to know these hex codes based on our location and our user personas with an understanding of color psychology and what they mean when it comes to food name the colors in the theme of our restaurant so just don't say fuchsia give fuchsia a cool name give it a special name based on our restaurant because what is a branding color without a special name to denote it just a color just a color just a sad little actually I I really prefer when there are special color names because it makes it easier for me to uh put the like understand what what they are in the in the code like obviously you can just put green but if you're like my Restaurant Green yeah it's much different it's just yeah so Bard came back with something amazing here yeah I've never run a color palette request into Bard and I'm already excited with what I'm seeing here sure here are three color palettes with primary secondary and tertiary colors with hex codes based on your location based on user personas and an understanding of color psychology so color palette one primaries red secondary is yellow tertiary is white here are the hex codes and the theme is Americana red is a vibrant color that is associated with passion excitement and energy it is a great color to use to create Lively and inviting atmosphere yellow is a cheerful color that is associated with happiness optimism and creativity is a great color to use to create fun and playful atmosphere White's a neutral color cool you know um that reminds me of McDonald's oh red yellow white yeah you see Ronald McDonald here somewhere yeah Ronald McDonald red overall yellow color palette too green green blue and brown uh green is a calming color that is associated with nature growth and growth and freshness there's a great color to use to create a relaxed and peaceful atmosphere blue is a Serene color that is associated with the sky the ocean and peace it is a great color to use to create a calm and tranquil atmosphere Brown is a natural color that is associated with earth wood and warmth very cool color palette purple pink and black with the hex codes theme romance purple is a luxurious color that is associated with royalty luxury and mystery is a great color to use to create sophisticated and elegant atmosphere pink is a feminine color that's associated with love romance and sweetness it is a great color to use to create a romantic and dreamy atmosphere black is a powerful color that is associated with sophistication elegance and mystery it is a great color to use to create dramatic and alluring atmosphere a few notes before we go on number one it definitely gave us like like the hex codes it gave us an understanding of why those colors were picked um I think it's it's interesting that the green and blue color palette like some of the people and some of those user personas mentioned um uh sustainability in their in their thing and those are certainly like Recycling and Earth and those are earthy type of type of colors like natural colors so that's very interesting I don't like that we asked for unique names for them and we just got green blue brown we just got red yellow white maybe it took that as name our name you should have been more specific but because it seemed to name the color palette maybe you should have said give us give us unique names based on this that and the other well let's see what let's see what gpt4 does for us so we're going to provide it the same exact prompt looking for three color palettes primary secondary and tertiary colors with the hex codes based on location and our user personas um and here it's responding back the first color palette Urban technical name the palette and we now have names to the colors yeah so this palette is inspired by the tech focus of your restaurant and the urban environment of Southeast Michigan primary a steel gray this color is Sleek and modern secondary is silicon silver a nod to the tech aspect and tertiary is electric blue a vibrant pop of color to engage the younger audience nice palette three is the American Bistro the palette is inspired by the classic American Cuisine you're serving with a contemporary twist so primary is apple pie red this warm inviting color is reminiscent and classic American dishes okay uh I was looking to see if we were going to the red yellow white again which were not the secondary is biscuit beige that's kind of cool a neutral color that suggests warmth and comfort and the tertiary color is craft beer Brown this adds a modern sophisticated Edge to the palette palette three you ready drum roll Michigan Lakefront context contact look at that those Powers inspired by the natural landscapes of Southeast Michigan primary is Lake blue soothing color secondary is Morning Mist a lighter lighter Airy color that adds a feeling of openness and tertiary of sunset gold a warm color that adds contrast and excitement to the palette so very cool I got our three color palettes we got names we've got hex values and we have what they mean so we know where to use them why to use them and how they impact our users yeah yeah I definitely like this more than with with Bard I think it's the names the names get me yeah there you go that's really cool I I think it'd be cool if we put um put these pallets somewhere that people can see maybe we can pop them on the screen here or here yeah or something like that so we can see if the like obviously we just see the hex codes and I can't interpret hex codes in my brain right yet um but uh it would be nice to see if these these colors are actually like matching like do they look good right right that's that's certainly a thing to to think about right so that's that yeah I mean how cool is that we just kind of went through like a little miniature Discovery phase and we used AI to help us through the the majority of the aspects of of Discovery in a very high level high level weight yeah giving you the tools and context necessary to now go dive deeper into each of these aspects and it might be a little bit more work but you could you could use both of these tools in tandem you could ask them the same question sure get the different responses and pick and choose and combine the two responses that you get into into maybe what you need or what your inspiration is you know put it on a mood board or something like that who knows yeah get weird with it absolutely awesome so that's that's using AI that was that was really cool I think I think that that was that was a neat little uh experiment to do yep awesome so thank you for listening thank you if you're listening through this episode I'm sure it was a little bit difficult but oh you know maybe you want to go check out the the video uh sister uh file of this uh podcast over on YouTube but hopefully we've done enough to kind of explain and read through what the what the tools were giving us yeah that was fun I liked it awesome cool well thanks for joining us appreciate it see you next week [Music] foreign [Music]

2023-05-27 12:38

Show Video

Other news