Automating Content Repurposing: Multi-Channel Content Distribution Using Zapier & ChatGPT | Tutorial
Let's build with Zapier and AI (100+ videos)📅 2023-09-18
Transcript
[00:00] everyone I hope everyone's having a great day welcome back to Corbin AI where I'm showing you how to start leveraging artificial intelligence in your personal or your business life you clicked on today's video because you want to see how to start repurposing content for whatever the context may be this is going to be a pretty in-depth video but it's going to be a pretty valuable video at that so this is going to really apply to different businesses such as a real estate company where you get data when it comes to a property and we repurpose it for an article repurpose it for social media and stuff of this nature now you see behind me the uh automation form the Google form that we
[00:31] have here at webcaf AI where essentially we receive data on a you know an automation one built out so what makes this tutorial so special as well is we went ahead and took one of the submissions and we went ahead and you know inspected element and removed their names so for their privacy and we went ahead and found this as one of the submissions here and we're gonna go ahead and build out this specific flow here with a little bit of nuance and essentially we're going to make it so that if you are a business of any context you can still apply it for your business and as you're going to learn the fundamentals of this specific flow
[01:03] that was created here so the business that reached out to us essentially asked for a flow that automates Google Business listening offerings but essentially the way they want their flow created was a trigger event of they receive a specific type of email and then from that email we're going to extract the data and then from that extraction they want us to create a picture of ballet and then create a Google Business listing and then you know gave us the relevant software so what we're going to do today essentially is we're going to show you how to find a specific email we're going to show you how to extract the data from the email but we're going to take it one step
[01:33] further rather than creating an image with openai's Dila we're going to go ahead and show you how to actually extract the image itself whether that be through a Google drive folder or through the email itself therefore allowing you to put like real images rather than AI generated images if you want to do AI generated images I suggest you check out other videos on this channel as we dive a ton into that and then finally they want the end product to be a Google Business listing so we're going to show you how to do that on top of that we're going to show you how to do do it for you know different social media posts and then you know kind of wrap up the
[02:03] tutorial there with that being said these forms are typically handled by our back-end team and we build them out and place them on our Marketplace but if you want to fill it out yourself and see an automation that could potentially be put on this channel or on our Marketplace go ahead and either go to our link tree or go to one of our social media accounts like our Twitter here where we shout out different videos coming out click on this and go ahead and fill out the form yourself so let's go ahead and build this out the first thing we're going to do is stage one let's go ahead and create a pseudo email that has information and the way or the lens
[02:34] we're going to create to today's tutorials we're going to act as a real estate company so we're going to pretend that the trigger event is a specific email comes in because we have a new listing for our company and essentially we want to take that listing and all the information relevant to that listing and start pushing it out automatically to our different streams of organic traffic so I'm just going to go to a random website here so let's just look at Miami real estate so I can find the data for today's video so we're gonna go ahead and choose this picture right here so we're gonna go ahead and I'm gonna grab the picture here and I'm gonna grab some
[03:04] relevant information found here now for your specific use case you would already kind of know operationally what triggers your current event so this event is going to be based off an email so let's say your event happens on maybe maybe the back end of whatever operating system you have for your real estate company or your business you would be able just to essentially set up a trigger where when this event occurs you could send it as an email then everything else triggers as you'll see in this tutorial or maybe it really is just an email when a new listing comes but with regardless let's go ahead and
[03:35] I'm going to grab all the data associated with this and I'm gonna mock up a fake email real quick so I'm going ahead and just grabbing some relevant data here as you know whatever your wave handling backhand for new listings or whatever the context may be you'll know exactly how your data typically comes in so I'm just going to grab this last little bit here okay so I went ahead and sent that pseudo email here as you see you know the data Maybe not be formatted the best but essentially let's just say this is something that happens within our backend already we got the new listing here we got the relevant information about that enlisting and then you know
[04:05] we got our one image and you can probably have multiple images on your context here but we're going to use this one image just to kind of showcase how to use it but that works so essentially we got our underlying received email let's go ahead and jump over to zapier and start building this out so one thing I want to point out real quick before we jump over to zap here is with Google my business it already has a native integration when it comes to creating your posts all you have to do is provide a location summary and topic type what you'll see in this specific tutorial is we're going to just jump straight to basically extract information creating this information but pushing it towards a Google doc as right now in our backend
[04:37] we don't use Google my business so I don't currently have this app yeah I'm gonna come over here to zap room and say create News app and let's go and do this so we're going to do data repurpose and let's start with stage one here so stage one what was requested in our automation form was essentially an Incoming Email uh but we have to identify these specific parameters of the email so we don't you know essentially do for every inbox so we can do email matching search um you know some things you can do is a new label email this really makes sure that you know the individual email that
[05:07] comes in isn't going to be an issue when it comes to finding it but for most cases we can do new email matching search right continue here I'm gonna choose my courses and then when I continue here again and then we do a search string so essentially this is what's going to be what it's looking for when this entire flow gets triggered in most contexts what you can do is in inbox subject so essentially you know we're going to be looking inside the inbox that we received here and we're looking at the subject line so what is one thing in this subject line that seems relevant
[05:38] new listing so this is going to be a fixed text that we identify as we know that anytime the terms news listing occurs in our subject line then we know this is going to trigger this event so we're going to go ahead and hit continue here let's test this trigger and as you see here we got the underlying email here so this has all the relevant information and data associated with that email so what we can do now is we can kind of start working on data extraction and with get gbt we can do data extraction so let's go do a couple things here let's go ahead and continue here continue here
[06:09] now the context of your user message is going to be purely on the data that you care about so for us I'm going to go ahead and just create two chat gbt blocks first one is going to give me like a general overview of What's significant to me and then the second one is going to kind of give me more of um you know interpretation so let's go ahead and begin we're going to say base we're gonna say context uh we are a real estate company who just received
[06:44] an email for a new listing so you put in whatever your context is real estate company receive the new listing uh email subject email body so we want to give as much contacts as possible so for the email subject we'll do parentheses and we'll put in the email subject on here so right here and then for the email body we're going to go ahead and provide just a body plane that's fine and essentially from here though is we're going to say generate we're going to generate essentially what
[07:14] we want from that specific data what's relevant to us right so we're going to say generate uh the following we're just going to say generate uh would say listing address and then we'll say uh uh how many bedrooms how many bathrooms so this is just like very Niche down specific pieces of information you'd obviously look for
[07:45] whatever Niche down specific information that you care about I'm gonna go ahead and up this model to four this probably could work with 3.5 but 4 is a little bit less of a hassle here we're gonna go ahead and add a memory key so we're gonna do real estate um data this is just so essentially if the output looks good the first time it will make it so that it's consistently good for future outputs and then as you see from our output here we got the listing address the address that was provided and then the amount of bedrooms and the amount of bathrooms so now you learned how to take specific data from
[08:16] an email and extract what you care about that's going to take this a step further here and we're gonna go ahead just for my sake and for your sake to learn we're going to say duplicate here so a lot of that context that we already provided earlier is you know what we care about so I'm going to rename this to specific data points and then for the next part here we're going to go ahead and just do something more General so we're going to say generate uh in generate in three sentences
[08:47] uh a selling copy for ads advertisement and then we can add stuff like tone um uh persuasive see if I saw this right persuasive okay let's go ahead and do uh real estate tone this could be any randitude 32 characters here continue here so based off all this data I mean I could add more context but for now this should be sufficient for you to understand of how you can kind of
[09:18] manipulate this data for whatever the use case you may be using it for and let's go and see the output here but here's the output experience the Perfect Blend of style charm and comfort at our newly listed home in the exclusive Capri lace gated community the this single home boasts four bedrooms two and a half bathrooms sparkling ceramic tile floors private pool a picturesque Lake View offering high-end stainless steel appliances a one-car garage light taking all this information compiling into a three sentence so this
[09:49] in theory would have taken time for you to look at all this and you know come up with a copy and deal for that but now we can do it automatically this is great so now we have potential ad copy you could use within the future parts of this one other thing I want to say that is really cool here is not only can we organically repurpose this content for you know social media articles uh Google my business posts but we could in theory also create Google ad campaigns specific to this listing because of the fact that we integrate with Google ads on zapier keep that in mind that's something cool
[10:19] so from here though let's go to just do what that client had requested in that automation form and instead of jumping it straight to Google my business as we don't have an account here we're just going to do Google uh Google Docs and then provide what was what would have been needed anyways in Google my business which was location um let me see what else location summary and topic type so from here we're going to do create an event we're going to say create document from Templar say create document from text I continue here I'm gonna choose our account here so we'll do our courses
[10:50] account and then we're gonna go ahead and just put the document name to that address we had earlier here in order to identify you know that would be the location parts so we go ahead and actually before we can do that we actually need to format this data a little bit so we're going to format uh we'll do formided by zappy so formatted by zapier essentially what it is what allows us to manipulate data on charge GBC outputs as if I would have continued with that flow that's all in one output so we're going to do a split text function and essentially what this does is this is going to allow us to split the data into multiple
[11:21] different points that we can grab so this is going to make more sense as you'll see here um for now we're going to just do a separator of new a new line and you might be saying how did you know dispose like this uh essentially if you just come over here hit more this is some documentation you can look at let's go ahead and choose all as separate Fields this is going to allow it so that the output as you see here isn't just one big block of data we're going to separate it by each data point here so we can grab it and use it for whatever context we want to do so I want to test this step and as you see here we got three separate data points we got
[11:52] listing one or item one item two item three from here we can come over here so we're gonna say our document name it's going to be the formatted version we'll just save the listing address the document content will be the uh you know we're gonna come back over here we're going to use this Chad UT output like the ad copy we created as firstly reply here and then the footer uh this is essentially not the footer but the folder is where it can be placed within your Google Drive so maybe make a folder dedicated to this but in this context we're just like you know mimicking uh mimic a Google Business posts and then
[12:24] the other part is going to be the topic not in the context of choosing the topic um I believe it's called topic right yeah topic type that's typically fixed text right you already have you know specific topic types that exist within your Google my business account so this would just be fixed text you know topic type um new property and then essentially if I go ahead and hit continue here in this context as a pseudo Google my business post we should see a document created with that name with this information and essentially with the topic type of new
[12:55] property here you go so as you see we got our topic type of new property we got the ad copy paragraph we made earlier we got the listing here now one thing I want to show you real quick so I'm going to delete this I'm going to show you how to use some HTML real quick so if you want to use that in whatever context you would want to do it you learn how to do that as well so I you see here we have we're going to go ahead and make this bold and we're going to add a little bit of indentation for formatting reasons so in order to do that and what's great about uh document content is it can handle HTML so we're just going to real quickly add some HTML here if I can do that real
[13:27] quick so we're going to do b stands for bold and then essentially in order to make a line like that you just need to do a break so it's going to be BR and then I'm going to hit continue here and I'm going to test this step again and as you'll see It'll be it'll come in formatted which is nice and what's great is you can actually format Chad gbt outputs which is cool so as you see here it comes in formatted we got the Bold we got the break and so on from here so let's go ahead and see how we can keep repurposing this content for other use cases past a mimic of a Google my business post okay so to start off with the repurposing the content pass the
[13:58] Google my business post we're going to go ahead and just post to Twitter so as you know Twitter's API or maybe you don't know it got deprecated so we're going to use a scheduling software called buffer to do it buffer is completely free so you can go ahead and start using it right away we're going to say add to queue then we're going to go ahead and continue continue here and you know just to show you how crazy this can get I'm going to add another chat gbt block here let me go ahead and just do this real quick we're gonna go ahead and create a tweet specific chat gbt block so we're gonna do chat gbt and when I say uh conversation
[14:33] continue continue and then we're going to go ahead and say based off this property listing I'm going to add a context property listing if I can spell property uh generate a tweet and we're going to do then we're going to give the context here we're going to say uh content property listing property listing information and then essentially from here we can go ahead and provide the previous data points we
[15:04] got so for example we got address and then we got uh bedrooms actually we don't need to do that we can do address and then we can do um you know more information and essentially what we can do here is for the address we'll give that formatted piece of data that we had earlier here and then for the more information we'll go ahead and give that other Chad gbt output that kind of condensed all that information into one little block here and we're going to Upper model to gbt4 here I might need to
[15:34] do more prop structure here but let's go ahead and see real quick it's going to be real tweet real and I'm going to continue here and test this step and here we go so here's our tweet just listed discover luxury in this four bed two and a half bathroom home in the cabri lakes and then gives relevant hashtags Associated this as well you can also add fixed hashtags here what we can do here though so we can get away of these nasty little quotation marks to add another formatter block and essentially just say you know the output here but let's remove it so we're going to do text
[16:05] and we've already transform and we're going to do the events of replace I'm going to hit the input is going to be the output that we have here so let's go to reply and essentially all we need to do is find the quotation marks leave of empty space and then hit continue and test this step so now we should get the exact same tweet there we go I'm going to move this up here and then from here we can take the output of this formatted text here put it right there we're going to choose our Twitter account here and then we're going to choose our media so our medium media in this context is
[16:36] going to be an image the image is going to be that attachment that we had up here so let's go ahead and get all attachments and then from here we can add alt text we don't really need to do that method EQ we're going to say you know this would be whatever the context you may have do the fact that this is going to happen every time a new email comes in we can just share it now we're going to hit continue here and then let's go to test this step and we should see it on our Twitter okay so we actually went into an issue here so we got a message that's too long it's actually good that happens so what we can do here is we can ensure this doesn't happen in the future
[17:07] we're going to go ahead and say um generated tweets and add a format block here say formats Max uh uh either do a character limit or sentence limit I'm going to say max amount of sentences in tweet is we'll say two just for now uh in order to really make sure it works um I don't remember what Twitter's exact character limit is that might be more effective use whatever it is let's say the Character level is 250 go with
[17:37] something a little bit lower so go up like 220 210 this will really ensure that there is no issue in the future all right so we got our tweet here which is going to be a lot shorter and still has the hashtags we're going to go ahead and hit Test action we'll re-test this step all right perfect and then we're gonna go ahead and test this step again and now it should work so I'm going to jump over to Twitter here and reload and let's see if it shows up here and there you go so based off our information that we provided we got some hashtags we could add fixed hashtags here and we got our image of the link we can even add links here if we wanted to
[18:08] but here is one example of repurposing or a second example of repurposing content which is going to be for Twitter so I'm going to delete that and let's look at another example so the next example is gonna be pretty obvious here Instagram for business why would you not post on business uh on your Instagram so we'll just go ahead and say publish photo the photo will essentially just choose the all attachments that we had earlier here and then for the caption we can go ahead and just you know maybe re-proctor this original caption here so let's go ahead and come over here and we'll say generate a Instagram caption now as I
[18:39] said before you can add tones to this you can add you can request specific hashtags but I just want to show you how to repurpose content so we're gonna go ahead and say uh max amount of sentences in Instagram caption is two and actually we'll say one so we can kind of get a different output here and then we're gonna hit continue it already would in theory I'm out of one here to refresh the memory key so it's you know starts on a clean slate it would in theory already give us a different style of dictation for the Instagram because it's gonna be basing of its data and what it's you know
[19:10] obviously an Instagram caption and tweets are drastically different in the way people talk in them so from here though we got our nice little Instagram caption I can come over here and hit the caption part here we can add and this one didn't come up with the quotations we need to leave it in there and you can tag users in Instagram which is nice you can also provide the location specifically here as well I'm going to continue here and test this step and there you go so our Instagram is just like AI generated images but as you see here it pulls up here we got the relevant uh caption we
[19:40] could have requested uh hashtags as well but it pulls up here automatically really nice stuff I'm gonna go and delete this from here the capabilities are Limitless essentially now you learned how to repurpose content in whatever context you may receive deep data and need to repurpose it for other streams as you know the way this all started was just through a simple email we were able to extract the data from that email we gave a mimic of a Google Business post so leave a like if you feel like you've learned something up to this point as it shows us here that you'd like tutorials that are more structured like this we went ahead and took it one step further here where you
[20:11] restructured the incoming data to make it into a tweet make it into an Instagram post and essentially any other social media content you'd want to publish it to I want to kind of leave this tutorial at that essentially if you want to take this even further here and maybe you wanted to post a blog article on it or just an article showcasing the property and its different functions we do offer this AI article generator where essentially this in theory could grab the context that you provided from that email and create an entire article based off of it so use that at your discretion
[20:43] if you want to do something of that realm when it comes to content but without further ado this kind of covers what we receive in that form now you understand essentially how to take data extraction repurpose it for other use cases so you can stop doing that and saving you a ton of time I feel like you know something make sure to like the video it's completely free and it helps us here at corbini if you want to learn more about how to leverage automation artificial intelligence and stuff of this nature I'm gonna leave your playlist at the end of this video it's called integrating 5000 apps of zapier and showing you how to leverage
[21:13] AI with it really powerful stuff you can learn a ton more in that playlist without further Ado I'll see you in the next video thanks for tuning in and yes surprise I'm an AI Avatar make sure to explore more here at Corbin AI where we demystify AI for your personal and business life until next time foreign