what's going on guys it's jono and in this video today what we're going to be talking about is how you can webs scrape leads off of upwork and then automatically score them filter them out and then how you can create a message that is custom tailored to send off to every single job that matches the criteria of what you're actually looking for and how we're going to be doing that is essentially by using make.com make.com is an automation platform that connects uh thousands of applications together in this case we're going to be using something called
appify which is a web scraping technology where we're just going to scrape different listings off of uh upwork and then we're going to classify those listings based on your criteria what your interests are what you're good at all of that kind of stuff and then we're going to use chat gbt to filter it out and then we're going to dive into the listing pull additional information out of that listing like for example you know the hourly rate we're going to see how long the projects are the duration that kind of stuff just to make sure
that it it actually aligns with what you're looking for if you're in the United States and somebody's putting up job posting for $20 you're probably not going to want that so we'll we'll we'll definitely filter that out and then we'll create a nice little proposal and put this into a spreadsheet so that you can take all of those proposals once a day and you just apply right you click a couple buttons and you apply it's going to shave off so much time now just before we begin guys two things I want to point out here
if you like this tutorial and you want to follow along then uh if you want the blueprints rather you can just click this more button and import the blueprints straight into your.com account they are going to be free in the description down below if you don't have a make.com account please sign up via my affiliate link and I will send you over my top 10 favorite automations using m.com and if you guys are also new to this channel make sure to subscribe down below because if you like content like this that hopefully helps make your
life simpler then I'm sure you're going to like all the other content that I release as well so with that being said let's dive in right [Music] now all right so let's into this we're going to first create a scenario on the top right corner there and we are going to focus on scheduling before the trigger here right now in order for us to actually get to scheduling I'm just going to set up a fake trigger here just because it's not going to let us focus on scheduling until something set up here so I'm just
going to go ahead and click test and test for the value and name and save and this is important that we save the scenario so that afterwards we can start scheduling now my face was blocking the scheduling uh parameter down here but you can see down at the bottom left corner of your screen it's set for every 15 minutes which is great right but we want to change this so that it's at a better Cadence so for us what I'd rather do is I'd rather have this fire every single day and it doesn't necessarily matter
uh I guess it does I would schedule this before you kind of come in and look just making sure that maybe you're scheduling it for I don't know let's say 6:00 a.m. on uh on your day right so that's just an easy way for you to go ahead and um and make sure that every single day at a specific point in time this is always going to run collect new results so all you have to do is instead of going to upwork having to log into upwork find all the results you just have to take
a look at what's in our critical spreadsheet and say hey this looks good to me we're going to apply here copy and paste and you're good to go right so that's essentially it so that's the trigger in a nutshell we're going to make sure that this runs every single day once a day right so now that that's set up I want to move over to the actual trigger here so we're going to go ahead and delete this now that this is all functioning properly and we're going to use something called a web scraper to pull
these results in now before a web scraper we could have used like upworks RSS feed but the problem is is that they actually discontinued this so they removed in August uh 20th this year at least 2024 so that's not longer that's no longer an option so we have to find an alternative to essentially scrape these leads from upwork and how we do that is through a platform called ay.com now for the majority of you guys watching this you're going to be able to use this pretty much completely for free um they do give you $5
of free credits and that's typically like lots maybe hundreds at a minimum but more like thousands of web scrapes every single month and it resets every single month so uh just keep that in mind you can go ahead and sign up for this I do have an affiliate link down below if you guys are generous enough to help me out with with that and this is your first time you can use that link that would be much appreciated and once we're inside we're going to go over to the store tab here and we'll search for
upwork now once we search we have a list of tons of different upwork um web scrapers that we can use and for the purposes of this tutorial I want to use this one especially too because it has the most amount of people using it so I'm going to go ahead and use this amplify scraper upwork jobs scraper you can see that it's $4 per 1,000 results you can see down here at the bottom as well that uh you do get $5 recurring every single month so essentially you could have 1,250 WIP scrapes completely for free
so that's a lot of jobs to be applying for especially if you're a freelancer that's that's that's a crazy amount um so that's pretty cool right it's going to give you access to a lot of jobs so let's go ahead and let's set this up I'm going to go ahead and restore this to a def the default parameter so that we can work through this together so in this scraper you're going to see the job keyword is full stack developer okay this is pretty much good to go um what we just need to do is
click save right so down at the bottom here you can click save e either at the top right or at the bottom and the reason why we're saving it is because we want to essentially load this in to our make.com scenario so back in make.com now that we've saved it we can actually access it in make.com so we'll click this massive um plus button in the center of the screen and we'll load in appify here and we're going to run a particular actor right for you guys to set this up you you will need to
connect your account there's an ad button and you will need an API token you can get this on the free plan at least at the time of this recording you're just going to go to settings API grab it and install it there and then you'll have your account good to go now once inside we're going to choose the correct actor all you have to do is click this copy uh to clipboard button here and then paste it in the actor and because because you have saved it it should load up here now in my instance
it's not loading up because I selected the wrong account so I'm going to go ahead and change that account and I'm going to now find that and just in case you want to double check and make sure that it is the right one you can also click map see the ID here and it's going to be the same ID in the URL up top as well so we are good to go and the next thing we're going to do is we're going to run a uh we're going to run synchronously now what synchronously means here
is that if we take a look at this particular workflow it's not going to move to the next modules until the web scraper is complete and that's a big deal right because web scraping takes time it's not like an pretty much an instant request like some of these other modules it might take 30 seconds it might take 45 seconds or 2 minutes or 5 minutes or whatever right so we're going to wait for it until it's done and then only once it's done then we're going to go on to the next results because all of
these results require the data from the web scrape to move forward so it's going to draw an error if you don't wait now if you end up just pulling in so much data every single day and it takes over 120 seconds to load you're going to it's going to time out this and it's not going to work so essentially what you'd have to do is you'd run the actor synchronous or not or asynchronously and then you'd pretty much do everything else in this workflow on a new tab so you just start the workflow by essentially
having a new module down here and it's going to be appify watch actor runs and you just continue on so you have a two-part two-part sequence first step is just running it second part is just essentially grabbing it if you guys think you're going to be loading in tons of results for probably 90% of the people watching this that's not going to be the case so you can just continue to follow along with me but that's the one modification i' make if you're going to be web scraping in hundreds and hundreds of thousands of different
results every single time you're doing this now all of these we can pretty much leave as default it's going to load in uh the default build timeout and memory in appify which is in the Run options down here typically speaking you want to make sure you're using the latest build um you know because over time these guys are updating these builds and making sure it's up to date nothing breaks so obviously the best case scenario is the latest build unless it's super new which then it can also have bugs but usually that's not the case
they have a timeout a default timeout where it's just going to cut off after let's say 3,600 seconds and the certain amount of memory that it could use you can see here on your free plan and you're going to get 8 me gigabyt of memory so you could allocate more but the defaults are good enough for this particular run so that looks good to me all we really have to do is we have to just come in here and choose the job keywords so this can be whatever it is that you're looking for but if
you just you know come to upwork and go to find work and find work for your skills you can see tons of different job opportunities here and whatever floats your boat you can pick it and you can add it in there you could even have multiple runs potentially by having multiple scenarios or just having you know a list of different uh list of different job options that you're interested in running the scraper for every single one of them for this tutorial we're just going to focus on one at a time so with this input Json
I'll go ahead and just choose um automation let's actually see what they have here for automation okay they don't even have something for automation uh okay anyways we're going to we're going to go with make.com just for this scenario make automation sweet so that looks good to me we're pretty much all set here the only other thing that I will suggest is just putting 10 as your Max results especially for testing because if you don't you might blow through your $5 in like one run if you're not careful so just keep that in mind too
so we'll come back to the Json and all we have to do is just copy all of this and circle back to this input Json field here and paste it in there right so you can see all the Json here it's not formatted particularly great but you can see the search queries here make.com you could even go ahead and maybe do like go high level right so that's another way to to get multiple search listings in there I believe okay so that's good we'll click okay and just before we actually go ahead and run this
I'm going to add in one more module and this is going to be an appify module we are going to get the data set items so this once it's done running it's not going to provide us with any data right all it's going to do is tell us hey the run was successful here's the information on that particular run but if you want to grab that data we need to have another module here and that is the get data set items and all we need to do is just one thing here in the data set
ID we just need to find this field here which is the default data set ID and plug that in here and we're good to go we'll click okay save and we'll run the test results which is the Run button down at the bottom left side and we're going to wait until these results come in now keep in mind web scraping does take a long time we're only scraping 10 results here and you can see how long it can take so if you do end up scraping tons of different results and you hit that 120 second
threshold then it's a good call maybe to break this into two separate workflows [Applause] okay well that's that's pretty much it I have nothing else to say until this is done so looks like we're just going to be killing some time together right now man so how's your day been so far tell me about your life what's going on with you I'm just kidding it would be fun if you could talk through the camera okay here we go um we've got the results here and you can see that we have 10 different bundles now uh
bundles is like the jargon in make.com essentially just think about this as a list and you have a list of 10 different results right we just web scraped 10 different results off of make do or off of upwork regarding make.com so just because we added in make.com as the keyword doesn't mean that it's going to be necessarily in the title it could be in the tags like it is like like this for example could also be in the description it's really anywhere in the job listing that it's going to list make.com now before we continue
on with this you can see how big of a freaking pain and AET is to run this every single time you're going to want to like just honestly quit Automation in your whole life if you have to sit through two minutes every single time just running this all like like it's just it's just so frustrating right your computer might be broken by the end of you developing this workflow so what we're going to do is we're going to speed up this by loading in a parse Json module and we're going to use this as test
data so that we don't have to run it every single time so when we open up the results in uh this get data set items which is like this cloud like like thing here we can click on the download option up top and download output bundles and all we have to do is copy all of this and then Circle back and paste it in the Json data here click okay we're going to unlink the two here and this is going to be our new starting point by dragging the trigger from the appify module over to
Json we'll click save and we'll run this one time it's going to give us an error because typically when you're parsing Json or transforming data you shouldn't end a workflow with transforming data because it's like why would you end something why would you transform something and then not use that transform data because that's the whole purpose of these automated workflows for this purpose it's okay just to run anyways because we're just testing and we haven't built up the whole work but you can see all of the results in this parse Json that we also had
in this get data set items and we can just keep running this over and over and over again we'll ignore These Warnings um because my OCD is going to kick in I'm going to get frustrated with that as well but uh that's essentially how it works now we don't have to wait for the um po particular web scraping to be done okay cool so that's the first step that's that's completed right now the next step here what I want to do is I want to um filter out any results that might not be applicable so
what I mean by that is I want to make sure that the jobs that we're applying for actually make sense for you to apply to right there's no point in putting BS jobs on there that you don't actually want so let's filter out some of the lower quality options and those lower quality options doesn't necessarily mean the job is bad it just means that it's not the right fit for you so we're going to go ahead and choose open AI here and we'll create a completion for the particular model I'm just going to use 40
mini I think that's going to get the job done just fine so we'll go down to 40 mini down here and we're going to filter out the results now I'm just going to copy and paste them in here just so it's a bit faster so that you know we can speed up this video a bit but um we're going to have multiple different messages user system and assistant messages so the first one is going to be a system message what a system message does is it just essentially tells the system or it tells chat GPT
how it should you know essentially you're just giving data for it to understand the context of the prompt better so it knows how to better respond right so in this case we're just saying hey you're a helpful marketing assistant that is great at finding the right jobs to apply to and filtering out those that aren't the aren't right for the next one is going to be a user response this is going to be what you essentially provide to chat GPT these are the instructions that you're going to be giving to it so that it understands
what it is that you're actually trying to do and with the user message in this case we're going to say we need to determine whether or not this upwork job is a good fit for our company reply good or bad and then we're going to pass in the particular data here um such as the title description and also interests as well now you can see we're mapping stuff here essentially in make.com if you guys have never used this before you can reference past modules we're just referencing past modules the data that we received into the
future modules here okay so that's pretty much good here I'm just going to delete this interest and we'll Circle back to that in a few minutes here and then essentially with the last couple responses or the last couple um messages here they're both going to be assistant messages and we're just essentially telling chat gbt in an assistant mess Mage how we should respond right and we kind of laid it out up here where we're like hey please either reply good or bad those are like the two options that we want so we're going to give
both options here we're going to say good and we're also going to say bad so this is going to be an assistant again and bad so those are really the two outcomes that we're looking for and then in addition to that we just want to make sure that we add in some interests up here as well okay I'm just going to scan this make sure I didn't miss anything okay so that looks good to me the only thing that I want to do here is I want to add a module and this module is going
to be a tools module and it's going to be set multiple variables in the first variable I want to set interests right so in the interest section this is kind of where you can specify what it is that you're looking for what it is that you're not looking for right so I'm just going to come in I'm just going to copy in what I essentially created here is like test results but I said I'm interested in helping automate businesses or or sorry business tasks using software such as make.com and go high level I want to
make sure this is the main focus and not just a small piece of the job I'll click okay and pass this into the filter as well okay sweet and we'll go ahead and pass in the interests in here as well okay so that sounds pretty good to me so far now just a couple other things that I want to do here is I'm actually going to add some stuff before this particular module because I want to make sure that in this filter module we can filter out things that are not or that don't meet our
specific qualifications like the amount that they're paying the duration of the job maybe you guys don't want onetime projects maybe you want to work with clients for months at a time maybe you don't want to deal with clients that only pay 20 bucks your rate's $100 an hour right all of that kind of stuff so circling back I'm going to save here circling back into the parse Json we can take a look at some of these responses so we got title we have the description down here we have the URL we have all of this
kind of stuff so what I'm going to do is I'm going to take a look at this URL here and I'm just going to paste this into a new tab so upwork.com job salesf fun automation expert needed and here's the exact job now there's a couple things that I want to grab from this p page and it's ultimately up to you guys you're Happ you're you're welcome to grab anything you want to but the things that I want are these six Fields here I want to know how long is the job what the duration is
what the experience level is what the pay is remote work all of that kind of stuff right so this is what we're going to be grabbing in this next module here so we're going to go ahead and create another chat gbt module and actually instead of creating this what I'm going to do is I'm just going to clone this guy here and it's just going to speed up the workflow a bit more cool actually hold on we need to do one more thing before this we're going to add another module in here and this is
going to be an HTTP module we're going to make a request and what we're trying to do here is we're trying to find all this data right so we want to use chat gbt to scrape this data off um but in order to get chat gbt to scrape this data off we need to give chat gbt the information now if you just go into like chat gbt like chat gbt you can say hey analyze this website and they'll do that but in the API version of chat GPT which is these modules here you can't tell
it or instruct it to go to a website analyze it summarize it give you a response back it doesn't have that capability right now so we can't do that in order to overcome that we'll use an HTTP module and we'll make a request and the request is is going to be to a URL specifically the URL of the particular job listing here so we need to hardcode in here https upwork.com and then we're going to add in the URL here there's not going to be a backslash here because the URL as you can see already
contains that backslash so we're good and that is essentially it we'll click okay if we unlink this and just run it one time we'll go ahead and we're going to scrape this this data okay sweet I'll just stop this for now because we got quite a few results but in the output here we can see the data and it's a lot of HTML CSS all of that kind of stuff which is obviously not what we want right so we need to strip that HTML out of there but that's okay for now what we're going to
do is we're actually going to pass this in to the chat GPT module similarly to before I'm just going to speed this up by kind of copying and pasting what I've already built over here um so we're going to have multiple different messages user messages system messages and assistant messages now the first thing here is a system message your helpful marketing assistant that is great at finding the correct details and upwork job listings so you're prefacing it with the context that hey you're going to be coming to this website site you need to grab certain
job listing details and we're going to specify that in the user message which is going to be next here and the user message we're going to let it know hey please analyze this page and you can see here that we have the strip HTML tag here so the strip HTML tag can be found in the um functions in make.com so these are all kind of amazing amazing built-in functions that are going to be a GameChanger for you if you haven't learned the functions I highly recommend taking some time to learn them because you will spend
an hour or two learning it but then you'll inevitably win back days and days of your life um just by knowing this stuff right because you'll overcome blocks so much faster I do have a video I'll link in the description below where I go over every single one of the 74 make.com uh functions here but for this one we're going to use the strip HTML and this is just going to pull out all of that HTML we're going to pass in the data here so practically what that looks like is if we kind of come
back in here and look at the results um and we go to um we go to the output here you can see that there's just a lot of HTML like this isn't required this isn't required this isn't required it's going to strip the title tag here and it's also going to strip the the end case or like the both the start and the end of the title tags but it'll leave the text in between so we're just going to be left with essentially text at the end of this right so that's how it works we're
going to pass in all of the text of this page including this we're just going to ask chbt to essentially like get all of this information off of the page right and how we're going to go ahead and do that is we're going to essentially tell chap gbt that the output needs to be Json data so it needs to be Json data structured like this and we have six uh or seven different fields here and these fields mimic essentially what we find here so the first one is payment type the second one is duration and
then experience uh payment type uh location and project type and it's going to pull in all of these for us and then we're just denoting what uh type of data structure it is there's a lot of different data structures when it comes to Json Json is essentially this right here but also you can think about Json is this like key value pairs key is the title value is the title's name right the key will never change so every time we add a new field to this the key will always be title and that's how we
know you know where to put the title in but the value is always going to change depending on the particular job so this is just a way of formatting and structuring data so the key will never change but the value here is denoted by a string or for example a number there's seven different data types but yeah that's essentially how we're going to structure the data you guys can ultimately go ahead and add more data types in here if you want to um you could be like for example coming back here and you could be
like I don't know uh tags you could just add in tags and you could say tags equals and this data type will be an array right of multiple different values you could even put in a text array here just to know what exactly it is just make sure that when you're dealing with this kind of stuff the last value or the last key value pair doesn't have a column at the end of it or else the Json will break and she'll spend a lot of time trying to figure out why it's not working properly so
we have the Json structure good to go the last thing we need to do is show advance settings down here and we need to make sure that the response format is changed into Json object and we're parsing the Json response that way it's not coming back to us like text like this it's coming back to us like Json data that we can use and make so that's essentially what we're doing here and I believe we're good to go all we need to do is change this last field here which is the score so we have
a particular way of rating these job posts so that we know what exactly it is that we're you know we How likely it is to fit the criteria of what you're looking for so coming back in here I'm just going to copy this in one more time to speed everything up I'm going to click okay we'll save this and circling back to here this set multi multiple variables this is like where I'm just going to add in all the different things that you guys probably want to change about this workflow so all you have to
do is just come into this one location change this and you're good to go we'll add in an item here and we're going to call this score and then we'll Define what that score looks like so I'm looking for a job that pays uh you're just going to tell it what it is that you're actually looking for and then you're going to rate it on a fivepoint scale for points assigned to every single uh well some some variable right so in this case we're saying I'm looking for a job that pays a fixed rate is
at least $50 USD or more and that has a duration of over one month let's just say I don't know scoring is out of five and here are how points are awarded with a spelling mistake in there on purpose I'm just kidding that was definitely not on purpose two points if it's super related to make.com one point if it's over $50 USD one point if it's over a month in duration one point if it's a fixed rate and one last thing here that I'm going to change is I want to make sure that this can
get half points if it's potentially um somewhat related to make.com so I'll say or one point if it's somewhat related to make.com okay sweet so that looks good to me I'll save this and I'm going to plug in that score rating here so that it's going to automatically spit out a score when this particular chbt prompt runs so we'll come in here and we'll change the score and we're good to go so now what's happening is we are pulling out all the data and we're also pulling out the score as well and what we can
do is we can actually unlink this and we're just going to run this module and we're going to see how well this works okay sweet so we can take a look and we'll look at the results here so for a job like this it's saying hey this is like not really related to what we're looking for it's a one type project onetime project it's us only this is the location expert level less than a month hourly right so it doesn't really match the criteria so circling back in here what we could do is we could
also make sure that in this filter that it's matching certain criteria as well so um circling back we can say we need to determine whether this is um this upwork job is a good fit for a company reply good or bad title is this description is this and then we can come in here and we can add in um certain criteria here so salary is this and let's say score is this right and one more thing here is that in the interests we're going to come back in here and we're going to change this and
we're going to say I'm only interested in jobs that pay $50 USD or more and that are at least two out of five okay sweet and we'll just come back into the filter one more time and just make sure that in this particular prompting we'll make sure that all jobs need to have a score of over two and must pay over $50 USD to qualify as good and I'll just put that in quotes and click okay and we're done sweet so just two more modules here that I want to go over until we're essentially done
with this workflow then thing is is that like you know you get a job listing like this this is great right I mean it's it's just nice to be able to automate the process of actually finding this stuff because we all know it takes time it's a pain and it's also kind of stressful sometimes too especially on upwork if you guys have used it before I certainly have um where you're going through all these job listings you're like does this make sense you're dealing with all these other softwares and you know it can be quite
quite confusing and you just want to make sure like at it's just nice to at the end of the day have jobs that come in you don't have to think too hard about it you can allocate those brain re that those you know your brain your resources to more important tasks right so what we're going to do next is we're going to make sure that not only do we find these jobs but we also generate a custom proposal so that when you guys find a job in our Google spreadsheet which is right here and you
like you like that is freaking sweet like that is my type of job like that is what I want to be uh you know working on then we'll go ahead we also create a message for the person too that we can send over to them so that instead of you having to build this from the ground up it's going to at least have at a minimum like a first draft done so that either it's good enough just to send or it's good enough for you to just do some minor tweaks on and then send it
out after you're happy with the outcome so circling back here we're going to add in one more module so I'm actually going to uh copy this over so I'm just going to copy this module from this workflow and paste it in here and then we're going to connect this in sweet so okay and I'm going to move these um appify modules this is the one thing about trying to align your workflow sometimes it doesn't work that nicely sweet okay so we got this uh module in here I just realized I put in the wrong location
we're actually going to put it at the end here and move this in two things before we even set this up I want to make sure that we filter out responses here so we'll filter responses and the result here has to contain case insensitive good if it doesn't contain case andc or good then we're not going to move it onwards right we want to filter out jobs that you don't even want to spend your time looking at right because it's just not worth your time and it's not going to be the job that want to
apply to anyway so um we'll filter out those and then if the job's good enough then we're going to go ahead and we're going to create a job application or proposal for them so in this chpt module here there's bunch of different prompts the first one is a user prompt and we're saying you're applying for on upwork for this job and we'll list the title here we can pull that title in from the initial Json data here and you can say I'm an expert in whatever now coming back in here full circle we'll just set
another variable and we'll title this expert and you can add in like what you're good at here right I'm just going to put make.com and I'm going to link that back into to this particular chat gbt module here so I'm an expert in make.com I'll copy this again and I want to apply for this job please create a compelling response to so why be the best candidate for the role here's the job description we'll paste in the job description one last time actually you could put in the the job description here just to give a
bit more results or if you wanted to you could just put the job description as it is from the web scrape either or will work for you and make sure to have a clear intro provide a link to the portfolio answer any questions the client has thank them for their time and then sign off only answer questions if asked otherwise don't uh create questions and answer them right so this is kind of the structure that I like to take when applying to upor jobs is clear intro link to portfolio ask question answer any questions thank
them for their time sign off and also social proof in there the whole time right social proof is going to be the best uh way to essentially uh sell yourself now one of the other great ways you could do this if you guys are interested in how you can structure these first messages that I think is like a really really smart idea it's just go ahead and sign up as an employer on uh appw work and you can post a job and you know maybe you're even looking to hire something and or somebody and you
can see what the results are that you're getting you can see the first messages that people are sending you and then you can frame your message based on what you're seeing other people sending you so that's a great way to see what the top people are doing in your space to get some creative ideas on how you can structure this message properly okay sweet so coming back in here um sweet we're going to move on and the next thing down here underneath user is going to be another user message so please start by conly assuring
the poster or the job that I can help with the position while referencing the job title here so we'll paste that particular title in here again and please provide this portfolio link to the viewer so we'll go ahead and create another variable in the set multiple variables module and this is going to be your portfolio so this could be wherever it's located example.com and we'll paste that in here it's always really good to have a portfolio personally I would also include a video if you guys are offering a standardized service I'd always have this a
video for that service and included in these first um messages it goes so far I know that for myself when I was doing upw work um man just the the increase in responses that I got and positive responses and the willingness of people to jump on calls when you can deliver a high quality video describing how you can help them out it's amazing I wasn't doing customized videos for every single person I had a generic video for the service that I was offering that was sufficiently generic enough to really cover most the people that what
I would be reaching out to anyways so I can't say enough how uh how highly I think that uh that you guys should be doing that okay so we'll add in the portfolio link here as well and if we'll continue on with this if and only if the job posting has questions please answer these questions here's some basic questions here I'm going to zoom in um please go ahead and add this to your question bank or add on to your question bank because this isn't complete this is just a couple questions to get you guys
going so that you can understand so that you can give inputs to chat GPT so that you don't have to be the one that's always responding to these questions the more like literally every single time you message somebody or every time you get a job posting that you're interested in and ask for a question you should put that question into your question bank here so that you don't have to reinvent the wheel every single time you encounter that same question doing this enough times you're going to pretty much probably be able to handle 90% of
the questions out there at least so this is going to save you a ton of time so I just put in a couple questions here so when can I start I can start immediately how many hours I can do 40 hours are you looking for shortterm or longterm um how do you define success why should I hire you over somebody else are you working as an individual or as a part-time or an agency um all of that kind of stuff right so I'd highly recommend that we'll also put in the um link to the portfolio
again down here and then just thank them for the time and then sign off at this and then we have a system message here that's just essentially saying how we should format the response so we shouldn't have custom variables um don't include HTML or markdown this is just stuff that's going to help you guys send the message faster instead of having to you know do more edits with it so we can click okay and then lastly here we're going to add one more module and it's going to be a Google sheet add a row so
first things first we need to connect into the right account here and then we'll search by um selecting from All In in order to select the right spreadsheet we just need to pull in the UR the ID and the URL up top here so it's just after the d slash and before the slash edit paste that in spreadsheet ID select the right sheet which will be sheet one by default down here um yeah it's just you know unless you change the name you can have multiple tabs but sheet one is just the default here and
then we'll just map in all the custom variables here so title will be the title of the particular job posting description will be description status I'm going to put for as Todo just because we haven't taken any action on it payment type is going to be the uh payment type so in response here payment type will be payment type duration will be duration experience will be experience salary will be salary location will be location project type it's project type we'll put score here uh message that we're essentially getting chat HPT to write which is the
proposal here the live URL will paste in here as well so https upwork.com without a forward slash and then just paste the URL the date is um going to be the date that it was posted so posted that and then found is going to be now right so this is when I just put found but this is essentially when we scraped it and we replied we created this listing for it we'll click okay save this and let's give it a shot and see how these responses come into this particular workflow here just going to delete
these and start with column three or Row three okay so p in the title description status it's hourly payment less than one month to complete it's 70 to $125 an hour uh you can see here that I put the experience in wrong but here's the salary it's only located in the US it's a one-time project here is the message that we generated together um looks like the operation failed here so what we'll do is we'll Circle back to this maker request we're going to do two things here we're going to show advanced settings and we're
going to have a timeout of five seconds and I think that's pretty much it we'll click okay and then we'll also add an error Handler here so we will add an error Handler so that in case there is an issue we'll break and we're going to break for three times so what this means is that if there's an error like this it's going to throw it back in the queue and it's going to try to execute this three more times and after three times if it doesn't work then it's going to give up so we
can click okay okay you see this error here this one we just need to make sure that we include storing of incomplete executions which is right here so that we can save it to recall it a later point in time and run the module once again so we're going to save that and I'm going to come in here and just make sure that the Google sheet has the right experience level set here so circling back experience level click okay save and let's run this one last time okay so it's filtering out quite a lot of
these results as you can see here it's not actually having a lot of pass so maybe what's going on here is our filter step is just a bit too intense right so that's okay what we can do here is we can just say we can just say that um I'm actually going to get rid of this and say all jobs have to have a score of over two right we'll delete this and we'll just start a little bit easier not have such extreme kind of expectations for these jobs right off the bat say I'm interested
in jobs that um have a score of at least two out of five we'll click okay save and run this module one more time and let's see if this passes the filter at least a couple times right so we'll look at the results here bad bad bad so okay we got one result there passing through we can take a look and see so for this it's the it's a full stack Dev with mad skills and a sense of humor okay wow I've never seen a job requirement for a mad sense or a sense of humor
okay got it um in that order full sack then Mad Skills then a sense of humor um if you don't have three I'm sorry you can't qualify for this job job so it's an hourly 1 to 3 months intermediate this is the salary location is us-based um contract to hire and you know here's the message score we can even bring this down one more time and we can just say that hey this needs to be having a score of at least like above zero let's just say to include even a wider pool of people obviously
you know you don't want to probably Target people that are a zero anyways but this can just increase the threshold for testing purposes because with systems like this it's important to we'll test and see what works and what doesn't work it's not going to be perfect out of the box and that's okay um we just need to figure out what works that has a score of over 0 to five and I'll just double check this to see so um all Jobs have to have a score of over zero to qualify we save and we'll run
let's see if potentially the filtering that we set up here is the reason as to why this isn't working properly okay so we're getting another job passing in here we can take a look and see what that job is leaden setup we're looking for an experienced lead genen expert to set up a comprehensive lead capture engagement system fixed price onetime project okay so now we have another job coming in right so these are starting to come in a bit more we have three out of 10 right now which is not bad maybe okay we'll get
a fourth one right so you can kind of play around with this but I think the the key that I want to emphasize here is that it's not a bad thing if you scrap 100 jobs and you only get two or three or four because at the end of the day like you're probably not going to be doing 1,200 Jobs a month you don't need to apply to every single one I would rather just apply personally to the jobs that I know I love and that I know I'd kick ass at because of the jobs
where I'm going to really enjoy the work I'm doing and I'm also going to get the best results for the clients which is going to get me a good score which is going to allow me to get more jobs in the future right because you're scraping and you're you're switching the uh frame where instead of you having to search and search you're hunting right for these jobs now you're just in the process of um you know analyzing being like yep nope terrible don't like it don't like it don't like it don't like it okay that's
good right so it saves a lot of time it allows you to be picky so that you get to choose the cream of the crop to essentially you know help you not only love your job more work on the jobs you love but also get the best results for clients so that's it for this particular video guys I hope you enjoyed if you found value in this video and you're new to the Channel please hit that subscribe button down below give me a thumbs up leave any comments down below if you thought I missed anything
or you'd like to see any future videos or any comments at all love to hear your thoughts and thanks guys so much really if you made it to the end really just appreciate you guys watching so thanks so much and I'll see you guys in the next one bye-bye