first copy this link paste it in here click on run and you've got clean data in just 5 minutes put that into chat GPT and we've got the most valuable thing of them all graphs look at this gra so in a minute I'll show you how to scrape almost any website turn that data into a beautiful CSV file and get quality insights from that data even if you never scraped before without any technical experience anybody can do this and by the end of this video I'll show you some valuable Chachi BT prompts to analyze your data to get valuable insights but first why use web scraping well imagine that you're an e-commerce owner and you have hundreds if not thousands of people that sell the exact same product as you with good web scraping technology you can check their price and you can check if they have available stock and for example if they're out of stock you can bump up the prices or if you check their price you can undercut their price which gives you a competitive Advantage just imagine how much more you would make then before this used to be very hard to do with a lot of HTML code that you had to pars through to get the valuable data but now since we have good tools that do this you can do it very easily and that's what I'm going to show you now step by step so you can get the most valuable data and the insights from it really quickly so step number one is to go to site. com they're happening to be sponsoring this video so I can go in depth and show it to you step by step they're one of the easiest and best web scrapers I've found all you need to do is click on this button try for free now that you're on the inside I want to show you how to create a new project and scrape a website and I've done this many times before to show you a wide variety of sites that you could scrape but all you need to do is click on create project then you just name the project so let's name it ebike Amazon and then I'm going to click on sites AI powered spiders this is a no code solution for web scraping and you can also deploy your own web scraping code to the cloud I haven't used this before so I'll go to the AI spider then I'll click on create project and the reason you want to have something like this is because if you go out there and try to scrape on your own what you run into is your a bot cap shot and things like that and the way that these websites go around that is that they go to the exact same website through many different IP that looks like just hundreds of people going to the website and there's no suspicious activity to continue we'll just click on e-commerce and here I can name it whatever I want I'll do ebikes Amazon and then I'll find the website that I want to scrape so let's go to Amazon as you can see enter the characters this is why we use web scraping I'll search for ebikes as you can see we have have a bunch of different ebikes and ebike related gear I'll actually say electric bike and then I'm going to copy this link go back into site again paste the link and then we could do geolocation so if it matters where you want to scrape from basically the browser is located in this country we'll do United States they say that this geolocation has a higher price so they recommend you to run a sample before scaling up the volume we can put the max request at th000 10,000 and that will just dictate how long it will take to actually get all the results but I did like 1,000 before and it took about 30 minutes or so so it depends on what you want to do I'll leave it at 100 for this example then we have extraction Source I'll leave that at default and then we have a crawl strategy full allows you to follow most links within the domain of URL and then attempt to discover and extract as many products as possible as you can see for Amazon this is very easy CL it will go all the way to the bottom extract all the links and data from these and then it will click on the next page and then it will start extracting the data from this page here as well if you would have a different link from a different website there may be times where you should use the navigation or pation only to extract the data more easily I'll keep it at full it worked for the previous example now all we do is just click on save and run and just like that you're a web scraping expert almost until we go to the next step take a look at the data as you can see it's already started it's runtime and in a minute you'll see items and requests start popping up here but I already did this before call me the Gordon Ramsey of AI I already baked the cake before I started filming the video so I scraped three websites before I filmed this video one of them being zillow. com searching for Manhattan New York and say you're a real estate agent and you don't want to sit here scroll on Silo for hours you can get very clean data the second website I scraped was from Fel Raven which is one of my favorite outdoor clothing brands and this is more for people that want to scrape Shopify stores or e-commerce stores that areen amazon.
com just to show you how that works and lastly I showed you Amazon so let me show you how the data looks like as you can see from before what I did was ebikes on Amazon with 400 requests and that took about 19 minutes which is pretty crazy when you see how much data we actually got I'll just click on the items here and as you can see this is the data that you're left with you'll get the availability the brand the breadcrumb from the URL you get the color the currency the entire description as well as it in HTML the features including all the images $999. 99 design size style and the URL and of course you didn't just get one item we got like 217 items and you might be thinking hey this data looks cool and all how am I going to use all of this well I'm going to show you that a little bit later then I'm going to show you how the data from Feller Ren looks like so here we got 89 items let me show you a clean one here we got a classic cap in G1000 ventilation holes an adjustment at the back small leather Fen logo in the front and if you're wondering oh what type of colors do they have you just start scrolling down and you can see all the different colors they have including the sizes and the price to just give you a taste of what it looks like here with websites like Silo you get the description for beds two bath the square footage you even get the price here as well as the street name so you can see how this is very valuable without having to look through this men ually or God forbid write all of this down manually or copy pasting data entry but this data is not as valuable as when we enter step number three analyze it with Chachi BT the beautiful thing with site as a whole is that you can just click in the top download and here you get CSV Json or XML I'll do the CSV file and now all I need to do is just for example let's do the ebikes data drag it into chat G PT and why it's important to do data analysis with chat gbt is to get insights from the data as you may seen it's a lot of data that you might not be able to look through yourself checking all the individual boxes and with jbt you can easily ask questions about the data it can read all the data effortlessly and give you the insights that you need like what's in stock what's the pricing what are the most popular colors or give you reviews so give you some prompts right now one of the favorite overarching data analysis prompts is act as an expert data analyst and give me graphs that can help me analyze this data set right away you can see that it opens in sort of an XL clone or if you like to look at tables here you go it it's really cool how they have added this into Chachi BT now and then it will start writing that some potential graph we could generate is brand distribution price distribution rate in availability price and so forth and then it will start analyzing and doing the job for you first it started by cleaning and structuring the relevant field and then it says let's proceed with creating these visualizations where it starts with brand distribution of electric bikes price distribution of electric bikes rating distribution of electric bikes a pie chart of availability where it's 100% in stock as well as a scatter plot of price versus rating this could easily help you hone down on what type of electric bikes you want to sell in your store even so we're already getting a lot of insightful data and the roads that we can take down this is endless who needs video games when you got this honey honey no vacation I want to sit at home and look at graphs all right please make a graph of the top 50 most expensive ebikes above four stars and here we got the data but it doesn't look so clean so I'll just prompt it a little more this looks a lot better even though it's not 50 you can see the most expensive is the free go X2 let's only ask it for highest rated and this is the graph that we get sometimes you can ask it to give you a smaller font and it will be easier to read but yeah all of these to the right are five stars and if you're not a graph person but I mean who you're kidding you could also ask it to write it out in words and here you will get it as well as you can see number one this has the highest rating with a number of ratings to 5 to 18 so this one seems like the highest rated with a lot of ratings as well as you can see giving you very valuable insights in just a couple of minutes but let's take a look at the other data set here from Feller Raven can you sort all these items by price and show me a graph with price and name and out of the entire data set set you can see the high Coast hydratic Trail jacket men is the most expensive a little bit above $250 again we can ask it for can you make the font size smaller and now you can read it a little bit easier but let's see if we can ask it for the most popular product on the website based on reviews it will analyze the review count and rating from the data set and it said that the product conen received 103 reviews with an average out of 4.
