How to Make Money by Scraping Craigslist: 8+ Profitable Strategies

Hey, has anyone checked in on Craigslist lately? Lucky for you, I just did, and it's doing fine. In fact, Craigslist is doing great, but we want to make money with it. So that's what this article is about. All the time, I'm talking about scraping this, scraping that. Today, I'm going to talk about over eight ways you can make money by scraping Craigslist and doing really creative, profitable things with that data.

My Early Craigslist Success: iPhone Flipping

But first, let me tell you a quick story. I was in college, and I was on Craigslist all the time. If you remember back in the day, you had to unlock phones if you wanted to use them on different carriers. Well, one day I'm on Craigslist and I noticed, "Huh, a locked iPhone is worth $300, but an unlocked iPhone is worth $350 to $400." Then I went to eBay and saw people selling iPhone unlocks as a service for $5. So I thought, "Huh, $5 to increase the value of your phone by $50 to $100? That's interesting."

So I went and bought two domain names. You can still go to the Wayback Machine or the Web Archive and see the original versions of these sites from like 17 years ago. It was `10dollarunlocks.com` and `20dollarunlocks.com`. What did I do? I spent hours and days building an elaborate scraping tool to go to Craigslist, scrape every single iPhone listed worldwide that was locked to a carrier (Verizon, T-Mobile, Sprint, and AT&T). Then I would scrape their email address and send them an email or a text that said, "Hey, I see you're selling an iPhone on Craigslist. I see that it's locked. Did you know it's worth $50 to $100 more if it was unlocked? Just so happens, I can do that for you at `10dollarunlocks.com`." The second version of that message was the exact same, except at `20dollarunlocks.com`. I just wanted to know how many more people were willing to pay $20 versus $10. It was an A/B test.

It took me way too long to get this kind of gimmick up and running, but once it was, I was making money in my sleep. Literally making money in my sleep because I created an automation where someone would go to my website, check out, put in the IMEI (which is like the serial number) of their phone. It would send that to a guy on eBay that I just messaged and said, "Hey, can you just bill me like once a week? I'm going to send you a bunch of IMEIs, and then I'll pay you five bucks per at the end of the week." He said, "Sure." I would send them all to that guy, and then he would unlock the phone within an hour. I didn't have to do anything. I didn't have any customer support, no follow-up. I just started making $10 and $20 at a time. I was making hundreds of dollars a day in pure profit.

The Power of Data Scraping in Business

Now, as all good growth hacks go, this eventually died and went the way of all my other creative growth hacks. But that taught me something very valuable: scraping is profitable. A couple of years later, I started an iPhone parts wholesaling business. Where did I get all my customers? I scraped 20,000 iPhone repair shops – every single iPhone repair shop in the United States in 2012. Then I had virtual assistants cold-calling them, and I turned that into a business worth over $10 million that I eventually sold in 2019. Scraping, scraping, scraping. So a lot of what I do starts with a scrape. There are a lot of really slick tools out there that enable you to scrape just about anything without any coding experience.

Introducing Oxylabs: Your Craigslist Scraping Tool

But Craigslist was always a tough nut to crack. And don't listen to people that say Craigslist is dead, or Facebook Marketplace killed Craigslist, because everyone can look: it's still getting millions of views every month. Just because you're not using it doesn't mean no one's using it. So let's talk about how to scrape Craigslist with a super easy tool that I wish I had 15 years ago when I built this myself. And then let's talk about exactly what to scrape, how to make money, how to get creative, and how to do cold outreach to all these people.

The tool in question is Oxylabs. It's basically scraping + AI, which equals very easy scraping for the non-technical person like me. They do have some parts of their website that can be more complex if you're more technically minded, but I'm going to go to `oxylabs.io/TKOPOD` because only there can I just type a simple prompt and have it scrape stuff for me. There may be other tools that do what I'm about to show you, but I've never used them. I have no idea if they exist or if they're any good. Whereas this is one I use regularly.

Step-by-Step: How to Scrape Craigslist with Oxylabs AI Studio

Once I'm here, I'm going to click on AI Scraper. Now, over here in this other tab, I'm going to pull up Craigslist, specifically the Dallas Craigslist because that's where I live. Before I talk about my ideas, I just want to start ideating. I just want to start browsing Craigslist as a user and see what comes to mind, and I'll walk you through how my weird brain looks at this. I'm going to start by going to 'for sale,' then 'household.' Okay, you've got someone selling a blender listed 10 minutes ago, 21 minutes ago, 38 minutes ago. You can see that this is very active. This is one of hundreds of different sections of Craigslist in just one of hundreds, if not thousands, of Craigslist markets, and people are posting every 5 to 15 minutes.

Now, this is mostly just crap, people are selling stuff. You do see some people offering services like countertops, flooring, etc. I also see a listing for a "Richard Petty Weetabix Box." So that makes me think, "Huh, I should build some sort of an agent that can cross-reference stuff on Craigslist with eBay." Because maybe there's some arbitrage to be found. If you're not familiar with arbitrage, it basically means finding something at a specific price over here and selling it for a higher price over there.

Okay, so now what I'm going to do is I'm going to type in one of the most legendary keywords ever for any marketplace, and that is "need gone." Those are the desperate, motivated sellers. Parakeets, bed frame, cool. A lot of this stuff is probably free, or they'll pay you to take it, like in the case of a piano. Now I'm going to search "wanted." Remember, if someone is posting to a marketplace with the word "wanted," they're saying, "Take my money! I want to spend money on something." So then what you need to do is go find the seller for that thing. You've got a buyer, which is usually the hard part, you need to go find a seller. Why not do that at scale with an automation? You can't build an automation until you scrape the supply side of the marketplace, which in this case is Craigslist.

Another thing you can do—oh man, this is giving me all kinds of ideas—is find both the buyers and the sellers on Craigslist. This guy right here: "Small and large motorcycles wanted in any condition." So, this guy's wanting to pay money for small and large motorcycles. Why? He probably flips them. Cool. Will he pay you to source them for him? Most likely, especially if he can still make a profit on his flip. So, what's stopping you from scraping and doing cold outreach to every marketplace you can find to find motorcycles and to send the leads his way? You don't want to be in the motorcycle buying business, but you could certainly be in the, "Hey, every time I send you a lead, it's going to be five bucks" business, right? "I'll swipe your card every time I send you a lead." That guy's wanting to buy old dimes and quarters. This guy's buying pinball machines! Freaking awesome. There are just so many little side hustles going on in here. But I'm getting distracted, so let's lock in. Let me show you how to scrape this stuff.

Okay, so to scrape Craigslist, I want to start with an active category where people are actively transacting. I don't want to go on the misconnections or the rideshare or the discussion forums, obviously. We're going to the 'for sale' category, not the subcategories like 'household,' 'farm equipment,' etc., just the entire 'for sale' category. We're going to copy the URL. Then we're going to go back to this AI Scraper and we're going to paste the URL right there. We're going to choose a location for the scraper server, basically, United States, because that's where I am. I do know that I want to check 'render JavaScript' always. And then output format, I want to do CSV—that's a spreadsheet. I can work with spreadsheets. Now, for schema, you can either provide your own schema or generate schema. What a schema is, I think, is like the output format of the data. What do I want to scrape exactly? Prices, listings, the name of the person listing it, the email of the person listing it—there's a zillion things we could scrape. So, this is where the technical knowledge would come in handy if Oxylabs AI Scraper wasn't as good as it is. But we don't need to worry about that, so we're just going to use plain old English and some commas.

So, let's do price, URL, location, title, date, and generate schema. And that's date listed, is what I'm referring to. Okay, there's the parsing schema, whatever that means. Now, let's see if this thing works. Scrape and parse. Click. I'm going to time it. It's 3:30 exactly. Okay, right here, what time is it? 3:31. This took about a minute. There's my CSV data. But is it any good? I don't know yet. I'm going to download it and I'm going to paste it into a Google Sheet so you can see it clearly right here in my browser. Boom, there it is. That's exactly what it scraped. It took a minute. We've got 168 different items. It has the location as requested, the URL, done it, and it works, awesome! The item title, the price, the location, items date. There's nothing there, and I bet that's because that's not what that data point is called. So, let's use AI, our little friend. We're going to go back to the original 'for sale' URL. Right-click it, view page source. I'm going to highlight all of it, Command C for copy. Then I'm going to go to ChatGPT.

Okay, so I want to ask ChatGPT, "What all can I scrape on this Craigslist page, and what are those items called for the parsing schema?" And then I'm just going to put the URL. We'll see if it can work with that. "All right, Chris, let's run with a comma-separated list of all the things you can scrape or derive from a Craigslist 'for sale' listing page and the individual listing pages, such as: title, URL, price, year, make, model, trim, mileage, condition, fuel, transmission, drivetrain, cylinders, engine_displacement, color, interior_color, VIN, category, post_date, image_URLs, tags, has_pics, map, and is_sold_or_removed." Okay, we're going to copy that. We're going to go back to the AI Scraper. We're going to click 'generate schema.' Paste in the new list. We're going to take the same URL that we just gave ChatGPT. Paste it there. We're going to choose 'render JavaScript,' 'CSV' for output, and 'generate schema.' Okay, there's the schema. Scrape and parse! And it is 3:39. Let's see how long this bad boy takes. Okay, 3:40. Took about a minute. We'll download this. We will put it in the spreadsheet. All right, this is beautiful. You are going to love this. So, we'll remove all the old stuff, paste in the new stuff, and look at all this. Okay, nothing here because there's no map. True if there's a picture. All of these have pictures. Location, which Craigslist refers to as 'hood.' Cool. You even have the timestamp of when it was posted right there. The price, tags if any, the title of the item, and the URL of the item, which it seems to work just fine.

Profitable Ideas: Leveraging Craigslist Data for Retail Arbitrage

Now, just for fun, let's go back to ChatGPT and say, "Hey, at first glance, which of these items are underpriced? Where can I make money?" Listen, I have a little soapbox for a second. All of us are using really powerful, interesting tools on a daily basis. And then at the same time, we have these awesome business ideas—or any idea, it doesn't have to be a business idea. And then we start thinking like, "Uh, how do I do that? How do I execute this? How do I start this business?" When we're already using the tools needed to start that business, right? Like, "Huh, I really want to start a business where I buy an excavator and rent it out." But how am I going to find customers? "I got to learn Facebook Ads, huh, I just don't, I don't know how to..." Dude, you're already using Facebook Marketplace to replace the couch that your kids ruined in the game room. That's a very powerful tool. Why not use that tool to find customers for your new excavator rental business, right? Like, we need to connect the dots. And so I'm over here thinking like, "How can I make money with scraped Craigslist data?" while using ChatGPT to help me scrape the Craigslist data, not even thinking, "Oh, I should use ChatGPT to give me ideas of how to make money with scraped Craigslist data." So, I'm just going to copy the exact columns of the price and the item title, paste it into ChatGPT, and say, "Where's the alpha? What's interesting about this to you?" Even if I don't like its specific answers, it's going to tell me things that help me generate more ideas on my own.

Okay. "Very likely underpriced: 600-800 mAh 6-cell battery - $60 (needs $100 to $150, used if large and healthy)." "Two Sony FDCE-16-65mm E3 lenses - $1650 (usually clears $1900 to $2100)." "Bundle should be closer to $2500 to $3000." "Marin OMTS-AC - $1200 (even with issues, these go for $1500+ deep off the day)." "Highest ROI on the sheet." "Clean title, minor scratches or minor issues but still worth considerable follow-up." "Traeger Pro 780 - $650 (usually goes for $800+)." "Traeger Pellets - $80 (often bundled with grill)." "Bowflex Home Gym - $450 (usually sells for $600 to $800)." "Heavy load capacity utility trailer - $2500 (good if you have space)." "Solid wood acoustics usually $500." So right off the bat, I'm thinking, "Oh, I could just flip stuff." I could just flip stuff. And that's another framework: if Craigslist is dying—it is dying, like I will admit, Craigslist is dying. Millions of people still use it, but it has been shrinking year after year. If that is true, then how can we use that to our advantage? Well, that means that there's an arbitrage opportunity. Most people listing on Craigslist are probably only listing to Craigslist because they're not on Facebook much, they don't have the Facebook app on their phone. Like, I don't want to say they're not on Facebook because everyone's on Facebook, but they just don't think to use Facebook Marketplace. Just like I don't think to connect certain tools to certain ideas, we're human, right? We're flawed. And so if a large percentage of these people are only posting their item to Craigslist, but there aren't enough people looking for or at their item on Craigslist, then that means it's probably underpriced. Because what does Economics 101 tell us? High supply + low demand = low price. So, that could be an entire business: just buying stuff from Craigslist and selling that same stuff without adding value to it or improving it on Facebook Marketplace. That's like the lowest hanging fruit, without any sexy apps or automations or having to do cold outreach, just piggybacking on these giant tech companies to find all the customers for us.

When someone's selling a Dallas Cowboys Tom Landry "Dree" rejection letter, first of all, what is stopping me or you or anyone from just taking all this stuff that we scraped from Craigslist and listing it as if we already own it to Facebook Marketplace? Not in an unethical way, not because we're trying to scam people or mislead people, but to gauge demand. "Oh, this Sony lens is $1,000 on Craigslist. I think I can get it for $800, and I think it's worth $1,200." Boom. Let's say we get it for $900. We meet in the middle. Go to Facebook Marketplace, you list it for $1,400, you settle on $1,200. You just made $300 profit simply by buying from one platform and selling on another. This is 1000% realistic and doable, and people do it all day, every day. It's called retail arbitrage. People will go to Walmart, they'll see something on the clearance rack 40% off. They'll scan it with an app. Then they'll look on Amazon and see, "Oh, wow, this thing is $113 at Walmart, but these sell for $225 on Amazon." I'm just going to list it real quick from my phone, and then I'll buy it. Or, I'm going to come back and buy it only once it sells, so I'm out of pocket $0. I'm not inventing retail arbitrage. I'm just showing you a new way to do it. Free couch, LG washer/dryer. Let's take this a step further. Let's go up the food chain, a little higher ticket, right? Because we don't necessarily want to flip $30 Pokémon cards. Why not buy 'lunker cars' for $2,000 and sell it for $3,000 without having to fix it up?

Identifying Desperate Sellers and Motivated Buyers

And hey, that actually reminds me of another business I used to have. I built a scraper that would go scrape Zillow (which, by the way, Oxylabs can scrape Zillow as well). It scraped Zillow every single day for the FSBO listings (for sale by owner), and I would auto-text them and offer on their house that was 40% less than the Zestimate. 97-98% of people either didn't respond or said, "Go kick rocks, leave me alone." And 1-3% of people would say, "Let's talk." Yeah, I'll take that. It's a numbers game, guys. And the higher the ticket of the item, the fewer numbers have to work in your favor for you to make good money. So let's do this with cars. Back to the Dallas Craigslist, 'for sale,' 'cars+trucks.' Now, let's do some filtering because I do not want to mess with dealerships. I'm going to assume that those are unfairly priced to their benefit, and I won't find any good deals there. I'm going to go to 'posted today' only, because if it was a good price posted four days ago, it's probably gone. 'Sold by owner.' And look at how many of them! Like 10 to 20 every single hour just in Dallas. And then I want to parse this other data like how many miles it has. So, we're going to take this URL and go back to ChatGPT and say, "I want to parse cars for sale data from this page. Give me a comma-separated list of all the things I can scrape, such as mileage." I also included: 'title, URL, price, year, make, model, trim, mileage, condition, fuel, transmission, drivetrain, cylinders, engine_displacement, color, interior_color, VIN, category, post_date, image_URLs, tags, has_pics, map, is_sold_or_removed.'

All right, here's my prompt. "Okay, I want to parse cars for sale data from this page." And I entered the URL. "Give me a comma-separated list of all the things I can scrape, such as mileage." All right, sure thing. Okay, we're going to copy that. We're going to go back to the AI Scraper. We're going to click 'generate schema.' Paste in the new list. We're going to take the same URL that we just gave ChatGPT. Paste it there. We're going to keep 'render JavaScript' and 'CSV' output. Generate schema. Okay, there's the schema. Scrape and parse! And it is 3:40. Took about a minute. I just want to harp on that point even more, guys: more eyeballs equal more money. Think of the dumbest business idea in the world, okay? Like, pet rock dumb, but even dumber. I'm looking out the window, let's say. I see my fence in my front yard. Let's say I want to sell fencing panels that have Pokémon characters on them, right? Giant Pokémon characters on fencing panels. Pretty objectively a dumb idea, right? What if Elon Musk tweeted about that idea with a link to the website? Do you think people would buy it? Would anyone buy it? Of course! Not because it's a good idea, but because of the Law of Large Numbers, which means that if enough people look at it, someone's going to buy it, right? Someone, because the marketing is there, the eyeballs are there.

Beyond Retail Arbitrage: Niche Job Boards and Real Estate

Now, similar principle, different example. Let's say I have a single-family home that I rent out, and the market rental price for that home is $1,000. It's three bedrooms, two baths. My neighbors are $1,000 a month. People across town are $1,000 a month. That's just what they go for. So how are you going to get more money for it? You're going to get more eyeballs on it, right? If you have 100 people walk through that house, do you think you could get $1,300 a month for it? Yeah, you probably could. Because a few of those people are going to really, really love something about it—the layout, the location, something—and they're going to value that feature more than they value an extra $300 a month, right? That's why arbitrage opportunities exist, because you buy from a place with fewer eyeballs and you sell it on a place with more eyeballs. Therefore, the price goes up. Okay, beautiful, my data is scraped. We are good to go. We're going to go back to our trusty old Google Sheet. We're going to clear it. We're going to paste it. What do we got? Okay, look at this: body type. Most people don't list out all these features, right? You can see sedan, pickup, four-door. Cylinders, condition, drivetrain, engine displacement, exterior color, fuel, make (Volvo, Toyota, etc.), mileage (almost every single car has mileage), and model. Price, title, etc., and year, and link. Okay, cool. Now, this is kind of a lot of data to work with. So I'm going to re-download this as a spreadsheet. We're going to go back to trusty old ChatGPT. We're going to upload it here. My prompt: "Where are the best car deals in this sheet and why? Be specific, and look at the titles to see if it's running or not." Because if they see a 2010 Toyota Camry listed for $2,000, and the title says it's not running, it might not be a good deal, right? Okay. "Best deals (running, flippable, mispriced): 2016 Jeep Wrangler Unlimited - $3,500. Title does NOT say parts only, not running, or mechanic special. 4x4 vehicles are currently hot. Even with issues, these go for $7-10K deep off the lot. Highest ROI on the sheet. Clean title, minor scratches or minor issues but still worth considerable follow-up. 2001 Corvette - $11,000. Title '2001 Corvette for sale' (manual, not broken language). Even rough title usually starts $15K to $18K. Running, emphasized (make sure). It runs and has clear title. $3K to $4K margin. 2008 Toyota 4Runner - $6,400. Title: 'This is a normal red flag.' 225,000 miles but that's expected for these. Good value vehicle, easy flip. 2009 Buick LaCrosse - $4,500 (95K miles). Low miles for age. Title: clean language (good). Very flippable commuter car. Likely $6,500 to $7,500 means $2K to $3K profit. 2016 Dodge Charger Pursuit - $8,500. Police package (details, which creates pricing inefficiency). Title does not have 'needs work.' Good if it runs and drives, this is underpriced. Likely $14K to $18K means $5K to $9K profit. 2007 Honda Civic EX Coupe - $5,999. Very reliable, good gas mileage. Title: 'runs great' (great language). These are 'money' cars. Very fast flip. 2010 Honda Odyssey EX - $5,299. CO title. Reliable car. These cars sell themselves. 1999 Dodge Grand Caravan - $1,500. Title: clean. Cheap people mover. There are always in demand. Likely $3,000 means $1,500 profit. 2009 Nissan Pathfinder - $2,499. This availability was 'fresh.' That simple word matters a lot. If it runs and drives, this is underpriced. Likely $4,500 means $2K profit. 2010 Honda Odyssey EX - $5,299. CO title. Reliable car. These cars sell themselves. Also, that reminds me, I've had another business in the car business where I scraped car data. It was called Presto Cars. We would go to auctions—Manheim and Adesa auctions—and we would buy cars for people, me and my business partner Nick. Before the auctions, the night before, we would scrape all the auction data and find alpha: "Which of these cars are probably mispriced based on previous auction sale results?"

So, I hope this stuff gets you as excited as it gets me. A lot of times in business, you see data like this, you get super excited, and then you go start launching something, and you find out, "Oh, this isn't as good as I thought." That's okay! Push through that. That's called the trough of despair. That's normal. You got to push through that. Don't quit there. Just keep going. Now, what other things can you do with scraped Craigslist data? How about you make a niche job board? There are a lot of people out there making a lot of passive income—and you know I don't say that very often—with job boards. You scrape jobs from Craigslist, you populate your job board, and then you drive traffic to your job board. How do you make money? You charge businesses to list their jobs. I hope by now you've realized that Oxylabs is awesome. It's an awesome tool. They're a friend of the channel. They have 175 million proxies. If you don't know what a proxy is, it basically helps you scrape better because it masks your location, so these websites don't think you're scraping everything from just one location. If you want to use Oxylabs, please use my link below at `oxylabs.io/TKOPOD`. Also, I neglected to mention, their AI Studio tool is completely free for you to use. If you want more advanced scraping, you can go buy scraping credits that are super affordable. I asked my team to go ahead and implement a coupon code for you guys if you want to use Oxylabs. Just do TKOPOD for 20% off. This works with both their AI Studio and all of their other products.

Conclusion: Embrace the Power of Scraping

Thanks for checking out the Koerner Office, and we will see you next time.