I have a faint memory from childhood. I was about 6 years old, and it must have been the early afternoon because my parents had taken off work early to scoop me up and take me to a nondescript office in a semi-depressing mini-mall near our home. It was all drab carpeting, cheap-looking desks, and fluorescent lighting. There were posters on the walls of islands and families holding suitcases; a nice woman typed some things into a computer and eventually handed my mom some thick, envelope-sized pieces of paper sprinkled with codes. We were going to Arizona, to visit family, and had stopped at a local travel agency to confirm the trip and pick up tickets; it was the first and last time I ever visited such a place.

Remember travel agencies? Now, I couldn’t even tell you where one is. Other services that time has tried to leave behind — post offices, libraries, copy-and-printing shops, Blockbuster Video — I can still muster a vague affection for and even direct you toward. (If you drive about 40 minutes southeast of Portland along the 26, you will eventually hit a Blockbuster in Sandy, Oregon. Godspeed.) But travel agencies were swiftly and deftly swept away by the common enemy of The Way We Used To Do Things: the internet.

Eventually, online travel agents — or OTAs — took over the jobs of their human predecessors. You could stay inside your home and personally harness — via a user-friendly website — the mysterious information that travel agents once plunked away at their keyboards to find. Suddenly, the unknown world of finding and booking flights was made clear to the consumer.

“We’ve really given the individual the power that the operator once had in essentially conducting the most exhaustive and comprehensive search possible,” says Bob Mann, an airline industry analyst and consultant. “Individuals now have more characteristics and filters through which to look at the universe of possibilities. It used to be you had to know what you wanted and know who to contact to get it, and be given a take-it-or-leave-it price. Now, you have this mobile-enabled, multiple-filter, very comprehensive set of options.” Mann says that flight searching and booking has “gone the full evolution.” Now, all that remains is perfecting the usability factor — and, of course, finding better deals.

That’s where a new class of services enter. They come in various forms: email newsletters, apps, subscription sites. Scott Keyes, the creator of the Scott’s Cheap Flights newsletter, sees himself (and the niche flight-finder market) as a consumer advocate service — like a sort of personal airline concierge. “It’s gotten to a point where it’s a paradox of choice,” he says. “There’s too much information available to everyone, so nobody knows how to make the best decision about this.”

Mann mentions one service called Routehappy, which looks at characteristics of a flight outside of price and departure time — things like in-flight meals, or whether a long flight will have Wi-Fi — and creates a rating system. Routehappy is more concerned with gathering air travel data about a potential experience rather than finding cheap flights for spur-of-the-moment international vacations, but what it’s doing is still significant to improving flying. And it’s different from what the tradition OTAs are doing when they show you a list of flight options. “There’s an increasing desire to use what data carriers have about their own operations and try to extend those as far as possible into the booking and tracking experience,” says Mann. “I think whatever price point and experience you’re looking to purchase, whether it’s business class or coach, you’re still looking for the best deal.”

For years, the best value was whatever you found via an online travel agent — Priceline, Expedia, Kayak, your OTA of choice. But that’s no longer true, and over time, a side effect of turning to these services over and over has surfaced: They had trained us to become used to a certain pricing scheme, even if that didn’t reflect market value.

“Once in a while you would run across a pretty good deal on the major sites — Kayak, Priceline, and so on — and most of the time, they were good enough,” says Next Vacay CEO and cofounder Naveen Dittakavi. “Before 2008, anything under $1,000 to Europe was a good deal. The airlines had bought fuel at a prior high price point, and they were passing those costs along to us.”

“We became conditioned to expect this, so even as the price of fuel has dramatically decreased, airlines have been able to keep prices high. “Flight prices haven’t necessarily reduced correspondingly,” Dittakavi says. “It’s not like you can peg the jet fuel index against the cost of flights and find a correlation. You can’t, because the airlines essentially made us believe that we would be used to high prices for certain kinds of flights.” Other factors have also made international travel cheaper: There has been an expansion in transcontinental budget airlines, and newer, more efficient aircrafts. But mentally, we’ve been conditioned to expect that travel is expensive.

“If the market will pay a certain price for a flight,” says Dittakavi, “then that’s what the market will pay.”

So most of us do. Scouring foreign databases for flights was once something we couldn’t do, and then suddenly we could — and consumers took the prices yielded to them from OTAs as the best there were. Some people are dedicated enough to filter down to specific hours, dates, and cities, but that’s a manual, time-consuming process, so most don’t. It’s a similar story for Google: All of a sudden, one day, we could search for anything. This has never meant the top results are the best or most accurate, and those who’ve taken the time to learn Boolean search or to filter Google results will get better results — but for many, clicking something on the first couple of pages is good enough.

It wasn’t for Dittakavi. He and his wife started Next Vacay after catching the flight-hunting bug. They’d seen a friend’s Facebook post about a $300 round trip from D.C. to India and they were sold. Friends noticed the couple’s frequent travels and asked how they were managing this. From that, Next Vacay became a subscription service that costs $25 a year and is personalized to members’ origin airports — and it’s one of the few of its kind that is questioning the old model of finding flights. What’s more, it might be challenging us to rethink how we travel.

The motivation to find cheap flights is not surprising. International travel is more popular than ever; more and more people have flexible jobs, or work remotely, or benefit from unlimited PTO. The world is becoming more accessible for the people lucky enough to be in this position, and they want to be able to explore it.

This is happening at the same time that disillusion with air travel is reaching an all-time high. United’s recent physical removal of a passenger is a particularly dark moment on an already bleak timeline, and the negative headlines have continued — to say nothing of the general inconvenience travelers feel during every flying experience.

And so a perfect storm of wanderlust, economics, and low morale have created opportunity for airline search savants like Keyes and the Dittakavis. Something cyclical is happening: The internet replaced travel agents with OTAs, and now internet-enabled, self-made travel agents are replacing OTAs. Well, not quite — in reality, OTAs aren’t who suffer most from these niche travel services, because they just aren’t as interested in selling flights. “OTAs … have access to the real raw data, and I don’t see why they couldn’t start to do [what we’re doing],” says Dittakavi. “But here’s why they don’t do it: They’re mostly interested in getting the hotel.”

Next Vacay doesn’t get a commission for getting a subscriber to book a flight. Readers are prompted to click through to Google Flights or Momondo or wherever the deal was found and book everything themselves. This sort of operation doesn’t make a dime on compelling consumers to buy something; it makes all its money off of subscriptions and premium members. But this is how OTAs make money — except that they aren’t so worried about the flights.

“The real money in this business is in hotels,” says Dittakavi. “Hotels pay a significant commission, a CPA, what’s called a ‘cost per acquisition,’ when you book through these sites. So it’s in the incentive of TripAdvisor, Expedia, and so on to sell you a hotel deal. For them, the flight is not the primary product; the hotel is the product.”

What OTAs don’t care about is something called a mistake fare — but airlines certainly do. Mistake fares happen when airlines load inventory — essentially, open seats — into a system that will be distributed to databases that travel agencies access. Either human or computer error can result in a mistake fare, pricing a seat incorrectly. Once those numbers go out to travel agencies where consumers can see and buy them, the airline has about 24 hours to correct the mistake or else it has to honor the price. “When we list mistake fares, we mark them, and tell people to hold off making additional travel plans until the next day,” says Dittakavi.

It was a mistake fare that started it all for Next Vacay. For $300, you could go nearly anywhere in the world on the specific airline. “That probably cost someone their job,” Dittakavi says. Keyes knows this, too, saying that if an intentionally cheaply priced flight is suddenly bought up thanks to the alerts sent out by operations like his, it’s detrimental to their business. At the same time, all that unused inventory is gobbled by consumers — certainly not at a favorable price for airlines, but at least it’s not going unused, he reasons. But on mistake fares? “They hate us,” he says.

Each morning, I wait in anticipation for my Gmail inbox’s Promotions tab to light up with alerts from my various flight-finding subscriptions. There’s Next Vacay, here’s Scott’s Cheap Flights, there’s GTFO. Today, Scott’s Cheap Flights is advertising nonstop flights to Oslo for between $200 and $300. Over on Next Vacay, there are flights to Hawaii for $375. GTFO has a flight from Seattle to Milan for $475. And on Kayak, my general flight search engine of choice, those flights during similar dates were about $200 more.

These alerts are personalized and written in a human voice. “I’m not a sappy guy, but being in southern Mexico (Oaxaca) for Day of the Dead was one of the more fascinating experiences I’ve had traveling,” a recent email from Scott’s Cheap Flight reads. “The parades, the fireworks and abundant food, the all-night cemetery visits. Can’t recommend it enough.” The email, of course, goes on to advertise flights between $213 and $298 to Mexico City and Guadalajara for the holiday. When I get a Next Vacay email for a flight originating out of Seattle, the note opens with, “We know you don’t live in Seattle, BUT these deals are so good that it may be worth a drive or short flight to take advantage of these savings we found here.” These sites use personality and storytelling instead of relying on nothing but numbers. (Even when those numbers are often so good that they could do just that.)

That personal bent is purposeful. Because these deals are so much better than what you can find on regular OTAs, it can feel like these services have cracked some sort of code and rely entirely on proprietary software — but for the most part, that isn’t how this works. It’s just a handful of dedicated, focused people, doing something that’s essentially very simple, albeit time consuming: scouring the internet. Next Vacay is a team of seven; there’s one developer, and the rest populate the “hunt team,” marketing and growth, and customer service.

“We’re monitoring about 40 different spots on the internet where people might be talking about low fares or mistake fares,” says Dittakavi. This includes Twitter, Reddit, and other forums. All of that data is pulled into a feed, and then Next Vacay’s proprietary software analyzes it to look at the airports and prices mentioned and then, according to Dittakavi, “does some basic filtering and then elevates a list of origin and destination points.” Also factored into this are cheap flights resulting from “fare wars,” which is when airlines look at cities their competitors “own” and try to undercut their prices. Next Vacay and other deal-finding operations can look at this and figure out where a cheap originating or stop-over city could be. For instance, if a user usually flies out of LAX but there’s a fare war going on for Burbank, the human element of these businesses will use that to tell buyers — yet another thing software alone can’t pick up on.

The Next Vacay team looks at all the data and decides what’s a good deal — from a human perspective — before sending the options along to members. “There are a lot of good deals out there, but there are a lot of deals that are bad good deals.” For instance, if there’s a good deal on a destination that’s undergoing political conflict, Next Vacay won’t recommend it.

Scott’s Cheap Flights works in a similar fashion. The company launched in August 2015 and just recently brought on its 13th employee (the workforce is mostly made up of “searchers,” with a few people working on customer support). For Scott’s Cheap Flights there is no algorithm involved beyond the search engines the team uses to find deals. “It’s all just boutique, search by hand,” says Keyes. “My job is to be on Google Flights eight, 10 hours a day just trying to scrounge up these cheap deals. No algorithm involved.”

Keyes says that while there are some algorithm-based deal sites that do a good job, there’s a certain amount of human discernment needed to decide what’s a good fare and what isn’t. For example, he doesn’t send out deals on Wow Air. “They put out these really eye-popping deals, like round trip to Iceland for $200 or something, but then you start to look into the fees, and the fees are a solid $300 at least. Not only is that not that a good price, but it also leads to a negative experience for the customer.” He and his team also scour for convenient puddle-jumper flights. A flight from New York to Santorini, he offers, can usually run you around $1,500 on Kayak, Expedia, or its competitors. But there are several round-trip flights from New York to Athens in the $450-$500 range, often nonstop. “Once you’re in Athens, Athens to Santorini can be literally $300 round trip, and so that’s going to save you something like 530 bucks. You just saved yourself some thousand dollars that a computer algorithm couldn’t pick up on.”

Next Vacay and Scott’s Cheap Flights have similar origin stories. The founders all happened upon incredibly inexpensive flights and were hooked. They started looping interested friends in on deals they would find — the Dittakavis used a WhatsApp group, and Keyes started a newsletter using MailChimp — and eventually both grew too big to manage. That’s when Next Vacay launched as a subscription email ($25 a year, and customized emails to your origin airport) and Scott’s Cheap Flights launched as a free newsletter with a premium pay option (which means no ads and more deals for $39 a year). For the Dittakavis and Keyes, this has become a full-time job. Keyes says his newsletter’s click-through and open rates are incredibly high but that, in all honesty, he doesn’t care. “I don’t care if people book a flight or not; it literally doesn’t matter to me.”

Scott’s Cheap Flights doesn’t use affiliate links or receive commission on bookings — all profit is made from its premium memberships. The service has 600,000 subscribers, and 10 to 12 percent are premium members. He says there have been inquiries into acquiring his business or investing in it, but so far he’s not interested. Next Vacay similarly profits off its membership fee, and says its users are in the tens of thousands.

Human-only (or -mostly) cheap flight finders are not the only option. There’s SkipLagged, a service that helps find cheap flights via “hidden city” tickets, which is when a connecting flight can get you somewhere cheaper than a nonstop (for instance, if a flight to Denver goes on to New York, it might be cheaper to book the flight to New York and just not continue on than to book a nonstop to Denver). It was sued in 2014 by United Airlines because the airline said this practice violated its carrier contract. (The suit was thrown out.) There’s SkyScanner, a metasearch site that also makes recommendations based on its algorithm. And there’s Hopper, a flight price watcher that predicts and alerts when you should buy.

While Hopper’s core feature is to watch trips you’re interested in and alert you when it’s time to purchase, head of product Maggie Moran says that its recommendation engine is also becoming popular. This tool will take the flights and dates you’ve requested and make suggestions. “If you want to go to Paris in June, it might say check out September instead,” she offers. According to its data, Hopper says people are becoming increasingly flexible with their travel plans. Hopper has made more than $1 million in sales off the recently added recommendation feature.

Hopper has an editorial and human curation team, but unlike its peers (and competitors) in cheap-flight finding, the company focuses on its algorithm and big data. “We’re confident the right way to find consumers deals is by watching in real time,” says Moran. “We’re confident we can get there in our algorithms. … If we only had humans looking for these deals, we would be losing [consumers] a lot of money.” The Hopper team also hacked together an app called GTFO (which stands for Get the Flight Out) in 2014. GTFO finds cheap, last-minute flights out of select cities. (It currently serves only Boston, Los Angeles, Chicago, New York, Seattle, and the Bay Area.) GTFO more specifically serves clients that aren’t married to a specific date or location, but purely want a deal on a trip. Hopper is the main focus now, but director of communications Brianna Schneider says GTFO is still beloved by its users, and that they might look at integrating some of its spontaneous travel features into Hopper.

Like Next Vacay and Scott’s Cheap Flights, GTFO pushes you to book flights outside of its own app; Hopper allows you to book directly. Unlike Next Vacay and Scott’s Cheap Flights, though, “We make money through airline commissions and we also charge users a $5 booking fee,” Schneider says. “The fee is always included already in all the prices we display and disclosed in the booking details. Our users have been fine with paying the fee since the app saves them an average of $50.” Hopper doesn’t use ads, instead relying on the airline commissions and booking fees for profit.

Despite their differences, all of these models either rely on or benefit from something of an attitude shift. Instead of picking a particular place for a trip, choosing specific dates, and then going to look for flight accommodations, these services ask that you keep an open mind and let the market guide you toward a destination. “We can’t predict when a location will go on sale,” says Dittakavi. “This isn’t going to work for people who have specific needs or need to travel at specific times.” This is what keeps traditional OTAs from worrying too much about this new class of super-powered online flight-finders: We’re always going to need to book that business flight in September due to that one conference, and our agenda isn’t flexible. These new services can’t do that.

What they can do, though, is offer a thrill at major savings on the travel we actually look forward to — the vacations, the explorations of new places we didn’t even know we wanted to go to. And the idea is to turn us into disciples of their services along the way. No one wants to brag about a $345 flight from New York to Dallas on Priceline — but a $230 flight from Seattle to Milan you found in your inbox? That’s worth talking about. And booking.