Scott's Cheap Flights: Growing a small side project into a booming business

Hate to piggyback but I noticed that whenever this service gets posted on Reddit there are dozens of accounts that reply addressing Scott by name and have little to say.

Upon checking their comment history, I saw that most users hadn't posted anything for months, or even ever. It was clear that there was some kind of voting manipulation going on, and as reddit goes, if something gains traction it's bound to get upvoted just because.

It's a bit like Candy Japan, who posts here once a month with very little to say, and nothing vaguely technical in his articles. These kinds of posts just seem like blatant advertising. What are the rules for self-promotion of services on HN? It just seems dirty when there is a half-hearted article advertising a service without being labeled as an ad.

Throwaway of course, because I'd rather not get attacked by some bots.

Out of curiosity, which airlines charge $60 for cabin bag?

The most passenger-squeezing-until-recently airline in Europe (Ryanair) now allows two cabin bags for free (one standard ~55cm bag and one small bag).

Apparently Wizzair charges 10-20€ for "large" (aka regular) cabin bag though, only small ~42 cm bag is allowed for free.

I have written to Kayak suggesting this in the past. Without such a feature, the hidden charges and nickle-and-diming will just get worse.

The best explanation I heard (from elsewhere) was that if Kayak automatically included any fees, then some users might see a lower price on another site and mistakenly conclude the Kayak didn't actually have the lowest prices. But this seems obviously solvable by making it a non-default option and displaying the price as "$X ($Y fare + $Z fees)".

Add-on fees have gotten out of control, both in the hotel and airline space. There is a race to the bottom for having the lowest price to show on booking sites. This is a form of SEO for airlines and hotels.

I am not familiar with any airlines that charge for carry-ons, but for example I know that many hotels in downtown Las Vegas have mandatory "resort fees" that on many nights actually exceed the nightly rate shown in booking engine search results - so the actual cost of staying is more than double the advertised rate. Most booking sites have heard so many complaints about this that they do show the total (including resort fees) just prior to booking. But the "headline rate" - the one shown in search results and used by the sites in arranging results by price - is the one sans resort fees.

Taken to an extreme level, hotels could advertise a $1 nightly rate and a $200/nt "resort fee". It wouldn't surprise me if this actually started happening, given the competition among hotels to show up first in search results. It sounds like some budget airlines are headed in a similar direction.

I 100% percent agree with you.

But on a practical note, some advice for those traveling on these budget airlines: if you pack your carry-on in a backpack rather than a suitcase, every budget airline in Europe and the US that I have flown on will not give you a hard time about it. I personally use a very large backpack that holds at least as much as a carry-on suitcase, yet the fact that it is in backpack form means they count it as my personal item and don't give me a hard time. When I travel I usually have a smaller laptop bag inside a compartment in my backpack, once I get on the airplane I take that out and keep it with me as a personal item (so I can easily access it inflight) and I put my backpack overhead like I would a suitcase. I have successfully done this on Frontier, Allegiant, Spirit, and Ryan Air within the last year and have not had any issues.

What is the general policy on the so-called 'Ryanair jacket'? Can you get in trouble by packing 20kg of stuff in your coat?

Hipmunk can do that.

Or credit card fees.

Pay by debit: 0

Pay by credit: 5 to 10%

Nice website. Fix this issue: 1) When I search with a date, it loses the depart and return dates every time. Save the dates and remember them. 2) After I select a depart date calender as Month September, it should open September when I select Return calendar.

Love the simplistic design of the website to give the information right away! Can you where are you sourcing the data from? I've playing with Flask and Scrapy yesterday and love to build something like it. What I have in mind, [Complete disclosure - I'm a total novice right now] Scrapy to get the data and Flask to integrate it on the web with a Boostrap front end.

I replied to one of his emails and we had a nice back and forth even laughing at a pic response I sent him. He seems like a good person and I signed up for his premium service. I was already very interested in his premium svc but that interaction really sealed the deal for me.

Tiny suggestion: I was looking at flights and it was not immediately clear to me that Santa Cruz referred to Bolivia and not California. It almost made me disregard it and not click. May want to put the country name on the homepage as well.

FYI: I am trying to update my user settings (for my airport preferences) but it won't let me without re-entering my CC information.

I'm an SCF premium, will you give me a premium 3 Month coupon in order to compare the services and maybe switch my subscription to your app?

Do you use a particular api to find these flights?

Suggestion for your site. I'd change the front page header text that says: "Discover cheap flights with Concorde."

It's not the text I disagree with it, it's how you're using it. You've made it so that you have to mouse over that text to make it react (segments change color) and figure out that each section may be clickable (how do I know where they go, if I'm an average user and don't necessarily know to look at the url hover at the bottom of the browser?).

Why is it an issue? It's extraordinarily bad usability, because there's nothing above the fold on your site that says: sign up. Not anywhere. There's a weird text in the upper right that says "Discover" that is changing colors, but gives no indication that it links to /sign-up. From a logical usability stand point, why would the "Discover" text be the /sign-up link, as opposed to just having a "Sign up | Login" segment.

You're trying too hard to be clever / creative, you're over-doing it and it's a bad usability outcome.

At the top, I'd keep the large text that is making a clear statement. Then place two or so lines of descriptive text below it, with blatant links to signing up etc. Or some variation on this. You have plenty of room to work with up in the header area.

confused. none of the flights are supersonic.

Gave it a shot. No results from SIN, my other two "primary" airports (BKK, CGK) don't even match in your interface! Even tried NRT and nothing. I assure you these airports exist and I'm not making them up.

Tried my old haunts of SYD, MEL and ADL - they match but no results. So that's 6 airports I tried with literally zero results, and half the time the (major international) airport wasn't even found. I know you're probably not focussing on my area but still not the most amazing first impression! At least a "coming soon" message would be an improvement.

A few suggestions:

- since you are using Stripe, allow me to log in and pay without submitting my CC info over again.

- when selecting home airports, how about prefilling it with the list of closest to you / busiest based on geolocation? The picker interface could be improved by allowing autocompletion as well.

All of their listings are roundtrip.

I have personally booked sub or around-$400 return trips to Europe from the US using this service.

Wow, thanks for the feedback! Almost every single deal we send out is roundtrip. This is actually why we put "normal roundtrip price" instead of just "normal price" to help clarify this.

Maybe instead we should do "Roundtrip NYC to Paris: $260" and then put "Normal Price: $900" to help clear this up.

Thoughts?

Huh? I subscribe to lots of travel sites and it's pretty accepted that a price is round-trip unless stated otherwise.

Another data point: I haven't seen any email indicating that Scott's selling the addresses.

Hey, I'm just going to post the same response I used in the IndieHackers comment section for someone with a similar concern. Here's what I said:

"Our subscribers are too important to us to ever sell their information to someone else. We never have and never will sell users information like that. Our only source of income is subscriptions and occasional advertisements in the free emails. Ruining our reputation to make a few extra bucks would be an extremely stupid decision on our end. With that said, we do run promotions with partners where we give away awesome vacation packages (Mexico and Tokyo recently). If anyone signs up for one of these promotions they are agreeing to the terms and conditions, which clearly state that by signing up for the giveaway the email address will be distributed to all of the partners involved in the giveaway (typically 4-5 other companies)."

And their response:

"Thanks for the reply! I looked through my emails and found that the emails started right after I entered a giveaway from you. So that explains it! You're right, it does say for the giveaway offer that you're signing up for emails. Sorry that I jumped to conclusions. It's just that I get a lot of spam and it's very annoying. Also, your partner emails provided no value to me at all."

--

With that said, it looks like we could do a better job making it even more explicit (which we try to do in the emails announcing the giveaway). We'll work on this on our end and never intended to mislead anyone.

This isn't really "selling" email addresses. It seems much more like it's a lead gen strategy that multiple companies are participating.

Providing an incentive to sign up isn't quite the same as sending all those subscribers to the giveaway that didn't opt-in to it?

Posting this from below because you're making sound as if they're selling emails, when the terms of the giveaway are stated quite clearly.

>Here's a version that's hosted on ProductHunt - https://giveaway.producthunt.com/landing?promo_id=83848bb5-f... > By entering this campaign I agree and consent to recieve emails, communications and promotions from Product Hunt Inc., theSkimm (a daily email newsletter to stay in the know), Scott's Cheap Flights, Journy, The Wirecutter and Conde Nast Traveler.

>You can hate the promotion, but you're certainly not required to join it. And, it's certainly not the same thing as Scott selling addresses of people already on his mailing list.

That's a great question. At a certain point... yes. But I think that number is much higher than a few hundred thousand users. If you think about departure cities X destination cities X available dates and times X number of seats, the number is huge. I'm not saying it won't happen, but I think we have a long ways to go before we reach that level.

While flight industry is booming, almost nothing matters - just keep telling people they can fly cheap. Dozens of aggregators are already doing this.

Even if it caps out right where it is... they're retired after not too long

On the other hand, does it matter how long the business lasts? It seemingly has minimal costs and is making bank.

Some deals require specific starting and end destinations to work properly with a specific carrier. Others can require a stopover in a certain city before continuing on.

The really good deals will even be site specific. All this can be done via good expertise. With scraping and this type of logic, you'd get blocked quickly from the major sites.

Kayak does have alert notifications.

The thing is that if you're looking for a very specific destination, you'll almost never get the best deals. Flight deals work best for people who are very flexible about destination and timing.

I use Kayak and Google as starting points for every flight search, and I still spend hours and hours finding good fares on good routes.

This is just anecdotal, but if neither Kayak nor Google gets it right enough for me to use them end-to-end, I suspect there's a lot of room for a high-touch, human-intervention kind of searching.

Tech provides leverage horizontally, not vertically.

As programmers, we often fool ourselves into believing the opposite.

Thanks Joel!

Yes, that is how it works at most companies AFAIK.

Yes. This happens in pretty much every company, even if the company has a corporate card. Employees rather buy the tickets themselves and expense the costs.

Yes, it's completely 100% standard practice as other replies already mentioned. What's funny is that the tone of your question suggests he's doing something fraudulent. If you were the company auditor or a government prosecutor, and you didn't know about the standard practice or simply wanted to be a jerk, you could absolutely make a case for fraud. If you were fired or prosecuted for keeping the miles, what would be your counterargument? Everyone else was doing it? That doesn't work for speeding tickets. This is an example of how arbitrary laws can be.

Frequent flyer miles (and hotel points) are a massive perk for traveling employees. Most consultants I know finance annual vacations completely on points. Every once in a while, a big company will try to pocket the points for themselves, usually to great outcry--because it's akin to cutting benefits.

The corporate thinking is: "Points are a reward for buying travel. And I'm the one footing the bill." The employee thinking is: "Points are a reward for traveling. And I'm the one on the damn plane."

Personal perks on a corporate card are similar, and again, do the math of how many points you get by channeling $60-100k/yr worth of corporate travel expenses through a credit card.

Wow these deals look too good to be true! But I guess I'm in the US lol

Further inspection, looks like terrible prices, the advertised rates are based upon such things as 7 people sharing.

Or fly4free or secretflying or ..

99% of the time it's just affiliate links slapped on some deals found on FlyerTalk

Speculation: it's possible that the source sites are actively hostile to scraping and intentionally mess with the layout regularly, or may even have ruled out any kind of automation in their ToS - so if you don't want to get sued or blacklisted, you can do no kind of (detectable) scraping.

Who says it's not automated?

https://workplace.stackexchange.com/questions/93696/is-it-un...

;)

Seriously though, automation may not be a good value proposition, even if the companies with the data that SCF is scraping manually want to cooperate. It seems the human scrapers are a fixed cost. It's not getting harder to do manual scraping as new customers come on board. The business is already paying this fixed cost. The business is currently doing well. All the business really needs to do is continue growing and the fixed cost will continue to shrink as a proportion of the business.

Why would the business care in that situation? Reduce from 12 mid-range salaries to 3-4 high salaries? They would do well just to break even on salary expense, so why bother? If it ain't broke, don't fix it.

This is in the article, search for Business Insider.

We have plenty of copycats out there (one of the downsides of us trying to share our learnings online haha). I'd just look at when the domains were purchased if you want to know who was first ;) Also, we send deals from almost every continent (launching Africa by the end of the year).

Can you recommend other similar podcasts?

Just saw thus addressed in the comments:

> Great question. We get this one a lot and have been approached by quite a few developers who say they could easily build us something like this.

Is it possible? Sure.

Would it be better? I doubt it.

Say we use Google's flight API. The cost per query is 3.5 cents. Departure airports X destination airports X available dates X 3.5 cents = a ridiculous amount of money.

Or we build a scraper. It would be cheaper than the API, but we'd still need the human element to make sure it's actually a good deal. We take into account number of stops, airlines, etc.

I see potential for a computer/human combo in the future, but right now our flight searchers are doing great and we don't feel that automating the process would make the business significantly better.

In most cases, A/B testing will actually do negative damage than if you didn't test it at all.

He did say they "don't do a ton," so they are definitely doing some, maybe that's where?

That said, I do appreciate the sentiment here. This is a different context entirely, but in my ecomm past a/b testing was often employed by those seeking the politically safest growth path. I considered it proactive CYA where Product Managers would prefer premature optimization over having an opinion on what's best for the user.

I witnessed what I considered an unhealthy amount of a/b testing resources spent on micro adjustments eg moving a buy button a few pixels or changing the color slightly. There was very little calculated risk taking, which can be enabled with A/B testing employed differently. There's absolutely a place for those kinds of optimizations but I think they come much later in product maturity than when many organizations actually begin using them.

Then there's an entire other discussion about understanding the math behind building a relevant test and interpreting the results.

what is the one for low cost Amazon items? I've been thinking about starting a daily store of weird items but this seems like a good idea also.

Like sub $20 gift ideas/wacky items

Solid ideas but I think they may struggle with lower margins. With flights you save X00s but with Amazon X0s. Saw this with all the Airbnb for...since hotel replacement is similarly a higher margin/value purchase.

On that note you may be able to this with cars, hotels, real estate, luxury watches, computers, etc

What is happening here? http://imgur.com/a/1c8Nj

You posted this identical comment both here, on an account that has only a few posts over the last 5 years or so, and on what looks like a throwaway.

As far as I can tell Theflightdeal is basically the same service -- a small army of people searches for fares (HOPEFULLY WITH SOME AUTOMATION!), checks whether there are any results whose price is very low relative to historical prices they have logged for that route, and publishes a blog post if the price is very low. They have little blurbs about each destination prewritten so it's very easy to make the blog post once they see a good fare. ("Good deal" is computed either on a cents-per-mile basis or on a low-price-per-segment basis for certain special routes.) Often a human reads through the fare rules (I'm guessing they just use expertflyer, heh) and adds a bit of extra color like booking class, advance purchase restrictions, purchase-by date, etc. to help save people some time. That blog gets money from affiliate revenue (click thru from their blog post and book travel and they often earn a percentage). Very clever, tremendously successful in the niche.

Looker pivoted into doing something similar but with a mobile app (after first trying to do REALLY hard big-data work I think). They seem to perform the same kind of analysis, but with more automation and less reliance on spreadsheet. They use a really interesting input stream (real-time search query results from a GDS -- so, actual booked flights) to alert people when a route they care about is cheap vs typical prices.

Other services exist which do the same thing but for premium-cabin travel; these services tend not to be public facing and try to charge ridiculous prices in part because J/F travel carries a certain cachet and also in part because those deals tend to be considerably more fragile.

There are afaict somewhere like 10-100 of these "deal newsletter" business models doing basically the same work as the deal blogs but trying to make money on subscriptions. Very clever, I definitely respect the hustle.