How do free services on the web make money? | Hacker News Reader

How do free services on the web make money?

forbes.com

Interesting coincidence - last night I finished my first post in a series of posts about startup business models. Specificaly, advertising: http://blog.chewxy.com/2013/03/06/startup-business-models-ad...

My startup is going thru a series of heavy re-thinks, so I thought I would blog about the different models, and how to approach them in a rather thorough fashion.

Thanks chewxy. The only issue is that advertising is getting increasingly sophisticated and contextual. The big boys (Facebook and Google) have way too much data and crowd out the smaller startups. So, startups should start thinking beyond advertising as a main source of revenue.

I maintain that if you know what you're doing, advertising is ridiculously profitable. Most of the time though, most startups that use advertising do it because: they don't have other business models, and so advertising is shoe-horned in.

On the other hand, a startup that knows what they're doing will know that building out the correct audience is the way to go - if you provide your clients (the advertiser) what they want (quality, relevant audiences, preferably with intent to buy), you can do quite well.

Any examples to support your case?

Google (the search engine) would be the prime example. A search engine is the service that Google provides, and by virtue of its service, they build their product (audiences) by making the audience self-select into niches (i.e intents).

Google isn't a startup anymore. If I were to be asked about more recent startups, I'd list Buzzfeed, the 9gag peoples, Ben Huh's company (can't remember the name) amongst those with quite decent strategies for advertising. You may notice that they're almost all content based companies/startups - that's because it's easier to build audiences with content.

There's a big one missing from this list that is somewhat recent.

Selling the placement of audience segmentation/retargeting cookies on a publisher's site to a data aggregator like Bluekai (typically on a CPM bases) who then sells that aggregated data to DSPs/DMPs for use in display advertising.

If you want to learn about a whole different layer of the ad tech industry, just start perusing some of the questions on Quora for Bluekai (http://www.quora.com/BlueKai) and then do some digging on the companies in the Data Aggregator and Data Supplier buckets of the LUMAscape for display advertising (http://www.lumapartners.com/lumascapes/display-ad-tech-lumas...).

I wonder how they're going to work around two of the four major browsers blocking third-party cookies by default.

Ironically, the original author (who has some sort of Quora affiliation) did not include "generate content and resell it to other free web services" in the list.

Here's one he forgot: I've seen niche content sites that after building a community hold conferences. They may have advertising but it's the conference that's their main money maker.

That's a good feedback, rmason. It is a niche model that could be added to the article.

Its not that niche eg TED.

quite a lot of enterprise opensource software companies do that. And their gigs are not cheap ;)

Disclaimer: I'm going to comment on my personal experience running a free web application, http://sleepyti.me. It's not meant to be self-promoting, but I feel that the experience is relevant.

There are a lot of ways to make actual profit (not just break even) with free applications.

The Hacker News crowd doesn't always show a lot of love for the "small, free application" crowd, and the echo chamber is correct that having a webapp doesn't mean you're a startup.

I know that I don't consider myself the "founder" of my app, because my site is not a company. I'm not trying to make it a business.

That said, though, this write-up glosses over what has been, in my experience, the easiest and most straightforward way to profit from free traffic: non-targeted ads. For me, that means a single banner advertisement served over Google Adsense.

Now, like I said, I'm not trying to make http://sleepyti.me my primary income. If I wanted to charge users, or enter deals with, say, large mattress vendors, that would be different. But that isn't to say that the traffic and feedback I've received hasn't helped me both personally and professionally.

Personally, I've received a lot of attention from my app. It's pretty popular with college students, and it's gained media attention from outlets such as Lifehacker, the Toronto Star, and most recently the New York Times.

I'm not listing these sources to self-promote, but to show that from a personal exposure standpoint, it's opened many doors for me.

The same goes professionally: I had a brief partnership with YC-backed WakeMate, which, although now defunct, allowed me to interact with an early-stage YC company that I otherwise would not have been able to experience. I get between 1.6 and 2 million hits a month, and the financial reward has been steadily increasing through my single banner ad as well.

People always say that "if you're not paying for it, you're the product," but I don't think that that's always true. I don't prey on my users, and I can think of several free apps that operate in a similar--even less obtrusive--manner. Trello, for example, neither serves ads nor sells my personal data (I hope!). Sometimes it's worth it for a developer to create something and truly release it for free, without having the immediate motive of profit.

I always tell college students who ask how to break into an industry--whether it's information security, professional programming, the startup world, or anything else--to do something. Sometimes a free application, such as the one I created as a free weekend side-project, turns into more than what it is. I'm more than aware that my experience is mostly a fluke, though. The take-home message should be that even if it doesn't become some massively-profitable, highly-trafficked app or site, it's still something that you built.

And that has value.

While I love your site, I have to respectfully disagree with much of what you are saying here.

First, I agree on one main point--your site is not a "startup" or even really a "company" at this point. That said, you still very much have a product, and you are actively monetizing it.

But then let's get into this point by point...

You say that your ads are "non-targeted" but that is not at all accurate. Just because you aren't doing anything actively to target it, doesn't mean Google isn't. Google's AdSense network has a TON of targeting on the back-end that is determining what ads to serve. This can be based on the profile of the user, retargeting lists they may be a part of, the category of site Google has you categorized as (possibly health?) etc. Your ads are VERY targeted. That is actually part of the beauty of the AdSense product.

The second point I really take issue with is that you don't seem to think your site falls into the "if you're not paying for it, you're the product" bucket. Or at least that's how I parsed what you wrote there (apologies if I was off on that).

People aren't paying for your product, and so you are making money by serving them ads. By the nature of it, any ad-supported site falls into that bucket. You require more user impressions/eye balls and that is what you are "selling" to advertisers/Google. This is not inherently evil, and I don't think one needs to prey on their users to leverage advertising. That doesn't mean the saying doesn't apply to them though.

Again, I like what you're doing and think the single banner is a tactful approach (although you'd probably make a lot more by recommending sleep/health-related products and using affiliate links). But those two things you said just aren't accurate so I wanted to clarify on those a bit.

  > you don't seem to think your site falls into the "if
  > you're not paying for it, you're the product" bucket.
  > [...]
  > People aren't paying for your product, and so you are
  > making money by serving them ads. By the nature of it,
  > any ad-supported site falls into that bucket. You require
  > more user impressions/eye balls and that is what you are
  > "selling" to advertisers/Google. This is not inherently
  > evil, and I don't think one needs to prey on their users
  > to leverage advertising.

Saying that users are "the product" implies a predatory relationship. Predatory relationships are considered evil (or at minimum extremely undesirable) when the prey are human beings.

Exactly. I did the very same thing, made a small app that showed off my coding skills. It only made about $1000 through Amazon Affiliate sales over the course of a year, but it got some press coverage and got me numerous job offers, one of which I ended up accepting.

There is a lot of value to be had from providing a free web app if you are a lone hacker looking for a job.

So - the people who make enough to pay you a salary - what business are they in, and how do they make their money?

I'm not trying to be snarky. Well, maybe just a little, but in a good-natured way - I think you got a good deal that probably met or exceeded your expectations, which is commendable!

But you did not create a sustainable business. To do so with ads is not impossible, but since each user generally gives you way, way less revenue than with something where they pay for it, you really have to parsimonious with resources per user.

Also a 1000 bucks is a thousand bucks.

In otherwords, you made sleepyti.me a loss leader for your personal growth. That is a fantastic way of looking at it.

(Also, I <3 SleepyTime and I would pay like $1 a month or something :D)

Agreed. I have gotten many job offers and even just a foot in the door due to running http://searchco.de because its interesting to other coders and requires a degree if tech knowledge to run. Every interview I have had since I started running it has asked about it, what I learnt etc... I see it as personal marketing more then anything else.

I use sleepyti.me five days a week.

At some point, im pretty sure that you had a notice on the site awhile back that there would be an native mobile version. What happened to that?

A mobile version could probably make a fair chunk of change. I'd buy it.

If you are a solo entrepreneur who is running a website part-time and furthering your personal brand, then advertising is not a bad idea. However, it is different if you are working full-time along with a team and building something bigger.

I found out a few years ago that my sleep cycle is actually 45 minutes, and not 90 as sleepyti.me suggests. Is there a way to make it work in 45-minute steps?

There is a lot of variance in sleep cycles between individuals, between nights, between different types of sleep. Saying 'a sleep cycle is 90 minutes' is like saying 'The left side of the brain is for logic'. There is a rough element of truth there, but nothing to hang your hat on.

Source: once worked as an overnight sleep lab technician, collaborated with sleep specialists in next job working for the lab equipment maker.

Also, a timer with 90 minute steps inherently includes 45-minute steps. If you do desperately want the in-between steps, it's a trivial mental calculation on the four steps you do get given at the site.

You're right that it is trivial to calculate it myself, but then again, it is trivial to calculate it without using the website. I thought the whole point is that you do not have to calculate anything ;)

Anyway, I believe the site should allow "set custom cycle". As you wrote above, it varies between individuals a lot.

How would it be different? show more times in the morning? The 90 minute cycles it shows you are still valid 45 minute cycles too...

I assume you need more cycles than people who have 90 minute cycles?

By 'hits', do you mean page views or visitors? Just curious.

Advertising sucks because worldwide inventory grows faster than the supply of eyeball-hours to view the ads. So there's constant downward pressure on prices.

Subscription sucks because you have to fight subscriber fatigue ... if you can get someone to sign up at all. People sweat over $5/month memberships; meanwhile they spend $20/wk on coffee.

Micropayments suck because everyone hates the mental overhead.

I have a solution I call "microsubscription". The only problem is that it's been tried by others and failed. Naturally I feel I've got the magic ingredients that the others didn't have, but ... we'll see.

And yet, there are still sites that will command a high eCPM. The difference is of course, the audience quality.

Like I mentioned in my latest blog post[0],

> Real time bidding engines are getting more and more intelligent, and would be able to guess with quite good results, how valuable a user is to the advertiser. This allows the advertiser to only pay high amounts for audiences that matter to their ROI, while paying low amounts to audiences that don’t.

> What happens is then this: a segregation of high paying advertisers and low paying advertisers form. When I say high paying advertisers I mean advertisers who would pay high for the right audience. They will pay significantly higher than the low paying advertisers, who would “spray and pray” their ads in hopes of reaching the right audience.

I maintain that if a startup is able to build the correct audiences (StackOverflow building an audience of IT professionals for example), the revenue from advertising will be substantial.

[0] http://blog.chewxy.com/2013/03/06/startup-business-models-ad...

My original motivating problem was: how do blogs I like make money?

Blogs that aren't about answering Oracle Obscure Platform 6.75a-R2 questions. Blogs that aren't about mesothelioma. Blogs that aren't "review" websites.

Honest, good old fashioned smart people writing thoughtful stuff.

Based on who I've talked to in areas of business and politics and the visible spread of ideas in Australian public intellectual life, the demographics for the blogs I directly host are amazing. Highly paid professionals all round. They're read in ministry offices and C-suites all over the country.

To scrape some of this high-quality audience $$ I'd need to adopt a very high touch advertising policy and use invasive user tracking to prove my demographics. I don't have time or the moral flexibility and neither do the bloggers.

Hence: how do the blogs I like make money?

(Answer right now is: at their dayjobs.)

The technology I've developed allows users to be tracked visiting participating websites without the websites being able to piggyback on my scheme to track users across multiple sites. It's resistant to the visit falsification attacks present in any standard tracking scheme.

Pointless without a business, of course. So watch this space, I suppose.

Ever thought of advertorials? Advertorials + rev share would be a good idea had you not listed "moral flexibility" in your list of constraints.

I noticed you host John Quiggin. I used to read him a lot. This is how he would use his blog to make money: sell his books to his blog audience. Tyler Cowen, the guy behind Marginal Revolution used to have this trick where he would give you access to his private cooking blog if you bought his book. Some form of managed affiliate system would be my best guess.

Also, mesothelioma is soooo 2005. :P

When I wrote my honours project (which was the first version of the tech I developed) I mentioned merchandising; it's pretty hit-driven though.

I personally make money on affiliate links to Amazon. By which I mean "I get a minute subsidy on my book habit".

Some of the bloggers make a few dollars on advertising, but the amount has steadily fallen because they aren't sufficiently narrowly focused to be worth finding a niche advertiser. And they aren't sufficiently widely read that a more general A/B brand-awareness advertiser would pay much over open-market CPM.

You seem to have found the problem by yourself. IMO this would be the trough of sorrow[0] for any startup with advertising as a monetization plan - the growth but no growth phase.

In my blog post I mentioned that here there are basically two paths to go - both of which you have identified. If you want to stick with advertising, choose one?

[0]http://www.avc.com/a_vc/2012/03/the-startup-curve.html

I am choosing to invent a new model, actually :)

pray share (microsubscription has been done before as you said). How would you do it?

Email me.

Edit: "invent a new model" is hubristic code for "independently invent a new model that actually can be analogised to some pre-internet models".

I have been using a microsubscription model for an MMO strategy game for iPhone/Android that I helped develop. (http://spaceuncharted.com) We charge $1/month or $10/year for access to the full game. Here are some things we've learned:

-- Apple only allows auto-renewing subscriptions for magazines, newspapers and other periodicals. So, for us, users have to re-subscribe every month, which has lead to very poor retention rates. On Android, we use auto-renewing Paypal subscriptions, and have had more success.

-- Having to manage separate billing systems for iPhone and Android adds a lot of development time and headaches. The Apple billing system is particularly time consuming because the errors you get if something isn't configured just right are not helpful.

-- Subscribers can have very high expectations for a service that is costs $1/month. When they're paying, they (rightly) demand a high quality product. We've been very liberal with giving out refunds or credits when bugs have arisen or we've had server downtime.

Is "microsubscription" already the term of art in game development circles? Because I'm talking about a quite different funding model.

If the term is already being used to describe low subscription fees I'll need to thinking of something else.

FWIW, the tech I've lodged with the patent examiners could be extended to cover games as well (and I made sure to write a game in as an example). Single payment/funding scheme for web sites, web apps, phone apps etc is possible.

Edit: some googling sees a multitude of definitions (I wonder why I didn't notice this before?). I guess good names are hard to find.

Nice list where the author is trying to make a comprehensive list, but then I also have some criticism of the article. It lacks some models, and it makes an implied relationship between for-profit and non-profit. In the English language, this is commonly refereed as making money vs raising money. A charity or non-profit (say Wikipedia) is commonly not talking about making money.

Some projects is funded by an indirect sponsorship model. Best example here is google and their model of giving 20 percent of an employees time to do what ever project that the employee wants. While Khan academy model is direct funding, projects that are an result of the google employee free time model is indirectly funded project.

Then we have non-profit and what they will consider to spend money and man hours on. Sometimes, that not even remotely related to their primary purpose, but is just something that interest the people in the organization.

Note that many of these fall into a bait-and-switch category. You use the product for free for a while, and then all of a sudden you are spammed to high hell, or functionality is taken away unless you pay.

Paying isn't really that bad of an option sometimes.

That's not bait and switch.

> "Bait-and-switch is a form of fraud used in retail sales but also practiced in other contexts. First, customers are "baited" by merchants' advertising products or services at a low price; then customers discover the advertised goods are not available. Other products are "switched" for them; however, these items are often costlier." https://en.wikipedia.org/wiki/Bait_and_switch

Bait and switch means that you're promised or enticed with one specific thing, but that thing is not available and you're offered something different. The idea is that you wanted something, so fuck it just buy it and get it over with. And of course some people won't even notice the switch.

Free now pay later and the other schemes are not bait and switch. When you get there/sign up, you have the free version right now, exactly as advertised. And there is no attempt (usually) to hide the fact that free is only for N days, after which you'll need to sign up.

Similar for free/paid for reduced/enhanced. You're getting exactly what's being offered.

But they are bait and switch when the terms and offers change after you sign up. Instagram for example.

Bait and switch model is a pretty old concept (promo offers, free trials, etc in consumer goods).

Those are not bait and switch unless the buyer is told that the introductory price was the permanent price. Bait and switch is a form of fraud where an offer is made and accepted, technically constituting a legal contract, but then the offerer reneges on the deal, claiming some sort of snafu beyond his control. Having not made the offer in writing and having apparently made some sort of honest mistake make the buyer unlikely to hold it against him, even though it is often technically a breach of contract. Oh, well, the buyer reasons, I may as well trade up and finish this thing, now that I've put so much time and energy into it. And that completes the bait and switch.

Simply offering "first three months free, then we start charging" is not bait and switch. It's a perfectly valid offer.

They don't. They hemorrage money in an attempt to get some claim of silicon valley gold.

Gullible VCs.

I clicked the link and got a full page ad and thought that was the joke.

Many times they don't. It's usually something like Instagram where they give out a product for free in hopes that piles of people will start using it, and they'll "figure out something later". Worked out great for Google and Instagram, not so great for 999,999 other startups who tried it.

Personally, I like the idea, but it's tough.

I always thought if you have a product that works but is free , you can commditize it by selling your infrastructure as a service ( paid API , hosting , etc ... ).

For instance , instragram could sell a way to make instragram clones in the cloud, like wordpress has a free blog script and sells blog hosting for that script, you get my point. Once you understand a domain and develop an app for that , should be easy to sell your app infrastructure as a service for thirdparty apps.

The problem? They can't really sell that.

Want to make an Instagram clone? Write a web interface for Imagemagick Youtube clone? Interface for ffmpeg... This gets boring quickly. :-)

Reply on news.ycombinator.com