Redeye VC

Josh Kopelman

Managing Director of First Round Capital.

espite being coastally challenged (currently living in Philadelphia), Josh has been an active entrepreneur and investor in the Internet industry since its commercialization. In 1992, while he was a student at the Wharton School of the University of Pennsylvania, Josh co-founded Infonautics Corporation – an Internet information company. In 1996, Infonautics went public on the NASDAQ stock exchange.

Read more or visit First Round Capital

Monthly Archives for 2010

View the older monthly archives »

The Penny Gap

Lincolnmemorialpennyreverse120_1Updated - See bottom of post

Here at First Round Capital, we see a lot of business plans for consumer-facing internet services. Most assume a significant portion of their revenue comes through advertising -- but almost all of them have a "premium/subscription" option.  Typically that subscription revenue accounts for 20-40% of total revenue, and is based on a very low ($1-5/month) subscription fee.  When asked to support these subscription revenues, entrepreneurs almost always say "well, it's very cheap ($2 a month) and we're only assuming 5% of our users take advantage of it)."  On the surface, a reasonable position. 

However, that is rarely how things play out.  Most entrepreneurs fall into the trap of assuming that there is a consistent elasticity in price - that is, the lower the price of what you're selling, the higher the demand will be. So you end up with hockey stick looking revenue charts that go up and to the right, all supported by an "it only costs $2  month" business plan.


The truth is, scaling from $5 to $50 million is not the toughest part of a new venture - it's getting your users to pay you anything at all. The biggest gap in any venture is that between a service that is free and one that costs a penny.   I can't think of a single premium service that has achieved truly viral distribution.  Can you?


Consider the pay-per-download music sites of the late '90s. None came even close to matching the widespread popularity of Napster or Kazaa. By 2000, Napster was estimated to have 40 million registered users, with as much as 80% of external network traffic on colleges consisting of MP3 file transfers. Kazaa has had almost 400 million downloads of its client to date. Assuming 2 downloads per user that means around 20% of all Internet users have downloaded Kazaa. And the same phenomoenom is occuring now with movies via Bit Torrent  -- as opposed to CinemaNow or MovieLink.  That's the power of free.

At some point, the cost to acquire a paying customer is so high, it makes sense to consider shifting from a pay model to a free model. In these cases, asking “who would pay to reach these consumers” (or "who can subsidize these users) creates an opportunity to build a more valuable business through the combination of exponential growth and targeted advertising. This is what we're seeing today with Jingle Networks, owners of 1-800-FREE411. By harnessing the power of free 411 calls, Jingle has already managed to capture around 5% of the overall 411 market.  And it's interesting to note that "cheaper" directory assistence services such as Easy411 and 4114Cheap existed for years before Jingle was conceived and have gone nowhere.

It happened in music.  It happened in movies.  And it's happening in directory assitance. Now I’m looking for other industries that are going to be converted. If you've got a plan that uses the free model to get that first penny and disrupt an industry, I'd love to hear about it. It’s a great way to shrink a market.

Update:  There have been a lot of wonderful comments/discussion about this post and I wanted to be sure I was clear on one point.  I am not advocating the death of premium services.  Nor am I stating that "free" is the penultimate business model.  Rather, my primary point was to state that many people (mistakenly) believe that getting a consumer to go from "free" to $1/month is just as difficult as getting someone to go from $1 to $2/month.  I think that there is a huge burden to getting a consumer to pay anything -- and entrepreneurs tend to underestimate the level of effort.  If you can deliver enough value to charge for your service -- and cost-effectively attract a large base of paying customers -- of course you should.  However, if you find that you are able to attract a large pool of free users, but can't convince enough of them to pay you -- perhaps you should look at other ways of extracting value.  GigaOm said it best

To be fair to these VCs, they’re not advocating doing everything without pay. They’re suggesting free as a tactic towards getting paid in other ways: through advertising, or by premium services (as in a freemium model), or maybe even through being acquired by a company with a large wallet. Free is only a tactic, though, not a business model.

Conflating the two misleads web application developers into thinking they don’t need to do the hard work of figuring out what’s really of value to users before they build and launch their online service.


Jason L. Baptiste

Josh, I think the enterprise market is going the be the real breakout for entrepreneurs and VCs. Take the amazing features and usability from the consumer side and bring it to businesses. You will get the viral word of mouth effect of the consumer market, but the licensing revenue model of the enterprise side. That isn't going to merely disrupt, it's going to tear the walls down.


Robert Dewey

I guess it really depends on the business. If a company started up and presented themselves as a free P2P marketplace (think eBay, but free), I doubt they'd win over many users.

I've been trying to position our startup, Wantsy, and decided to go with the advertising market -- specifically, the non-traditional stuff like offline advertising and graphics-based (where I myself have felt the pain).

Our model is reverse of the typical model -- it will be the advertisers who list their need, and the distributors who bid down (even automated both ways by RSS). Rather than charge people to list, we'll employ a CPA method... If a distributor can fill the advertisers needs, we get a small cut. Otherwise, it's free.

Ex: I'm a local business and would like to advertise on a local radio station. I set my attributes, tag my demographics and medium type, and upload my audio file. Radio stations could find my listing via their RSS feed (or a site search) and upload an offer. Other radio stations could counter offer. The advertiser chooses an offer, and the audio file is sent to the station.

It works the same way with other mediums... And we don't take a cut unless an offer is accepted.


I'd say that flickr is a pretty good example of a successful viral premium service. I've had many many friends and acquaintances pay for their pro flickr accounts simply because all their friends have existing accounts. Once they hit a threshhold, most of them end up paying the annual fee to upgrade to pro accounts because of their existing network and because of their initial investment of uploading their photos with their free accounts. These are people who rarely pay for any premium service online.

Richard Johnson

I agrre with Dave.Flickr es a good example of Freemium model. However, Flickr is one per thousand successful case. If you not offer a massive services like photo storage, how it'll achieve overcome the penny gap. Is that possible?? Will Basecamp be the the best case for non-traditional service with freemium business model??


"I can't think of a single premium service that has achieved truly viral distribution. Can you?"

How about Next to nothing per month sub model but they do volume.


Couple of thoughts...

#1: With a paid service you get a smaller customer base, but I believe loyalty over time is higher.

#2: Pricing is always tricky. On-line I think this is even more true. I'm not sure there is much difference for a customer between charging $2 and $5. It's the same amount of effort to get out a credit card and pay and is still an impulse purchase.

This pushes pricing online into groups with somewhat elastic upper and lower limits. The groups vary based on whether it's a one time or monthly fee. e.g. for a one-time fee the groups could be $0.01 - $5, $5.01 - $25, $25.01 - $50 etc...

And that means many who are charging are probably leaving money on the table...

Alex Iskold

I think that this is unfortunate that we've gotten to the point when people think paying for software is not a good thing.

Looking at traditional media business which is ad-supported, it still has a fixed price component for the most part - major magazines charge annual subscription fee. Why? Because there is nice scaling with number of customers.

Another unfortunate thing is that ad-supported businesses need to have two sales fronts - one customers and another one - advertisers. This makes it complicated because many software entrepreneurs (myself included) do not understand the ad game well.

Anyway, we can't ignore the facts, this is where we are today, but I would actually prefer 37signals model.


Andy Swan

eBay and Paypal.

Good post and right on despite a few exceptions.

Andy Swan

Oh...also PartyPoker. :)


I really enjoyed this post that made me think about my E-business MBA class took in 2001. At that time the advertising model had already started to be completely banned and startups would never been considered seriously with a business model based on that revenue stream. Things look quite changed today!!

Nik Cubrilovic

"Getting users to pay you anything at all"

So true. I also believe that there is such a thing as 'too cheap' for some services, so the price/demand will have a bit of a hump - and you have to find the apex of that hump. We saw this behavior with Omnidrive with licensing deals

trading is prime for a disruption with a 'free' service, see:

'blown to bits' is the bible for this topic:

Kevin Farnham

I thought a lot about this in conjunction with my Physics background. The discontinuity in the price/demand curve is similar to the discontinuities studied by Quantum Mechanics.

The Web induces a kind of "Quantum Economics" where the standard economic price/demand curve necessarily has a gap, because the true value of a unit of Web service product is much less than the smallest unit of currency (1 penny in the U.S.).

Just as quantization of mass and energy proved highly disruptive for 20th Century Physics, the quantization of currency that Web services highlight is proving highly disruptive in our 21st Century global economy.

Bruce Franco

Charts without data to back them up are are just as arbitrary as entreprenuers with a freemium business model. Without additonal infomation it only appears that you are making self-serving arguments to reduce an entreprenuer's business valuation.

The extent to which this type of business model works is dictated by how much choice the consumer has. If the venture has good barriers to entry in a non-crowded space, then it is more likely to work better.

Your points would be much better made if you gave some supporting data on how some of these business models have fared....or do you just encourage the venture to drop their premium service approach altogether?

Jim Bursch

"Looking at traditional media business which is ad-supported, it still has a fixed price component for the most part - major magazines charge annual subscription fee. Why? Because there is nice scaling with number of customers."

The revenue that comes in from magazine subscriptions is incidental. The real value of a subscription is the signal that it sends to advertisers -- we have motivated, qualified readers (who else would subscribe?), therefore we are entitled to command higher ad rates.

The advertising media market is extremely distorted. It is a third-party payer system, and accordingly very inefficient and does not conform to obvious supply/demand assumptions.

I discuss this at the blog for MyMindshare ( and I am building a true mindshare market at


Great post, but your last line confuses the issue. Yes, you're right that the demand curve is not linear. There are still plenty of online companies that have built successful premium services - that doesn't disprove your claim. And services that are truly viral tend to be more successful than those that are not, whether their business model is fee or free (or freemium).

eBay is viral (sellers derive more value from the service if they spend their own time and money recruiting buyers), fee-based, and undeniably successful. But to your point, there's no middle ground between eBay (quite expensive) and Craigslist (free).

Mike Sabat

Yes, free will spread faster, but can this model work?

In the post you asked "who would pay to reach these customers?" and meant, where can we find advertisers that want to speak to these customers.

When I ask if the model works I see the question as "who will pay to reach these customers (you know, these 'customers' that won't spend a penny with me)?

Account Deleted

The start-up I am a part of (Atomic Brokers) is one of those companies that shrinks a large market (i.e. the $60 billion residential real estate brokerage market). The nice part is our revenue streams break into referrals, revenue, and a 'secret sauce' piece that really holds the key to our long term success. Definitely seems to fit the criteria you speak of.


I don't really know anything about business, so most of this post and the comments went over my head.

That being said, comparing Napster and Kazaa of the late 90s to pay services in one paragraph, and then beginning the next paragraph "At some point, the cost to acquire a paying customer is so high, it makes sense to consider shifting from a pay model to a free model" in an attempt to make some overarching point is ridiculous. Napster and Kazaa were services that facilitated stealing, not business models.

Jeffrey Zwelling

Efax is a good example of premium subscription service that went massively viral.


Completely agree that free vs. subscription is binary when viewed from the perspective of an online media consumer. Services are either free or not free. Pricing is not as important, as other comments state.

However, the thing that makes all this so interesting for me is that your curves would be slightly (but meaningfully) different if you drew them for South Korea or Japan.

Premium services with viral distribution are easy to find in those countries.


RE: "I can't think of a single premium service that has achieved truly viral distribution. Can you?" Just look at just about any top ranked sites list and you see some paid subscription services. Still, your point is well taken, although I’m surprised it comes as a shock to some of the startups that you’re talking to. Cheers, chrisco


my take on this is that ever business should have a “free” strategy but not necessarily a free business model.

That is, figure out how to give away as much of your product as possible, decide where the line of value is, and then charge people to cross it.

At Second Life, we first thought that line was admittance to the world (i.e. subscription to get in). Once we realized that the line was beyond admittance but more about creation, ownership, etc, the access model became free (since there’s a low cost to each incremental casual user) with various upsells and other ways of monetizing your activity in-world.

Scott Rafer

@ Update:

"before they build and launch their online service."

Yes and no. Whatever you think about the value before launch is wrong. Only the first para of that quote universally applies for consumer apps.

Don MacAskill

Great post!

As for a "single premium service that has achieved truly viral distribution", I suppose that depends on your definition of "truly viral".

At SmugMug, 60-80% of our customers come from referrals. We think it's close to 80%, but there's some "unknown" data in there, so we're not positive.

We've always had >100% growth year-on-year, have a basically non-existent marketing budget, and have a fanatically loyal following. Oh, yeah, and we've been profitable since Year One.

By my own definition, we're a premium service with truly viral distribution - but maybe your definition differs.


Howard Freidman

I agree completely that there's a chasm between free and cheap. But its not one size fits all. The key for a new business is understanding the size of "their" chasm, which I believe varies based on:

1) (obviously) Whether the item in question is worth paying anything at all for. There are some things that simply won't warrant enough payment to make it worth paying, or in some cases collecting.

2) How painful is it to pay for? People pay more money in tiny chunks for ringtones than anyone probably ever imagined. But they're painless to pay for. The charge just shows up on a bill I already pay. I'd buy lots of things with a universal tab that I wouldn't take the trouble to buy otherwise.

3)Expectation management. Customers who pay money expect support; customers who get something free don't. There's an organizational critical mass required to support paying customers; for large user bases the resources must be front-loaded. That's expensive and requires commitment to, and focus on, paying customers. If you're built to be free its hard to deliver service that meets the expectations of paying customers.

The great news is that there's a *ton* of money in traditional mass advertising that's a complete waste. This time around though most advertisers understand that and so there should ample sponsorship dollars available to pay for innovative free services that earn peoples attention.



Jason Wood


Thoughtful piece. I'm on the board of a company facing a related dilemma.

Entering their 5th year as a full-subscription service, growth has stagnated despite fantastic customer feedback and an unbelievably compelling (and quantifiable) end market.

This has led the a discussion of whether to transition away from a subscription model (where pricing has gone up at least 10% per year with almost no churn) in favor of an ad-driven model.

The founders were spooked by an ad-based business model having started the company right after the bubble, but now (obviously) times have changed.

The founders would be most comfortable with a "hybrid" approach; but how will long-time paying subscribers react to having X% of the content they viewed as propriety now being offered for free?

I'm not sure it shouldn't be all or nothing.




The only way to close the penny gap, especially in a business where people are already giving it away (music is a great example) is to add significant value, usually to abstract concepts like experience and interface.

Apple's iTunes is the best example. They entered the game way late, after thousands of college kids were already downloading non-DRM music for free. Bring the iPod, iTunes, tight integration and slick design and BAM...

those same kids are downloading songs for $2 a piece, even though the feature set and technology of apple's suite is not significantly more advanced than their (free) competition.

Andrew Parker

This is not an problem of economics, it's a usability problem. The act of paying for something online (regardless of the cost) requires collecting so much more information (CC#, Paypal Acct, etc) which is subject to data entry errors and form fatigue. Plus, users still don't have a means of paying for things on the internet that they can trust unconditionally. (In user testing, I constantly heard the complaint that users don't like giving their CC# out on the internet).

If paying $0.01 for a service had the same barriers to entry as paying nothing for a service, then I think the cliff on your graph above would almost completely vanish. There would still be a tiny cliff to account for no being able to use "FREE" in branding, but the cliff would be an order of magnitude smaller.

Dawn Douglass

I like the model of "let the buyer decide what price they are willing to pay." This is what Amie Street is doing for music. I'm doing something very similar for comics. Rabid fans who put in the time and passion and find the best pieces can get them free or for just pennies, while those who come in late and wish to ride the popularity wave that others have started, pay more. I think it is a monetization method that is both fair and elegant.


You wrote: "Nor am I stating that "free" is the penultimate business model."

"Penultimate" means "second to last". It's not a variation on "ultimate" as "very best."

Dawn Douglass

Ian, I was actually impressed that Josh used the word "penultimate" correctly. I took his statement to mean that "free" isn't necessarily the next to last business model, with advertising revenue or subscription -- or whatever monetization method you come up with -- as being the last one.

William currently using the "free" model with its Superdistributio serice. You can read more about it on my blog here

Hugh J. Sloan III


All good save for the domains that can be classified in the synthetic opiate category of viral user must-haves: online wagering, fantasy sports, porn and dating. Only Werner Heisenberg could truly plot these on an x/y coordinate system. You will soon be learning about a consumer internet play whose trajectory will need to be measured by redshift spectroscopy via Hubbel. If its narcotic to the consumer, they will fork over significant coin.


Ultimate Problem!

One edit desperately needed: Penultimate means second-to -last, not ultra-ultimate. Use either ultimate or for truly the ultimate,
Please--it distracts spelling nuts from the excellent content.

Tom Hobson


Makes sense.
The charts deliver a relevant economic concept.

How is the payment component of LinkedIn working out? I think there might be more revenue available in a model where the user pays per instance, rather than bundled packages [ie $3 per InMail or something to that effect].

On the flip side..
Why does i-tunes work? Is it because it is "pay as you go", or because people are invested in the hardware?
Does a site like count?
What about the
Perhaps each of these fall into the "narcotic to the consumer"[good phrasing] category.

Altogether, I think it is a good exercise to dream up atypical revenue streams- but if subscription is feasible, then go for it..

Tom H


Exploring the the best way to invent new ideas...might be this opportunity will open new doors in your life. No.0001146g.html, is giving the feel of this to all its franchisee and member. We are giving helping hands to make the future of India bright.


Gucci "treasure" black small boston bag delivered to anywhere

"treasure" small boston bag with double handles, key/push lock closure, signature web, chain and circular charm with engraved gucci trademark, bottom feet, and inside zip pocket. 9.4 x 5.5 x 7.5 inches. black guccissima leather with red velvet web, black leather trim, and light gold hardware.

Please check out our website: There are top quality of replica handbags for sell
with perfect weight, feel, and like the originals.or email us :

Keith Johnson, M.S.

Yes, the "penny gap" is an important concept for all web entrepreneurs to understand. What ultimately matters is value, and if that value is strong enough, the customer will pay, I believe. That is the true challenge.

Brett Hurt

In today's WSJ, the article about eBay's writedown of Skype illustrates the points Josh makes in his post very well.

trademark registration

"A penny saved is a penny earned." It's always important to provide customers with good value, especially when it comes to web sites.


where o where is the redeyeVC these days?

Kevin Burton


Your second graph should be a power law distribution.

However, if you focus on the tail (after the elbow) you'll notice it forms a more linear distribution like your first graph.

So you're both right essentially.

Also, it depends on the product. For First Round you're betting in companies that would IDEALLY have a $1B exit.

These numbers are different for midsize companies. There are a lot of companies that make a pretty decent amount of cash by charging users.

Of course these would typically be consider 'lifestyle businesses' by VCs.


Dimitar Vesselinov

"Currently LiveJournal has five account levels: free (comprising approximately 95% of the network); sponsored with advertising; "early adopters" who were registered prior to 2000; paid and permanent."


nice site! and interesting topic. if some wants to understand better and needs the book- go to rapidshare search
they have much info including electronic books


Spent time at a top tier school studying exactly this question. Coupled it with research on internet consumer behavior. Pretty sure I have it figured out. When I solve the free-riding problem of incurring all the costs to retrain customers for businesses who don't contribute, then you'll see it roll out. Swear to god you'll laugh when you see it.


good nice

Mick Liubinskas

I think the maths here is right, but sometimes it means that you just need to charge more for the premium service given that if people are going to pay, they will generally pay more than you think.

But I'm halfway through a blog post on this, so I'll just finish it off.

Keith Johnson

This is one of my favorite articles/posts on the entire Internet, because it addresses the challenges to defining a working business model, especially for business on the Internet. I agree with many others here who recommend one offers a basic service or subscription for free, and then you charge at a pre-defined higher level of service or subscription. It costs money to stay in business, and if the customer is unwilling to pay, then they do not deserve to receive your higher quality good or service. I am sure all of my college economics professors would agree with that, even the liberal ones. Regards, Keith J.

2moons dil

It is a boring thing.

Post a comment

Comments are moderated, and will not appear on this weblog until the author has approved them.

This weblog only allows comments from registered users. To comment, please Sign In.