How to create a profitable Freemium startup (spreadsheet model included)

http://andrewchenblog.com/2009/01/19/how-to-create-a-profitable-freemium-startup-spreadsheet-model-included/

by prakash • 18 years ago

80 26 18 years ago

pclark• 18 years ago

a lot of "social" startups are services where advertising online does not generate users.

Fuca• 18 years ago

You mean the users of such a startup are not the target for any market? I would think there are more profitable profiles than others but i would think anyone has to buy something.

jfornear• 18 years ago

I'm not sure what you mean by that.

I run a small social network, and advertising has 'generated users' at a much higher rate than any other method.

steveplace2• 18 years ago

That revenue model in the graph is quite Madoff-esque.

jwesley• 18 years ago

I was hoping the included spreadsheet model would be enough to secure funding.

steveplace2• 18 years ago

First you have to overlay an exponential curve of user growth, and then you're set.

fallentimes• 18 years ago

Welcome to finance!

Seriously though, when I worked in finance I only saw _one_ model that understated what actually happened (Projections Vs. Actual). The rest were "optimistic".

rokhayakebe2• 18 years ago

If you must spend 80k to make 120k you are in a bad business. EDIT: Really Bad.

utnick• 18 years ago

really?

a 50% ROI seems pretty good compared to traditional businesses

ejs• 18 years ago

But everything else is not profit, its 120-80 = 40k left for all the other costs, including a wage...

rokhayakebe2• 18 years ago

Lets note that 95% of small businesses make less than 250k.

If a business spends $800 and makes $1200 that is great. The risk is small. If the same business spends 8k to make 12k that is great. The risk is bigger but relatively small. If the new hired gun, self-proclaimed-marketing-guru, decides to jack up ad spending to 80k to make 120k, someone is betting too high now.

Asking if this model is scalable is bringing emotion into this scenario which will guarantee you failure sooner or later.

At one point you will reach the amount at which spending more on ads will not increase your revenue. I guess that is when justifying "branding" comes.

apsurd• 18 years ago

the article is analyzing "cost of acquisition" relative to revenue. As noted in other replies, this is not including total operating expenses. So for the business, the actual net profit margin is going to be lower than this.

tjr• 18 years ago

Would this method scale? As in, if I spent $8 in advertising, could I make $12?

andrew_null4• 18 years ago

in paid user acquisition, unfortunately, the common case is that you run out of ads to buy. So then you have to start buying crappier ads that perform worse, just to get to the next level of revenue.

One of the biggest examples of this is eHarmony, which is rumored to have $100M+ revenue, but terrible margins because they have to plow so much of that money back into advertising.

bmelton• 18 years ago

I dunno. If it scales, I'd consider myself lucky to find something with a similar return.

andrew_null4• 18 years ago

I'm the original author for the blog - the numbers are fake, and were chosen to make the graph interesting :-) Your mileage will vary!

Note also that you pay upfront for acquisition but your revenue base builds up over time - the more momentum you get, the better off you are.

It is however true that a lot of businesses that rely on paid acquisition end up with low margins - just look into the leadgen industry as a good example of this. This is particularly true in the Google Adwords world where auctions systematically drive up prices.

That said, if you can pay $80k and get $120k out, and that scales up to LOTS of revenue, then making $1.2B in revenue per year on $800M in cost may not be that bad. Certainly better than a lot of businesses.

run4yourlives• 18 years ago

Sounds like a profitable business to me.

daveungerer3• 18 years ago

As someone who's been building financial models for a living up until recently, I have to say that this model isn't very useful as it doesn't answer the most important questions.

The OP focuses entirely on the time period where the revenue line finally catches up with the cost line, i.e. profitability has been reached. But guess what? See all the time periods before that, where the business is running at a loss? Where is the money coming from to fund those losses? That's the up-front investment required - VC, loans or digging into your own pocket.

So that's the first important yet unanswered question: what is the up-front investment before the business can sustain itself?

The second is to calculate the ROI for the above investment, but with a twist: calculate the average ROI at every time period, taking into account all periods before the period being calculated for, but ignoring all future periods. This will provide you with a very important result: you will see that the average ROI remains negative long after your business becomes profitable. You will also see in which time period your ROI reaches an acceptable level - a very important factor when risk is considered.

Yes, it's just a model driven by assumptions. But with the above metrics, you can play with the assumptions (worst, best case etc.) and get meaningful results for decision making.

Disclaimer: The above explanation has been simplified quite a bit. Liberties have been taken with definitions. All your base are belong to us.

andrew_null4• 18 years ago

I am the author of this blog post - good points on this from a financial perspective. I can tell you were building financial models for a living, because you think ROI is the most important question for entrepreneurs ;-)

For startups, the most important things are good products and good markets. Successful angels and VCs know that, and will invest based on product and growth, NOT financial metrics. Thus, in very few VC pitches will you ever see stats on ROI or payback rate, that's more something you'd seen in the accounting or I-banking world. (I know this from personal experience, having spent the last year at a silicon valley based VC and seeing dozens of pitches)

So if entrepreneurs are primarily focused in the world of products, features, etc., then the key thing is to figure out the metrics that measure how these products tie to external market value. So things like conversion rates in funnels or cost per acquisition become super important, because the features are the steering wheel to your revenue engine. As a result, the spreadsheet model is mostly focused on things that are granular enough to relate to product and functionality, rather than things like valuation multiples or ROI or other overly-broad financial metrics.

daveungerer3• 18 years ago

If financial metrics are not important for entrepreneurs, and if the most important things are good products and good markets, then why did you write a whole article on creating a PROFITABLE freemium startup? Do you think profit is not a financial metric?

Of course it is, just not as useful in this situation. Hence my suggestion to use metrics that are more meaningful. The metrics that are useful should show you what the risk / reward is for pursuing the business.

It is very important for any entrepreneur to estimate how much investment is required to cover initial operating losses and capital expenditure. It is just as important for them to determine whether this investment is worth the expected future profits.

Just to drive the point home: you can spend 1 billion dollars in year 1 on advertising, if you had the money. And year 2 will be profitable. But you have a terrible business if the profit curve over the next years isn't steep enough to offset the original investment. Simply making profits is not enough.

DenisM• 18 years ago

Maybe you could build a correct model with explanations? That could easily be single most valuable article on HN for the whole year.

daveungerer3• 18 years ago

That's very tempting. Could be a very good catalyst for an idea I've had:

Been considering creating a web 2.0 type financial modeling site that wizards you through creating various types of models and allows you to collaborate with your colleagues.

Important features off the top of my head: * Export to excel * Ability for others to plug in alternative assumptions to your data (i.e. different scenarios) * Generates a deck of powerpoint slides. * Once the wizard has been completed, models will be edited graphically as far as possible, with the minimum of formulae.

Anyone interested in collaborating?

Anyone know of other websites that make it as easy as possible for a layman to do financial models without resorting to tedious work in Excel?

sayrer• 18 years ago

so, this guy who founded an ad network is telling you to spend 35k on ads when you first start your business.

who voted this up? do you people even know what Y Combinator does?

andrew_null4• 18 years ago

I'm not suggesting that.

You should run small experiments (<$50) per day until you figure out that you get can profitable. Only then do you scale up. The point is understanding what the microeconomics look like, so that you know how to grow the business over time profitably.

If you can go viral, that's even better obviously. I've written extensively on that also - and it's my preferred approach. But it's just true that for most subscription products, it's easier to buy the users than to acquire them virally.

rrhyne• 18 years ago

His post is very valuable for those apps where value is not derived from how many friends you have using the service with you.

How to create a profitable Freemium startup (spreadsheet model included)

Related Stories