Wallen's

The cost drivers impact on so-called freemium models

There is quite some discussion on the freemium business models these last days with Jott and Sprout killing their free services. Charles, Mark and Shai have very good posts on the topic. I had written a post a few months ago on freemium where I said that there were 4 key conditions for this model to be viable:

• A "free" service with real value to build a user base
• A good balance between "free" features and "paid" features
• Well-thought price points that match the willingness-to-pay of certain segment of the user base
• Good marketing tactics to encourage conversation from "free" to "paid"

These conditions where looking at the revenue side of the equation - not the cost side. But it is key to distinguish between two type of cost situations with “freemium” models:

1. Marginal costs of serving “free" customers is close zero (or is offset by revenues directly attributable to the “free” users, e.g., ad)
2. Marginal cost are significantly above zero

In the first situation, freemium is a "real" business model in the sense that it is an "endogenous" system that can be sustainably profitable with limited investments (provided that the 4 above conditions hold of course). This model is particularly attractive for services where free users provide indirect value to the overall product (users network effects).

In the second situation, freemium becomes an operational marketing technique: it is an investment  to acquire new users of paid services. It is actually similar to the traditional free sample/trial commonly used in numerous industries in particular the consumer goods sector. I am not judging whether it's a good or bad marketing technique. Actually, as long as you have cash and that the customer acquisition cost is below the expected revenues, it can be profitable. However, a company cannot build its entire revenue model on one single marketing technique. Free sample/trial is part of a portfolio of available marketing techniques. Various techniques maybe more suitable at different times depending on the situation. For example, Jott services necessitates the transcription of oral messages into text requiring human labor (which I translate into significant marginal cost). With hindsight, in a way, Jott executed a 2 year “free trial campaign” to acquire new customers.

In other words, when marginal cost are significantly above zero, the business model is not "freemium" - it is "subscription/payment for a service"... just as in 99% of the economy. It is vital for start-ups to keep this in mind because the way a company defines itself impacts the solutions it looks for. And in the current environment, solutions they will need. You can bet that we will see more and more companies killing their free services in the coming months.

Advertsing prices relatively stable in Q4 2008

PubMatic just released its Q4 2008 figures for eCPM advertising pricing. You can find the full report here. While it is not a fully representative benchmark, this report seems to be directionnaly correct. The main message I take out is that Q4 prices are relatively stable vs. Q3 (-3.7%, i.e., within the margin of error of such a benchmark). This is a good suprise given the economic environment but it could be due to seasonal effect (i.e., holiday advertising). There is in any case a continuous downward trend of ad pricing when we look at the figures since Q4 07.

Technology conferences - worth the cost?

Healy Jones just published a blog post building on a comment I wrote in his previous post. The topic is the "ROI" of technology conferences. The post is an interesting read to get a VC perspective on the fit of conferences for a startup financing strategy. The key messages from Healy for entrepreneurs are:

• Preparation is key
• If at pitch stage, entrepreneurs can line up informal meeting with VCs (while it should not be the sole goal)
• If at due dilligence stage, arrange meetings between the VC and customers or other critical partners/players in your ecosystem
• Leverage your (potential) VC to get introductions to potential customers or key players

Healy's post was voluntarily focused on the fit with a VC financing strategy. There are obviously other potential benefits. My feeling is that the potential main value-add of a conference for a web startup is to increase your overall visibility/awarness ("anchoring yourself in your ecosystem") rather than directly impacting new users growth or pitching a VC. Here are a few examples of how to do it:

• Line up before hand a series of meeting with players in your field
• Organize or sponsor an "unofficial" event related to the conference
• Buy a booth to showcase your product (sort of expensive)
• If you are blogger, try to get a blogger pass to "officially" cover the conference either for your own blog or for a bigger media/blog
• Participate to the startup competition
• Organize a workshop on a secondary stage on the problem you are adressing with your product (if conference permits)
• Become one of the sponsors of the conference (sort of expensive)
• Be on the plenary stage... which is sort of difficult for early stage startup, except if you are among the startup competition finalists

There is probably a lot of other ways but there is no silver bullets. It's the combination of several actions that really increases the ROI of a conference. But it requires a lot of preparation for an uncertain return (especially for early stage startups).

What's your experience?

Update: there is post from Futuristic Play with an extensive list of tech conferences here

2007 Web start-up funding (2/2): Investment size and implied valuation

I said in my last post that I would make a series of 3-4 posts before year end on 2007 Web start-up funding from CrunchBase database. Well... time flies, so that's the second (and last) post of this series. It focuses on Web start-up exclusively on investment size by stage and rough estimates of implied valuation.

Web investment size

Series A, B and C & above represented in 2007 respectively 55%, 31% and 14% in terms of number of investments and 36%, 33% and 31% in dollar terms. In 2007, the median round size was $~4m for series A, $~10m for series B, $~15m for series C & above. These figures are similar to the 2004-2008 figures which is normal given the 2007-2008 bias in CrunchBase. 

There is still a large variance. For example, there is a ratio of 3.2 between first and third quartile  series A ($2.1m vs. $ 6.5) with 10th percentile investment size going up to $10.5m. Series B and C & above have a lower variance (ratio of 2.3 and 2.5 between quartile 1 and 3). Series B remained in the range of $6-15m and series C $10-25m.

  

Implied Web pre-money valuation

By doing some assumptions (series A tend to go for 25-50% of capital, series B for 20-30%, series C & above for 10-20%), we can infer some rough estimates of pre-money valuations that should be directionally correct for 2007. The figures are summarized in the table below.

Series A median pre-money valuation is around $8m and the average around $12m. Series B estimates are around $30-35m and series C & above are around $95-115m. However, these figures are just indications not a recipe to follow. For example, top series A pre-money valuation can be >$~20m (10th percentile).

 

 

It would be interesting to compare these 2007 figures to 2008 ones. I'll try in the coming month to repeat this series for 2008 Web start-up funding.