An interesting post this morning made me think about the differences between EBIT models for traditional and SaaS businesses.

If we look at the expenditure side of the equation, we can see that SaaS tends to smooth expenditures for developers. Rather than creating a product in its entirety prior to earning any revenue off it (the traditional model), SaaS developers can launch a “bare bones” product (witness the Xero pre pre pre launch) and rapidly and incrementally develop their offering, all the while making some return on the product. Thus their development costs are naturally smoothed over a period of time rather than artificially amortized for accountancy’s sake. It seems a much more appealing prospect – from the expenditure side anyway.

If we look at revenue, a similar thing occurs here. In the traditional model, revenues are massive at launch (hopefully) and then very much peter out until such time has elapsed as to enable the business to charge an upgrade fee (and sure development times are being contracted but there is only so much business will take in terms of short term usability prior to requirement of upgrade). In the SaaS model however, revenue is (usually) incrementally earned on a monthly basis, making a much smoother and more stable model.

None of this is rocket science of course, but it would seem to make SaaS businesses a surer investment bet than traditional software ones – SaaS should tend to reduce volatility in EBIT which should in turn reduce volatility in share price.

The following graph is interesting – it comes from McKinsey and co (via this post) but it is, I believe, flawed in that the R&D shown for the traditional players is merely incremental upgrade R&D, their initial R&D to get the product to market has already been sunk. Therefore the true R&D cost to traditional business is, in real terms, higher than for SaaS players meaning that, once the SaaS players have matured, EBIT’s should be as good or better for a SaaS player than a traditional one. You’re also seeing some real early stage, high dollar selling costs that the incumbents don’t have so the analysis is flawed in favour of the traditional businesses.


Ben Kepes

Ben Kepes is a technology evangelist, an investor, a commentator and a business adviser. Ben covers the convergence of technology, mobile, ubiquity and agility, all enabled by the Cloud. His areas of interest extend to enterprise software, software integration, financial/accounting software, platforms and infrastructure as well as articulating technology simply for everyday users.

1 Comment
  • A couple of comments about this post and the MacKenzie chat in particular. I’ve commented at greater length on my blog (

    First, what I find most interesting is the comparison of the 2 bars on the right vs the one on the left. The lefthand bar is hugely skewed by the likes of Oracle, SAP, and Microsoft, who have unfair advantages of scale and monopoly. No SaaS companies have reached that stage, so I don’t find comparisons with the lefthand bar compelling yet. The primary takeawy is that the LBO/Merger trend is well justified because software companies below $1B are much less profitable than those above and debt is currently cheap.

    Second, if you dig a little deeper, you can see that the Sales and Marketing expense for SaaS companies is less than comparably sized perpetual companies. See my blog for the details, but its easy enough to pull the data from EDGAR filings. The conclusions one takes away are that SaaS is real and growing rapidly, and that SaaS CEO’s are making hay while the sun is shining by spending every penny to compound that growth.

    Whether from the standpoint that mass is worth exponentially more (the lesson of the lefthand bar) or that it is harder to grow than shrink software companies, that’s the right strategy.

    Do you find the comment that it is harder to grow than shrink odd? I don’t. Once you hit a critical mass that is fairly small, customers are loath to turn of the lights on software. It drags on for years of gradual decline with minimal effort. Computer Associates and Rational (now IBM) have made good money relying on this and it looks like Oracle has stepped up to take their place.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.