is SaaS such a great business model?

I have to admit that I love SaaS (Software as a Service), both from a user perspective and from a supplier perspective., and those are the main reasons why:

SaaS is a new way (granted it’s well over 10 years old but new in most people’s mind) to deliver application over the web. Instead of downloading or buying an application and installing it on your own machine or server, you simply connect to a web site and use the application online. What’s the benefit as a user:

Instant access, no install cost. Using a SaaS solution is basically a matter of creating an account. When you use your gmail or Yahoo mail, it’s email using the SaaS model.. You don’t need to buy a machine.

Pay as you go.. Pricing is usually based on a subscription model: If you use the application (assuming it’s not free), you pay a monthly fee, but if you want to stop using it, just stop paying. The investment required is definitely lower than paying an upfront fee for “On-premise” software.

No Maintenance. when you have a software installed in your enterprise, you have to handle it’s maintenance. Upgrade, problems, extension is needed, etc.. With the SaaS model, your supplier takes care of all of it, including Backups. Upgrade are done seamlessly, on a very frequent basis, bugs are solved faster (no need for a big release), storage is usually expandable at will. All this for your monthly subscription.

One version to deal with. If any problem occurs on the application and you have to deal with your suppliers hotline, you won’t hear the usual: “Upgrade to the latest version and call us back”. You are always using the latest version and support is therefore simplified.

From a Supplier stand point the benefits are also great:

Easy to do trials with customers. As the start up cost is close to zero, it’s easy to push your potential customers to try out your software and hopefully convince them that it works for them and that they should subscribe.

The Subscription model allows for recurring incomes that are more predictable and that add up over time. Every new customer will bring an additional revenue to your business which, over time can become very significant and less dependent on sales effort.

Only one version to support. If you think that companies like SAP can maintain something like 10 versions of the same software, imagine the  hassle (and cost) of development, upgrade, testing (Quality Assurance – QA), and training. Dealing with only one version is fundamentally less costly for a supplier.

Easy to up-sale customers with add-ons. Want extra storage, want a Gold Access to certain features? the user can be one click away from those up-sales by making it very convenient for them to buy. SaaS makes it very easy to come up with version pricing, bundle pricing, and get a little closer to the economist dream of one price per user..

I could go on and on about those benefits and the growth of the SaaS market as a whole has been impressive as a consequence. Gartner projected that  “by 2011, 25 percent of new business software will be delivered as SaaS” and IDC projected that “by 2011, this opportunity will reach $14.8 billion, representing a compound annual growth rate (CAGR) of 32%”. Fast growing market (one of the key point for a successful business) so why isn’t everyone taking that route then? What is hidden behind the scene?

People talk about as one of the biggest company to have achieved a sustainable revenue/profit with SaaS. makes almost $1B a year but if we take a closer look at another company with similar size (at least in terms of market cap): BMC, what do we see?

Some rough numbers first:

SalesForce BMC
Market Cap $8.01B $6.84B
EBITDA $145M $633M
Revenue $1.24B $1.8B

If you look at all the benefits from SaaS, a SaaS company should have way larger margins than any other non SaaS company, but the numbers say otherwise:

SalesForce has a profit margin of 5.96% where BMC shoots at 18.29%, the operaintg margin of Salesforce is 8.53% where BMC flies at 27.62%..  (all data come from Yahoo Finance). BMC has been on the market for a lot longer, so the learning curve for Salesforce might not have allowed them to reach an optimum position yet.

Of course I’m not a financial analyst (call to all my friends who are in the finance sector for a quick analysis that I can add here!!) but there is something wrong in this picture.

Here are a few guesses that would surface in their 10K:

Spending on the sales force itself and on marketing effort must be huge at Salesforce. Despite all the benefits of SaaS large company who can bring large revenue with smaller sales force, are still convinced that (for security/control reasons) they are better off having everything internally. There is a huge need to convince decision maker that SaaS IS safe and that you can trust a third party to host your data, secure it and maintain it, better than you can do yourself and for cheaper.

The SaaS market is still very attractive to Small and Medium businesses too, and reaching those companies requires large sales force (or going through resellers), which is costly..

Another explanation is that Saleforce might not follow the SaaS model as much as it wants. What if trying to gain new customers (especially big ones) they break their own rules and start building custom solutions (different versions that need to be maintained separately..). I wouldn’t be surprised to see that happening.. but doing so on a true SaaS architecture makes it a nightmare (and expensive) to maintain and grow.

The SGA of both company could backup those guesses:

Sales Force SGA: 693M (growing steadily with revenue)

BMC SGA: 739M (stable for the last 3 years with a $400M increase in revenue)

When I look at this more closely I realize that SaaS might not be the major success people think int he coming year. It is a good business model that’s for sure, and startups should definitely consider it, but let’s not take things for granted and think that building a SaaS model over an On-Premise model is the recipe for success. I would go even farther than that and say that a company structured to build on-premise solutions can more seasily build a SaaS offer and maitain it as an extra version that the other way around.

Finance people, I am turning to you for comments.. what’s hidden behind those numbers? is SaaS such a great business model?


12 thoughts on “is SaaS such a great business model?

  1. Hey Antony,

    Note that the accounting for SaaS revenue treats it as an ongoing service, so even though SF sells a big deal to a customer, they only recognize the revenue slowly over the next 12-24 months as the customer uses the product. SF had another $113M in deferred revenue that doesn’t show up on the income statement because of GAAP rules. I suspect this revenue lag leads to a bit of the variance you’re looking at here.



    • Thanks Adam for the clarification on that front, it is important to note. I looked up BMC for that and wondered if I read well when I saw that BMC had $1.7B in deferred revenue? I know they have to account for revenue as their software implementation goes along and not only on the total deals they sign (license + implementation) which might explain that very important number..

      Also, I have to admit that sustainability in a downturn would be better because of this model.. If you don’t renew or upgrade your On-Premise solution you can still use it and that might be a budget that gets cut pretty quickly (and one of the big issues of SAP for example), whereas a SaaS enterprise like SF still gets revenue because people do still need to use their tool. The deferred revenue for SF took a pretty nice hit lately and I wonder if we are seeing the consequence of 2008 not being as good as before..

  2. good post Antony. I’m not a finance person, but I agree growing the monthly recurring revenue base is the key driver of success for a subscription business model, so looking at past financial performance can be deceptive if you’re trying to examine future growth potential. I think it’s better to focus on booked revenue and marketing expense from previous quarter (

    Alot of these SaaS companies that sell into the enterprise rely heavily on sales people to sell into these companies, which erode margins. I imagine over time as Salesforce gets its foot in the door of more companies, it can rely less on human sales force and more on its in-product marketing infrastructure to generate revenue.

    • Hawro,
      thanks for the link, the Magic Number is pretty cool. I think I’ll play with it a bit. That can also be a good thing to incorporate in a Business plan forecast, especially if you end up talking to Lars Leckie at Hummer Winblad. I’ve met Lars several times and he knows his business.. good to have some thoughts from him.

  3. Pingback: Tweets that mention is SaaS such a great business model? « Entrepreneurship Talk --

  4. a couple of thoughts, not necessarily well structured but hopefully that will clarify some of your questions.

    Revenue recognition / deferred revenue. The general contract model for companies like BMC is usually multi year with multi products bundled together. The contracts usually include license, maintenance and sometimes professional services. Upon contract expiration, maintenance gets renewed. GAAP has very specific rules around revenue recognition. The short summary is that you can’t recognize your revenue until the product or service has been delivered to the customer. Assuming the company has VSOE in place ( has a good explanation how it works), it can then recognize individual elements as revenue when they are provided. The deferred revenue piece is the piece sold (“booked” in the finance world) but not delivered yet (i.e. maintenance for years 2 and 3 for instance). That’s why companies like BMC or CA have huge amounts on these DefRev accounts and the real instant performance metric to follow is actually not revenue, but rather bookings which reflect the actual amount sold (but not necessarily recognized as revenue or even billed as cash flow). Not all companies provide that guidance though.

    Stable revenue stream. Building on that, indeed, once you have booked a contract for 3 years and most of it is in deferred revenue, you get a nice visible revenue stream since you know what’s coming ahead. But i think this creates only a timing effect because eventually revenue will be impacted. On the other hand, your company doesn’t grow as much from a GAAP standpoint (since it takes 3 year to fully absorb the performance). Companies tend to trade on multiples, most of them being GAAP multiples (revenue, EBIT, EBITDA, P/E), with the exception of the FCF multiple. so one could argue that you “lose” some market value if you indeed trade on a revenue multiple and assuming your sales are growing, you get only 1/3 of the growth (assuming a 3 year contract) recognized that year.

    Profit margin. Indeed, SaaS companies tend to have lower profit margin. i know very little about saas businesses, but i attribute this to:
    1/ the cost of hardware + its maintenance. That’s a cost that is normally pushed onto the customer in an on premise model. I don’t have figures to prove my point but maintaining a data center is certainly costly and SaaS model companies probably carry that cost.
    2/ indeed, per your point above, i don’t think large companies (am talking the JPMs or Citis of the world) trust yet the SaaS model. Scalability (can you handle 70,000 users worldwide for each customer?) and security concerns are a huge hurdle to adopt this technology. I also believe the pricing structure is not appropriate for these large customers (there is moment where you actually gain economies of scale I think).
    3/ because large companies might not be quite ready yet (i think it’s a matter of a couple of years), targets are SMBs and it indeed requires a huge sales effort, hence S&M costs… (I am making this point based on intuition, i actually never looked at financials to back this up).
    4/ probably many more points i cannot think about right now…

    Companies like BMC on the other hand have minimal inventory cost (they sell software delivered by CD or internet I assume), target the largest 2000 customers worldwide (sales force cost is reduced per dollar of revenue I guess), have less infrastructure costs (only for development), probably less marketing costs as well. Also, i tend to believe that barriers to exit for customers are pretty high: these systems become so entangled in the customer’s IT that it becomes very expensive to kick them out. That certainly drives in part the high renewal rate for these companies. That’s certainly not a complete list, but i imagine these factors contribute to explain why the profits are higher at large software companies like BMC or CA who have a lot of renewals and deferred revenue.

    Now, does it make it a great business model? I guess it depends what you mean by great (in term of financial returns or in term of user experience?). 😉

    • After talking to Antonin I realized that I have not thanked him publicly about the tremendous effort he has put in his reply! THANK YOU Antonin for the time you have spent to clarify my analysis. This is great stuff!

  5. Pingback: Is SaaS such a great business model? – 19 months later. « Entrepreneurship Talk

  6. What’s Taking place i’m new to this, I stumbled upon this I’ve found It positively helpful and it has helped me out loads. I’m hoping to give a contribution & aid other
    customers like its helped me. Great job.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s