Friday, September 30, 2011

Disrupt Yourself Before Others Disrupt You: DVD To Streaming Transition Is Same As On-Premise To Cloud


Recently, Netflix separated their streaming and DVD subscription plans. As per Netflix's forecast, they will lose about 1 million subscribers by the end of this quarter. The customers did not like what Netflix did. A few days back, Netflix's CEO, Reed Hastings, wrote a blog post explaining why Netflix separated their plans. He also announced their new brand, Qwikster, which will be a separate DVD service from Netflix's streaming website. These two services won't share the queues and movie recommendations even if you subscribe to both of them. A lot has been said and discussed about how poorly Netlflix communicated the overall situation and made wrong decisions.

I have no insider information about these decisions. They might seem wrong in short term but I am on Netflix's side and agree with the co-founder Marc Randolph that Netflix didn't screw up. I believe it was the right thing to do, but they could have executed it a little better. Not only I am on their side, but I see parallels between Netflix's transition from DVD to steaming and on-premise enterprise ISVs' transition from on-premise to cloud. The on-premise ISVs don't want to cannibalize their existing on-premise business to move to the cloud even if they know that's the future, but they don't want to wait long enough to be in a situation where they run out of money and become irrelevant before the transition.

So, what can these on-premise ISV's learn from Netflix's decisions and mistakes?

Run it as a separate business unit, compete in the right category, and manage street's expectations:

Most companies run their business as single P&L and that's how the street sees it and expects certain revenue and margins. Single P&L muddies the water.The companies have no way of knowing how much money they are spending on a specific business and how much revenue it brings in. In many cases, there is not even an internal separation between different business units. Setting up a separate business unit is a first step to get the accounting practices right including tracking cost and giving the right guidance to the street. DVD business is like maintenance revenue and the streaming is like license revenue. The investors want to know two things: you're still a growth company (streaming) and you still have enough cash coming in (DVD business) to tap into the potential to grow.

Netflix faces competition in streaming as well as in their DVD business, but the nature of competition is quite different. For the enterprise ISVs competing with on-premise vendors is quite different than competing with SaaS vendors. The nature of business — cost structure, revenue streams, ecosystem, platform, anti-trust issues, marketing campaigns, sales strategy — is so different that you almost need a separate organization.

Prepare yourself to acquire and be acquired:

Netflix could potentially acquire a vendor in the streaming business or in the DVD business and that makes it easy for them to integrate. This is even more true in the case of ISVs since most of the on-premise ISVs will grow into the cloud through acquisitions. If you're running your SaaS business as a separate entity, it is much easier to integrate the new business from technology as well as business perspective.

Just as you could acquire companies, you should prepare yourself for an exit as well. Netflix could potentially sell the DVD unit to someone else. This will be a difficult transaction if their streaming business is intertwined with their DVD business. The same is true for the enterprise ISVs. One day, they might decide to sell their existing on-premise business. Running it as a separate business entity makes it much easier to attract a buyer and sell it as a clean transaction.

Take your customers through the journey: 

This is where Netflix failed. They did not communicate to the customers early on and ended up designing a service that doesn't leverage existing participation of the customers such as recommendations and queues. There is no logical reason why they cannot have a contract in place between two business units to exchange data, even if these two units are essentially separate business entities. The ISVs should not make this mistake. When you move to the cloud, make sure that your customers can connect to their on-premise systems. Not only that, you need to take care of their current contracts and extend them to the cloud if possible and make it easy for them to transition. Don't make it painful for your customers. The whole should be great than the sum of its parts.

Run your business as a global brand:

Learn from P&G and GE. They are companies made up of companies. They do run these sub-companies independently with a function to manage them across. It does work. Netflix has a great brand and they will retain that. As an on-premise ISV you should consider running your on-premise and cloud businesses as sub-brands under single brand umbrella. Branding is the opposite of financials; brand is a perception and financials is a reality. Customers care for the brand and service and the street cares for the financials. They seem to be very closely related to each other for a company looking inside-in but from an outside-in perspective they are quite different. There is indeed a way to please them both. This is where the most companies make wrong decisions.

Wednesday, September 7, 2011

Freemium Is The New Piracy In The SaaS World

It is estimated that approximately 41% of revenue, close to $53 billion, is "lost" in software piracy. This number is totally misleading since it assumes that all the people who knowingly or unknowingly pirated software would have bought the software at the published price had they not pirated it. RIAA also applies the same nonsense logic to blow the music piracy number way out of proportion. The most people who pirate software are similar to the people who pirate music. They may not necessarily buy software at all. If they can't pirate your software, they will pirate something else. If they can't do that, they will find some other alternative to get the job done.

Fortunately, some software companies understand this very well and they have a two-pronged approach to deal with this situation: prevent large scale piracy and leverage piracy when you can't prevent it. If an individual has access to free (pirated) software, as a vendor, you're essentially encouraging an organic ecosystem. The person who pirated your software is more likely to make a recommendation to continue using it when he/she is employed by a company that cannot and will not pirate. This model has worked extremely well. What has not been working so well and what the most on-premise vendors struggle with is the unintentional license usage or revenue leakage. Customers buy on-premise software through channels and deploy to large number of users. Most on-premise software are not instrumented to prevent unintentional license usage. The license activation, monitoring, and compliance systems are antiquated in most cases and cannot deal with this problem. This is very different than piracy because the most corporations, at least in the western world, that deploy the on-premise software want to be honest but they have no easy way to figure out how many licenses have beed used.

In the SaaS world, this problem goes away. The cloud becomes the platform to ensure that the subscriptions are paid for and monitored for continuous compliance. You could argue that there is no license leakage since there are no licenses to deal with. But, what about piracy? Well, there's no piracy either. This is a bad thing. Even though a try before buy exists, there's no organic grass-roots adoption of your software (as a service) since people can't pirate. In many countries where software piracy is rampant, the internet access is not ubiquitous and bandwidth is still limited. This creates one more hurdle for the people to use your software.

So, what does this mean to you?

SaaS ISV: It is very important for you to have a freemium model that is country-specific and not just a vanilla try-before-buy. You need to get users start using your service for free early on and make it difficult for them to move away when they work for someone who can pay you. Even though you're a SaaS company, consider a free on-premise version that provides significant value. Evernote is a great example of this strategy. It shouldn't surprise you that people still download software, pirated or otherwise. Don't try to change their behavior, instead make your business model fit to their needs. As these users become more connected and the economics work in their favor, they will buy your service. It's also important to understand that the countries where piracy is rampant, people are extremely value conscious.

On-premise ISV: Don't lose your sleep over piracy. It's not an easy problem to solve but do make sure that you're doing all you can to prevent it. Consider a freemium business model where you're providing a clean and free version to your users. If the users can get enough basic value from a free version, they are less likely to pirate a paid version. What you absolutely must do is to fix your license management systems to prevent unintentional license usage. Help yourself by helping your customers who want to be honest. The cloud is a great platform to collect, clean, and match all the license usage data. You have a little or no control over customers' landscapes but you do have control over your own system in the cloud as long as there's a little instrumentation embedded in your on-premise software and a hybrid architecture that connects your on-premise software to the cloud. In nutshell you should be able to manage your licenses the way SaaS companies manage their subscriptions. There are plenty of other benefits of this approach including the most important benefit being a SaaS repository of your customers and their landscapes. This would help you better integrate your future SaaS offerings and acquisitions as well as third-part tools that you might use to run your business.