If you think objectively about pricing trends in enterprise software you can easily come to the conclusion that almost all software should be free or nearly free. I’ll sidestep the entire topic of open source because while that segment is certainly a participant, I don’t believe it’s the main driver. Also please note that this post is pretty rough in terms of having thought it all through… in other words, I’m thinking out loud here.
Let’s look at a couple of realities in enterprise software, first and foremost being that license sales are simply a proxy for the future annuity that a maintenance base is. If you could offer core application functionality that is competitive neutral as free, not only would you go a long way to eliminating a major obstacle to building a large installed base – the actual sales cycle – but you would also be able to reconfigure your entire strategy for going to market in the first place.
Enterprise software selling is a nasty and inefficient process that favors the large vendors at the expense of the rest of the market (one reason why Oracle bought all those companies, to become the largest “share of IT wallet” in their customer base). Current enterprise sales processes don’t facilitate selling across the entire market spectrum and that’s the only reason software companies ever had partners selling for them in the low end of the market, they couldn’t afford to do it direct. Salesforce.com is turning that on it’s ear but even they have some challenges ahead of them as they grow up in enterprise accounts, and in the end they may end up looking a lot like Oracle and SAP rather than different.
If you could dramatically lower the cost of sales while horizontally scaling the customer base you could, ideally I admit, transform a hunter/killer sales force into a farming operation where the objective is the upselling of components and maximum penetration of existing accounts. In just such a scenario the sales force becomes an extension of customer support because what they are monetizing is the support part of the deal and that’s just long since moved away from the promise of upgrades. Customer advocacy in these companies could take a dramatic turn for the better if the software license component was removed, but there is the risk that it could worsen because the investment a single customer is making is considerably less and therefore the vendor may in fact re-calibrate what the expectation of satisfaction is. Another scenario is that the support side of the business would be subject to a new range of competitors that could price cannibalize the entire business… it’s a possibility but I’m not going to pretend that I can predict every unintended consequence.
The other aspect of enterprise software sales is that customers are already conditioned to expect a huge discount on the list price for the software, so going to zero, or close to it, synchronizes the actual price list to the official one… it’s like acknowledging that nobody pays MSRP so why bother having it.
There are considerable obstacles to this concept, the most direct being that you are effectively cannibalizing billions of license revenue for the prospect of rewards in the lower margin services business. Not an easy decision, I admit, but if you accept the premise that license revenue is a dead man walking then it would be irresponsible to not anticipate how to structure your business when it’s gone because of market conditions. Having said that I am also forced to admit that taking out a few billion in license revenue, even if the unit economics are improving, creates a major problem that is only fixed through dramatically upsizing the support and services side of the business, which being more people intensive could end up eroding the economics (I guess this could the call “the Dell dilemma”).
The other obstacles are tactical, such as how to compensate a sales force and report financial results. Largely solvable but hugely disruptive. Another interesting challenge is how to price support/maintenance when you don’t have the license cost denominator to work with, but again this is an opportunity for a new and creative approach to support pricing.
Lastly, not all your software should be free, just the what we would call the “core” which in our case is ERP. The segment of software products that do deserve the traditional pricing models are the “context” pieces, such as industry extensions and truly proprietary components, including analytics. I’m hesitant to even suggest this, but in many ways this is like the bubble-era tactics for “buying eyeballs” but in this case there is a rationalization that is rooted in the true economics of our market, not just the hypotheticals of a B-school case study.