This is the final market we look at in our Creative Destruction 7 Act Play series.

Enterprise software is a huge market - $222.6 billion in 2009 according to Gartner. Put that in perspective. The online advertising market in the USA is only $50 billion. Even assuming the global online ad market is 2x the US market, enterprise software is 2x online advertising.

That $50 billion online ad market has come from nowhere in the 15 years since the start of the web. That is why online advertising has been the fun growth market in the last decade.

And most VCs have shunned the enterprise market in that same decade. They have had two reasons. One is the perception that the enterprise market is locked up by a few giant firms such as SAP, Oracle and IBM; actually they dominate their market far, far less than Google dominates online advertising. The other is that the cost of sale is too high; that remains true in most cases.

That lack of VC attention is a blessing for entrepreneurs! The market is less crowded. And customers do want innovation. They do not want to be reliant on a few large vendors; they know that these vendors will exploit that position at the customer's expense.

Even a small slice of the enterprise software market is big. Most VC want an addressable market that is $500m. There should be plenty of $500m niches within that $229 billion! And a bootstrapped entrepreneur does not even have to get that big a market; they could get a very healthy $10m business in a tiny $50m market.

And enterprise softwate is a market where ventures have historically not needed a lot of capital. Most of the current giants were bootstrapped.

So, at the macro level, enterprise software is a good place to be. But start-ups don't live or die at the macro level. They live or die by having a stunningly strong value proposition to overcome the corporate risk aversion. Think 10x. You have to be better, faster, cheaper by 10x orders of magnitude. That is the only way to get cost of sales to a reasonable level - or even to get to the point where you worry about cost of sales!

The question is, can Semantic Web technology have a 10x scale impact on enterprise software?

Image courtesy Flickr and BirdOfTheGalaxy.

The Big Change: Users Are Finally In Charge

Search for "enterprise software" on Google and 2nd link (right after Wikipedia) is a 2007 post by Jason Fried at 37 Signals (Basecamp) entitled Why Enterprise Software Sucks.

"The people who buy enterprise software aren’t the people who use enterprise software. That’s where the disconnect begins. And it pulls and pulls and pulls until the user experience is split from the buying experience so severely that the software vendors are building for the buyers, not the users. The experience takes a back seat to the feature list, future promises, and buzz words."

User experience is a driver for SaaS. You don't need IT to approve the use of a product. Start with a free trial and pay for a few users on your credit card and expense it.

The Semantic Web does not make software more user friendly. In fact most semantic web user experience is pretty awful; great for geeks, lousy for the rest of the planet.

This is a big problem for enterprise semantic web vendors. Too many vendors sound like they are pitching to IT. They talk the language of IT and yet claim technology that will make traditional IT irrelevant. Turkeys don't vote for Thanksgiving. It is unlikely that IT will lobby hard to get budgets that overturn the technology that they lobbied so hard for in the last budget round.

The Current Growth & Buzz Around Collaboration

The growth pocket in enterprise software has been around collaboration, sometimes known as "Enterprise 2.0". This grew out of earlier markets such as Intranets. One piece of this, Online Community Management, which is very new is on track to be $1.6 billion by 2013 according to this analyst.

The broader market of "content, communications and collaboration (CCC)" is pegged by Gartner at $2.6 billion in 2009 (compare that with CRM at $2.3 billion).

Unfortunately that is not a market that currently uses much semantic technology.

That is two strikes against semantic web technology in the enterprise. It is not part of the user driven SaaS wave and it is not part of the Enterprise 2.0 online collaboration growth market.

The ERP Reality Check

Despite all the buzz around collaboration, ERP is still what pays the bills. ERP accounts for $67 billion of that $229bn vs $2.3bn for CRM and $2.6bn for CCC.

Why is ERP still so big? What does this tell us about future opportunities?

Its The Integration, Stupid!

If you spend any time selling a nice shiny new SaaS solution within an enterprise, after winning over a bunch of end users, you end up in a meeting where somebody says:

"OK, but how does it integrate with...."

Enterprises are all about integration. It is the defining characteristic of an enterprise. Without integration, an individual or small partnership could do the same thing. Integrating multiple departments, profit centers, divisions and country operations is what makes enterprises big.

That is what drove the now $67 billion ERP market. That ability to integrate all the key processes was what made SAP a winner and other ERP champions followed suit.

This is primarily driven by the growing needs of the enterprises worldwide to integrate all the back office functions in order to achieve control of information flow.

Integration. Now THAT looks like an opportunity that is ripe for semantic web. Data integration is what we are all about. But this is hardly a new challenge. Lots of smart folks have been working on the data integration challenge for a long, long time.

The Evolution Of Enterprise Integration

Cheerfully glossing over a few critical details, here is my summary of the last few decades in enterprise software:

Phase 1. The ERP Era. Integration has to be at the level of business process. This is basically the ERP mantra. It is horribly complex but, when it is implemented really well, the result is a much more efficient enterprise. But ERP fits a world where change is slow and predictable; ERP is not so good when business changes too fast for the IT folk to change the ERP systems to meet the new business reality.

Phase 2. The Data Warehouse Era. Leave the processes where they are, just integrate the data. This worked well with acquisitions as it was quicker than integrating processes. But it tended to take a long time and the result often became "shelfware", so the ROI was much weaker than planned. It was too hard to forecast what data people would want and how they would use it.

Phase 3. The Message Bus/SOA Era. Leave the data where it is, just send messages around and let loosely coupled adapters figure out how to convert the messages into useful data. This is the current preferred approach. It is better, but it still does not enable the on the fly access to data that business users have come to expect online.

Phase 4. The Search era. Just plug in a search appliance and tell users to enter their queries into the search bar. This is is the bar as far ease of adoption is concerned. Approaches from any of the earlier eras will be now judged against this ease of adoption. Search appliances require no custom integration project and no training for users. But as those of us in the semantic web world know, word search alone will not work in complex enterprise domains where lots of money is at stake.

So all the semantic web startups have to deliver is something that is as easy as Google but with the sophistication of the earlier eras. That is clearly easier said than done.

In the next post, we will look at how semantic web vendors can position within the enterprise.

-----------------------------------------------------------------------
• Don't forget to propose your startup for our Semantic Web Impact Awards. The deadline is Sept. 15.