Enterprise Software: Returning To A Market That Has Been In A Coma For A Decade September 5, 2012

About 10 years ago, I wrote a post called Enterprise Software R.I.P. The venue that I published in has long since disappeared into the digital dustbin, but my trusted laptop files retrieved it, so it is published below. I decided to revisit this post because I have come to the conclusion that enterprise software did not die, it just went into a coma and it is now coming out of that coma.

To a casual observer, there is not a lot of difference between coma and death. First let me say what I mean by coma/death in this context. Of course there is still lots of enterprise software and a few huge vendors doing very well and lots of small niche vendors operating in the cracks between those behemoths. But where is the Oracle or SAP of the last decade? Or even the BEA or Cognos or TIBCO of the last decade? Where is the start-up that broke into the mainstream and became a multi-$billion success story? I am referring to the death of innovation in enterprise software. This decade has not been conducive to enterprise software ventures. It is no wonder that most VCs ignored enterprise software during this decade.

Ten years ago I sensed that this was happening. It was disturbing to me because enterprise software was the world that I understood so well. I felt like the Polish Cavalry in the Second World War. The Polish Cavalry was renowned as the best in the world. They could shoot with great accuracy at full gallop and turn on a dime, nobody could match them. None of that helped them when the Nazi dive bombers and tanks rolled into Poland in 1939. I had mastered a game, but the game had changed.

So I set out to understand the consumer web. A decade later I am not a master of that game, but I understand it “enough to be dangerous”. That helps now that I am coming back to enterprise software because the consumerization of software is a big part of the renaissance of enterprise software. However it is only one part of that renaissance. Enterprises are more than the sum of their parts, they cannot simply empower every employee with consumer web type tools and hope they all pull together to grow the profits. That is why I remain sceptical of the hype around Enterprise 2.0 tools, all those “Facebook tools within the enterprise” ventures. These are “shiny objects” that make Gen Y employees happy and can have an incremental impact on productivity, but they are hardly game-changing in the way that say relational databases and ERP were game-changing in their day. Sure it helps to have better user interfaces that encourage collaboration. But there is a lot more at stake for enterprises and therefore for the employees. Enterprises face a perfect storm of three tsunamis hitting at the same time – Digitization, Globalization and the Debt Crisis. This is an existential crisis for large companies, their very reason for existence is being called into question. Business-as-usual won’t help them navigate this perfect storm. Therefore software-as-usual won’t help them navigate this perfect storm. That is why there is a huge opportunity again in enterprise software.

But I am “getting too far over my skis”. I want to return to the R.I.P post from ten years ago. There was a lot that I got right. I could see the dismal grind of consolidation. However some of my gloom was due to the terrible, but transitory, backwash from the end of the technology nuclear winter. These were the days when nobody was buying any software, innovation was dead, the only answer was to get into real estate speculation. Thankfully I resisted that temptation, it ended in a bust worse than the Dot Com bust, one that we are still living through. In hindsight the biggest thing I got wrong was:

“The “IP everywhere” rollout is exactly that, an implementation of proven technology by big vendors”.

That was wrong because the “IP everywhere rollout” fundamentally changes the rules of the game. The IP everywhere rollout makes transaction costs cheaper externally than internally. This is the practical realization of Coase’s Theorem. Coase, an economist writing in the 1930s, posited that firms grew big based on the fact that transaction costs were lower internally than externally. The Internet makes external transactions dramatically cheaper; this is the “frictionless commerce” that is now becoming reality. This challenges the very basic idea that scale is always an advantage. All those roll-ups, acquisitions and mergers in the industrial age were based on the theory of economies of scale; in Coase terms they were based on the theory that transaction costs were usually cheaper internally than externally. This goes to the very heart of what makes an enterprise big, why it needs to be big to win. This is as far from business-as-usual as you can get.

I also lost the plot a bit here as well:

“The return of the single vendor stack. IBM, Sun and HP are putting together complete solution stacks that look suspiciously like the pre-Wintel proprietary solution stacks provided by hardware vendors such as IBM, DEC, Data General, Olivetti, Burroughs, Univac, Wang and other dinosaurs that once ruled the earth. “

I had that both right and wrong. Yes, the vendor stacks forced consolidation. I got that bit right. But of course this is no different from the pre-Wintel proprietary solution stacks and they all disappeared into the dustbin of history; only IBM survived and thrived, DEC, Data General, Olivetti, Burroughs, Univac and Wang are all history. I saw that but I did not see how it would end. The answer is that when IP everywhere rolled out enough, the proprietary solution stacks started to become threatened. That is happening now. Sun has already fallen, it’s great technology is now one part of the Oracle stack. HP is a seriously troubled company. Again, only IBM has emerged stronger from this wave of change.

Perhaps the most fundamental mistake I made was in my definition:

“First a definition; enterprise software is the core, mission critical stuff that manages transactions, accounting and management information.”

If you define enterprise software that way then certainly it is “game over”, but that is only a definition of the first wave of enterprise software. The next wave, enabled by IP Everywhere will tackle much more critical issues than basic administrative functions. These critical issues will be the subject of anther post but they will address the existential question for enterprises which is how to grow when economies of scale is no longer the driver of growth.

I also mistook the fact that real time enterprise was only in the “slough of despond” that always comes after a period of hype when I wrote that:

“real time enterprise” is a fancy name for what the industry is gradually evolving towards

Real time enterprise needed to wait until the IP Everywhere rollout was more complete. For example, now that about 50% of the 7 billion people on the planet have mobile phones, you have to operate real time to thrive. The IP Everywhere rollout also enables real time enterprise solutions to be implemented practically.

—————————————-

Enterprise Software R.I.P

(Note: this was written late in 2002 and is copied here unchanged).

This is a receding Tsunami. Thousands of companies rode this one to fortune, but it is now crashing on the beach and the backwash is pulling a lot of companies underwater.

We are still in the early stages of the enterprise software consolidation and the most sensible option is to sell out for the best price you can get. Then you can find another wave that is growing. Or you can get out of the industry as thousands of talented, experienced executives have done in the last few years. For those who love the industry but hate the idea of working for one of the gorillas, this article highlights how to find a reasonably protected niche market

First a definition; enterprise software is the core, mission critical stuff that manages transactions, accounting and management information. The industry has been doing this for decades and there really are only so many ways you can slice the cake.

Of course it is a huge industry and is not going away. The issue is whether this is an environment conducive to start-ups. Look at the things that customers are now focused on such as data center consolidation and integration. These require big companies. The “IP everywhere” rollout is exactly that, an implementation of proven technology by big vendors.

Attempts to hustle up big new growth waves within enterprise software have failed. Wireless is a simple another delivery option and “real time enterprise” is a fancy name for what the industry is gradually evolving towards. These are add-ons to existing products from big companies.

What is driving this consolidation?

• The proximate cause is the after effects of the bubble bursting. Massive over-investment and the dramatic drop off in demand puts the buyer in control.

• The buyer has always hated the traditional enterprise software model; too many small vendors blaming each other for projects that don’t deliver business results. In a buyer’s market, they get what they have wanted for a long time.

• Investors demand the earnings visibility that comes from a recurring revenue model. When customers and investors both demand the same thing you can be pretty confident that it will happen.

• The return of the single vendor stack. IBM, Sun and HP are putting together complete solution stacks that look suspiciously like the pre-Wintel proprietary solution stacks provided by hardware vendors such as IBM, DEC, Data General, Olivetti, Burroughs, Univac, Wang and other dinosaurs that once ruled the earth.

• ASPs with solutions engineered from the ground up for the Net, such as SalesForce.Com and Intranets.com are getting real traction and proving that it is possible to deliver real solutions over the Net for a monthly fee.

So will all the customers simply plug into a few giant Con Edison style utilities? Is our only option to work for/invest in these utilities? Thankfully the answer is an emphatic no. The utility analogy can be stretched too far. IT has a far bigger impact on a company’s profitability than electricity and there are a lot more variables. So how can smaller independent companies prosper in this new world?

• Leverage the stack for your own high growth niche. Offer the total solution on-line for a monthly fee. This reduces the buyer’s risk and thus enables start-ups to get that critical early traction. The good news is that it is now much easier and cheaper to put together a total solution from a mix of outsourced data centers, open source frameworks and offshore developers. All you have to do is find an emerging growth market.

• Operate right at the top of the stack where you are dealing directly with end users (aka a vertical market solutions focus). Look for business sectors that are growing fast but that are small enough today to fall below the radar screen of the gorillas.

• Web Services based “features”. Experienced venture builders look at most new ventures and say, “that is not a product, it is simply a feature”. The best that can happen to these “companies” is that they get sold for R&D value. It is possible – but not yet proven – that Web Services will enable small companies to thrive by offering these features on a pay as you need basis over the Web.

• Mine the backwash. There is a lot of money in maintaining old systems, catering to the conservative customers and forgotten niche markets. These forgotten markets last much longer than you would think from listening to industry analysts. This is low on the glamour stakes but if you are in business to make money it is worth remembering the old saying “there’s brass in that old muck.”

• Private label commodity providers. This is another low glamour business, suitable for low cost operators. You sell your product through other solutions providers without your brand being visible. Of course you may eventually pull off the “Intel inside” trick and move up the stack, but even pure commodity players can make good money if the market is big enough and you focus on efficiency and being the lowest cost provider. You will need to bundle excellent support and show that your TCO is lower even than open source.

This a good time to take stock and get ready for the recovery with a new positioning. Markets will recover and IT will remain central to business. But don’t expect it to be like it was before. Prepare for dramatic change and find where you can add the most value.