SmoothSpan Blog

For Executives, Entrepreneurs, and other Digerati who need to know about SaaS and Web 2.0.

Archive for September 25th, 2007

Interview: Concur’s CEO Steve Singh Speaks Out On SaaS/On-Demand

Posted by Bob Warfield on September 25, 2007

I had an opportunity to talk with Steve Singh, CEO of Concur Software about SaaS.  Steve’s company Concur is in the business of automating corporate travel and expenses through SaaS (or On-demand as Steve prefers) offerings his company provides.  Concur is also one of the few that’s successfully undergone a transition from being a conventional vendor selling On-premises to being a pure SaaS vendor.  Did I say succeessful?  Concur is now one of the powerhouse SaaS plays, second perhaps only to Salesforce.com.  Steve’s a real gentleman to talk to, extremely smart, and as you’ll see, he is passionate about what Concur does and about SaaS in general.  He provides a number of key insights that companies thinking about entering the SaaS world should consider.

Below is my transcript on the interview, with my parenthetical remarks (meaning my reaction but not something Steve and I talked about) in parenthesis.  Any mistakes are mine, and any great insights are Steve’s!

What’s your elevator pitch to customers about what Concur does?

Steve:  It’s an amazing opportunity, so good I have to pinch myself sometimes.  We bring innovation to a stagnant market by automating business processes around booking travel and getting expenses reimbursed.  This creates a more efficient supply chain around these areas.  This kind of processing is amazingly expensive.  It costs companies anywhere from $45 to $50 per trip to book travel without automation.  We bring that down to $5.  It costs $40 to $60 (let’s call it $48 on average) to get an expense report reimbursed and we can automate it to do that typically for under $8 a transaction.

In addition to the savings, we eliminate a ton of paper.  We pull data together for mining.  We help companies address governance and SOX issues.  Think about how big a portion of budget travel and expenses can be.  CEO and management have to sign of on the integrity of these transactions.  Concur spends something like 2 1/2% of expenses on Travel and Expense reimbursement.  That’s a big piece of our budget.  And, it’s very hard to govern a process without automating it.

How is your product sold?

Steve:  Our metric is transactions.  Think about it like a cell phone plan.  You get volume discounts with lots of minutes, and incremental minutes cost more.  A volume discount can be thought of as your free minutes.

Concur has 5,000 customers and we’ve sold about 28 million transactions.

Bob:  (This differs from companies like Salesforce.com that charge by the seat month.  It’s an interesting model to consider where the transactions are what the customer is thinking about.   This makes perfect sense for Concur, who want to align their charges with the customer’s perception of savings based on Steve’s economic figures on costs per transaction.)

What crystallized your decision to go SaaS?

Steve:  I would love to tell you there was an epiphany, but we just paid attention to our customers.  Our business and stock were doing well, but, there was a realization that our business served large customers well but nobody else.  This is true for all conventional software companies.  This limited our market and growth.  We wondered how best to solve that problem?

The only real example serving customers of any size and with an accountable 2-way relationship was On-demand.  Licensed software has no accountability.  Our examples were companies like ADP or Paychex.  Technologists think they invented everything, but On-demand has been around for decades.

Bob:  (It’s great that the proximity of ADP to the space helped give Concur an example to learn from.  Others have to stretch further, but now that SaaS is out there, everyone should “get it.”  Steve Singh is not the first SaaS executive I’ve talked to that brought up ADP.  Dave Thompson, former VP of Marketing for WebEx, told me their sales model was formalized around many small transactions because they hired their VP of Sales from ADP and he kept asking how to make WebEx more like ADP.)

Steve:  To make the transition involves a lot of pain.  Only some technology is common, everything else has to be reinvented.  The best example of what can happen if you get it right is to look at Concur vs Extensity.  In March of 2000, both were the same size.  During the downturn, Concur went On-demand and Extensity stayed On-premise.  Extensity was betting the downturn was the source of their problem, that it was just the economy and their growth would resume as soon as the economy turned around.  Fast forward to today and our guidance is almost $35M, Extensity is at $3M.

Bob:  (I wonder sometimes whether the conventional software model has ever really flourished except during bubbles of one kind or another.  Look at Big ERP.  They had the benefit of several major bubble waves:  the rush to switch to client server architectures, the Y2K debacle, and then right on the heels of Y2K was the Internet Bubble.  Would these businesses have grown nearly so large without any of those bubbles?  Maybe SaaS would’ve gotten here sooner.  The choice between the two is a Tortoise (SaaS) versus the Hare (Old School ISV) race.)

Steve:  It’s about driving a long term business, not short term success.  We knew we faced a lot of pain going SaaS, but we wanted to focus on the long-term.

Our first quarter of transition expanded new customers sales faster than ever before.  This was like getting instant feedback that our decision had been a good one, so I’ve never looked back.

What Were Your Biggest Fears About the Transition, And How Did They Turn Out?

Steve:  Our stock cratered and we wondered in the first 6 months.  We replaced all but 2 executives over 9 months.  The only 2 that stayed were the founder and myself.  Nobody else wanted the job.  We had massive changes in sales, support, prod development.  We created a new hosting organization.  We needed a new consulting organization that changed from traditional SI-style consulting to smaller deployments.  Essentially, we started a brand new business.

Our biggest fears:

  • We didn’t know if people would buy SaaS.  Even though customers bought payroll that way from ADP, they weren’t buying anything like what we had.  There was no proof they’d buy software as on-demand.  We did do customer research, but it was a bet.  Our first quarter told it all.  We couldn’t have asked for a better response.  This is why SaaS is doing so well.  It’s cost effective.  The vendor is accountable to the customer.  The customer’s life is easier.  And there is a much lower cost structure.
  • We also had a fear of whether our people could transition.  They had a lot of choices outside Concur.  It was the height of dot com.  Many wouldn’t take the pain.  We lost a lot of folks who wouldn’t do it.  Some were fine to lose and some it was very painful.

Bob:  (You can hear the pain in Steve’s voice when he talks about how hard it was during the early transition.   It’s awesome that they got instant positive feedback from the market, but you wonder what happens to companies that try a SaaS experiment without really committing and don’t get enough instant feedback?)

What surprised you as you moved into the SaaS market that you didn’t expect?

Steve:  We knew we’d do well in the middle market bringing enterprise class services that were unavailable to that market before.  But we didn’t think global accounts would embrace as quickly as they did.  American Express.  BofA.  Alcoa.  They came early.  They thought we had a much better solution than any other choice, including our own on-premises software. 

The other surprise was how things went with sales.  On-demand delivers reduced cash flow versus perpetual.  Initially, we tried to incent our sales orgs to tackle big accounts perpetually to offset the cash flow issue.  Eventually we made comp completely neutral.  Same commission no matter what the customer buys.  We let the customers decide whether to go SaaS or On-premises.  When the sales comp favored On-premises, it was a 50/50 split.  When we made it neutral it went to 99% On-demand versus 1% after comp change.

Bob: (Having been in the incentive compensation software business, I can’t tell you how powerful compensation is as a tool for aligning sales behaviour.  Steve Singh’s example points this out.  His salespeople where swaying almost half the customers to do something they didn’t really want to do because of the comp plan the sales people were on.  CEO’s take note: the next time you want to turn your supertanker company on a dime, think about using compensation to push that turn through faster!)

Do you think SaaS is an inevitable bridge that every ISV has to cross in some form or fashion?

Steve:  Absolutely.  Look at what’s happened in the last 30 years.  We went from mainframes and sky high transaction costs with few users.  Then we went minis.  Lower costs, more users.  Micro/PC.  Web-based On-demand  is the lowest cost structure and most users yet.  It’s inevitable continuation.

Bob:  (It’s a fascinating way of looking at it.  If Singh is right, there’s a lot more action in store for the SaaS world as we go forward!)

Any other advice for those who want to convert?

Steve:  This is just one guy’s prediction.  I don’t believe large companies can make the conversion.  Forget their genetic code.  How many will take the pain?  Companies won’t reinvent themselves. 

Think of taking a $40B company to On-demand.  The value of the business will go through huge negative change.  It will get crushed.  Cash flow will get crushed.  You have to layoff.  The transition is really hard and its very sudden.

If you’re north of $100M its hard.  Over $1B its impossible.

I’d like to compare businesses that have gone through fundamental shifts in the business model.  Very few have done it.  Intel is one.  But for every Intel there are 100 DECs.

You will see a next generation of leaders that don’t look anything like the last generation.  Keep your eye on the SaaS leaders.  Lots will happen there.

Bob:  (I think Steve is probably right, but we’ll see whether a few don’t decide to walk over the hot coals to SaaS anyway.  The amazing thing about SaaS is that it is better for customers and much worse for the incumbent non-SaaS ISV’s.  This is what makes it such a disruptive new business model.  It literally has the potential to completely change the landscape.)

Next?

Check out Part 2, where Steve gives his view on Salesforce, their Force platform, SAP’s ByDesign and acquisition strategy for SaaS companies,.  Part 3 discusses Sales and Marketing for SaaS companies as well as some observations Singh makes on future trends in the SaaS world.  Be sure to click the “subscribe” link at the top left of the blog page so  you don’t miss out on these future posts!

And special thanks to Steve Singh for taking so much time to talk with me for this interview.

Related Articles

To Seth Godin’s point, Steve Singh is the guy that said “follow me” at Concur and converted the company from Seth’s green curve to the blue curve.

Submit to Digg | Submit to Del.icio.us | Submit to StumbleUpon

Posted in business, saas, strategy | 18 Comments »

This Facebook Deal is Too Big! This Parakey Deal is Too Small! Maybe They’re Both Just Right…

Posted by Bob Warfield on September 25, 2007

Everyone needs to be careful what they wish for, lest they get it.  There’s a buzz in the blogosphere over two deals, one that many think is way too big and one that seems to raise the ire as being way too small.  Ironically, both involve Facebook.  You know things are a bit frothy when you can simultaneously generate many TechMeme hits around the same property, and when the hits are not that interesting.  Although what passes for popular gossip is often uninteresting.

Let’s start with the deal that was too small.  Michael Arrington tells us that Parakey’s investors got a bad deal because they sold the company to Facebook for chump change of “less than $4M”, and they didn’t even get shares, they got cash.  Meanwhile the founders of Parakey got “handsome stock options”.  Apparently the invested got a 2x return after 6 months.  I’ve no doubt the investors were disappointed at the time, but their disappointment at the present is completely irrelevant.

Why?  Play it back without the 20/20 hindsight.  The investors accepted a 2x return after 6 months.  That is well what they need to get to make their numbers.  Check out Will Price’s excellent post to understand why VC’s want 10x multiples.  Worse, they want 10x after a couple of years when it’s 10x on several rounds, not 10x after a few months on very little capital invested.  Now granted there were angels involved, but mighty Sequoia was also involved.  Let me tell you, those guys are smart.  They did not get taken advantage of here, at least not in their minds at the time the deal was struck.

That’s not to say Sequoia was wholly enamored of the deal, only the insiders know for sure whether:

– Sequoia had lost confidence/interest in Parakey.

– Parakey’s founders told their investors they were moving on to Facebook in any event.  The acquisition purely salvaged some value for the investors.

– Investors and Founders became convinced Facebook would crush Parakey if they didn’t give in to being acquired.  Convincing the acquiree of such an eventuality is very typical in tech M&A situations.

As for the founders getting “handsome stock options”, hello?  They got a job.  Jobs in Tech carry stock options.  Yes, those options appear to be worth an incredible fortune now that Microsoft is tickling the value of Facebook, but who knew back then.  The options still have to finish vesting, and Facebook has to generate a liquidity event before they’re worth anything. 

I’ve been involved in a lot of M&A activity in the software world, and deals like this are pretty common.  At Borland, we used to look for companies that had great technology but failed marketing.  The usual formula was to pay the investors similarly to Parakey and give the employees jobs.  Most of the time, everyone was happy about it because great technologies found a real marketing channel where they could finally succeed.

In short, it’s silly to kvetch about the Parakey deal.  This is why Wall Steet says we ought not to look back on stock trades.  We all have a list of those things we should have done.  So what?

The more interesting chess game is Microsoft’s rumored offer to invest in Facebook at an astounding valuation.  Apparently they’d like to acquire a 5% stake and are willing to give up circa $500M for it.  This churns up a lot of questions:  Why would Microsoft want this?  Why would Facebook want it?  Is Facebook worth $10B, let alone the $15B they’re rumored to be asking for as a counter?

Let’s start with Microsoft.  First, they need to find ways to stop the Google freight train from taking over the Internet completely while they get their act together.  Remember, Microsoft historically needs 3 releases for that to happen.  I’ve lost track of which release of their Internet strategy they’re on, but the Internet is not a product, so may require even more releases.  I would count Microsoft’s first Internet release as the browser wars, wherein they thought of the Internet as a product and fought Netscape hard to win the browser wars.  Now they have the leading browser but its entirely unclear how that monetizes.  But we digress. 

Facebook is the hottest available property right now, so making a 5% investment would give them some ability to interfere with Google taking it over while Microsoft figures out what to do.  Remember, they still have $21B of cash in the bank and generate a staggering $17B of cash flow each year.  If you had to deploy the astounding cash flows at Microsoft (and they dont’), $500M represents about a week and a half out of the year.  How many properties are as interesting to spend it on as Facebook when looked at as a week and a half?

GOOG does okay too, but Microsoft will win a cash bidding war.  Yahoo, meanwhile, is completely out of the game on these kinds of numbers.  It’s hard to imagine this had anything to do with slowing Yahoo down.

What’s the Microsoft downside? 

  • They can potentially make a return on a $500M investment or not.  They don’t care.  Look at their cash situation again.  Microsoft practically has to bury cash to keep investors from shaking them down to distribute it as bigger dividends.  They are almost too profitable.
  • They can end up buying Facebook.  If it looks right, why not?  I bet they wait until some catalyzing even takes place though.  Meanwhile, 5% may give them visibility inside Facebook they wouldn’t otherwise have had.
  • They can keep others from buying Facebook.  That’s the nightmare scenario, that a competitor + Facebook can lock Microsoft out of further Internet growth.  $500M is cheap insurance to avoid that.

Okay, we can sort of see how it makes sense to Microsoft.  It boils down to the rich having so much money that prices don’t matter versus aspirations and strategies.

On to Facebook.  Their aspiration is likely to go public.  They want to be the next Google.  This deal is great for that.  They get a bunch of fresh capital to fund further expansion–you can buy a lot of Parakeys to launch into your channel for that!  They get a head start setting a huge valuation on their company for the public markets.  Will Microsoft interfere with their going public?  I would think not.  Seems like its easier for Microsoft to acquire them after they are public, all things depending of course.  It certainly would have been harder for Larry Ellison to buy PeopleSoft had they been private.

This brings us to our next question:  is Facebook really worth $10B?  Very unclear from this transaction.  Microsoft has a vested interest in driving the price up.  Indeed, at one point Oracle told SAP and others they could and would pay crazy prices for acqusitions.   The answer to the real value of Facebook will have little to do with mundane issues, however.  It has to do with predicting the future.  If Facebook continues for long enough on their current meteoric trajectory, they will indeed be worth $10B or even more.  Waiting a little longer to see is another good reason for a company like Microsoft to invest in 5% and not the whole enchilada.  Growth curves can often flatten out when you least expect it.

Fred Wilson says this isn’t about setting a real price at all.  It’s about Facebook selecting a strategic partner and charging them a premium for it.  Fred sees it as similar to moves AOL made.  He’s right.  It is a strategic partnering opportunity, and from that perspective, Microsoft looks a better/safer partner than Google, who wants to build a competing network.  The strategic partner angle is also why Apple makes a bad partner, despite what Scoble thinks.  Yes, they are the l33t style leaders, and they have a vocal community, but it is largely an island.  Such a partnership offers more to Apple than Facebook.

Om Malik gets it with his “put option” concept, but I think he’s wrong to say Microsoft will never sell advertising anywhere else.  Where was the advertising going anyway?  Smaller players will always want to play with grown-up Microsoft.  Most of the bigger players are choosing up sides against each other.  Here Microsoft is at least gaining a dance partner.

Getting back to what you wish for, I’d say all parties are getting what they wished for at the time of the transaction on both deals.  Be careful what you wish for…

Posted in business, strategy, Web 2.0 | 3 Comments »

Halo 3’s Sophisticated Game Movie Capabilities are Key to Social Networking the Game

Posted by Bob Warfield on September 25, 2007

Josh Catone has an interesting post on Read/Write Web about the sophisticated game movie creation features of Halo 3.  Basically, you can capture video during game play and view it later.  But, it’s more interesting.  Since Halo is a simulated 3D world, you can change your camera position after the fact, so it’s really recording game state.  The other angle Josh focuses on is a file sharing network Halo’s creators provide for sharing this sort of thing.

Commenters quickly point out to Josh that this is not anything new–you could create videos for Quake 10 years ago (albeit not with the changing camera angle) and in addition, there are many communities sharing such videos for various games. 

The interesting thing about this is to step above the noise of who invented what when and consider what this means and why it makes sense to do it.  Yes, one could argue that since other games have done it the Halo guys just build it in, but it’s because other games have done it and it’s been a popular feature that wins the day.  Why is it popular?  Because it enables social networking for the gamers.

Think about it.  People who play these games love the rich 3D first person shooter (or driver or whatever) experience.  Virtual reality.  If you hand them some lame textual forum, it only takes them so far.  They want richer media.  They want to extend the game experience into the social network.

The moral?  If you’re trying to add Social Networking to another online experience, make sure you capture the personality style and media choices of the participants in the other experience.  For a systematic way to think about this sort of thing, check out my Web 2.0 Personality Style posts.

Posted in user interface, Web 2.0 | 3 Comments »

The IT Appliance World Has Multicore Problems Too

Posted by Bob Warfield on September 25, 2007

Rational Security has a great post on the challenges of multicore for vendors of security boxes.  You’ll recall that the Multicore Crisis comes about because the impact of Moore’s Law on microprocessor performance has changed.  We no longer get a doubling of clock speed every 18 months, instead we double the number of cores.  The problem with this, and the reason its a crisis, is that conventional software isn’t designed to make use of more cores.

There are a heck of a lot of different IT Appliances out there performing every imaginable task.  Cisco has made a tremendous business as have its competitors selling a variety of them.  Inside the black box one often finds a computer running software.  It should come as no surprise that:

  1. Performance and throughput matters to these black boxes.
  2. Since they’re just computers, they can run afoul of the multicore crisis too.

Apparently the majority of these boxes are built on software that’s at least 5 years old.

It seems the multicore crisis will find its way to every corner of the computing ecosystem before too long.

Posted in saas | Leave a Comment »

IDC Says Software as a Service Threatens Partner Revenue and Profit Streams

Posted by Bob Warfield on September 25, 2007

Welcome to the party, IDC.  I’ve been saying that SaaS is toxic to old school ISV partners for some time now.

The problem is threefold.  First, reselling SaaS is problematic–the partner would receive only a small portion of the monthly revenue streams which is much less than they’re used to when they resell perpetual licenses.  To make matters worse, SaaS already has a difficult time making a profit in many cases so there isn’t a lot extra to go around for partner/resellers.  Second, the Services piece of the pie is increasingly bundled into SaaS.  In fact SaaS without Service is just hosted software, an entirely different and much less successful beast.  Third, SaaS vendors have largely made it difficult for partners to create value added IP around their offerings.  Most SaaS offerings have no way to tie other software into them, and the SaaS vendor has worked hard to eliminate as much customization as possible.  What’s a poor partner to do?

For starters, partners should be working with SaaS ISV’s in their space to hammer out areas they can add value.  And the ISV’s should be listening.  These same partners are frequently going to be in customer accounts attempting to sway the customer one way or the other.  If you’ve totally alienated the key partners in your vertical, they’re not going to be pushing hard for your solution.

What sort of discussion can be had between would-be partners and SaaS ISV’s at one of these sit downs?  The key question is to identify what the distinctive IP for the partner is going to be and how they can participate and gain access to the SaaS vendor’s customer base.  Distinctive IP can follow a variety of forms:

–  Best practices within the space.  These partners often know as much or more than the ISV itself about the best practices.  The SaaS vendor can facilitate the partner’s ability to peddle those wares by providing access to the vendor’s community.

–  Adjacent modules.  Inevitably, some of these the SaaS vendor will have earmarked for themselves.  But others are things the SaaS vendor will never get to.  Having a discussion about how to deliver such functionality can lead to API’s on the SaaS product that benefit everyone.  These API’s need not be complex.  SOA is the watchword for SaaS.

–  Integration with other vendor’s software.  Inevitably, there are interesting integrations to be performed.  Once again, the SaaS vendor can play a role in opening up their system to make it easier for their partners to perform such integrations.

In addition to enabling, partners need the usual care and feeding they’ve always been after.  They want to feel like they occupy a special position with the ISV.  They want briefings and training and sales leads.  Put yourself in their position; this is all pretty basic stuff to you, but it’s life or death in many cases for a services partner. 

In exchange for supporting their partner’s activities in building new businesses, the SaaS vendor can be rewarded with a vibrant ecosystem around their product.  They pick up more advocates out in the world and hopefully the partners will be helpful in delivering more sales leads as well as helping to close the leads that are already out there.

A SaaS vendor that doesn’t provide a viable ecosystem for their space’s partners runs the risk that someone else will.  Remember too that most people operate tit-for-tat in partnerships.  Start out treating them well, and they treat you well.  Do them wrong and they stop treating you well immediately.

Thanks to the excellent SaaSWeek blog for bringing this IDC story to my attention.

Posted in saas | 2 Comments »