Week 6 in Startup-Land

I joined Pensa six weeks ago and it has been the best thing I could have done. Mentally my wheels are turning fast and I am learning every day. I have done a lot different jobs over the years, and I come into a new situation like this knowing it will take me a long time before I really know everything I need to know about the business and the customers. As I have written in previous blogs, I have shortcuts I use to start fast, make decisions, deploy proven business and operational models, and then adapt them as I learn. That is what I have been doing at Pensa, and we are making progress at faster rate than anything else I have been a part of. The team here is surprised and excited about what we have been able to do so far, and a bit scared because of what remains to be done. I have pushed the “GO” button hard, and they have responded. I have been basically working 24–7 and I am not tired (yet).

In a few weeks we will launch the “new” Pensa. The company has not technically been in stealth mode, whatever that really is, but when I joined Pensa I was the first full time business person in the company. The company had a tremendous group of engineering people who had been solely focused on technology development. We have since hired a great VP of Marketing, sales leaders for three US areas, and a few additional engineering and operations people, so we are going fast toward the next phase.

I can’t wait to do the company launch, and I would love to tell the story now but I think I can wait a few more weeks. What I can share is that we will bring to market a SaaS offering that will help IT organizations deal with the complexity of delivering IT infrastructure and applications in an all-software, multi-cloud, DevOps world.

I have been talking to a lot of potential customers, partners and others, and here are some of the things I have heard (I would love your feedback):

  • Companies are much more interested in replacing traditional networking with software-defined networking, especially VWware NSX. Many are buying into the idea of using NSX micro segmentation for security. But they are nervous about deploying it in general purpose networking to replace Cisco because they are not sure they have the skills.
  • Regarding DevOps, when I travel outside the Silicon Valley bubble, there is a lot of confusion about what DevOps means, and a range of opinions about how much it matters.
  • Very few of the VARs and regional solution providers I talked to are doing much with DevOps, and they are concerned about not knowing how to play it.
  • VARs in the VMware space are trying to figure out how to build solutions around NSX and software-defined storage but many struggle to build the skills, especially around multi-vendor software-defined networking.
  • IT is still pouring boatloads of money into all-flash storage arrays despite budgets staying flat. Many want software-defined storage to change this.
  • A couple of customers I spoke to said they had planned to do software-defined storage this year but ended up doing refreshes with their existing storage vendors for one more round because they did not think they were ready. They indicated that they want this to be the last time. We’ll see.
  • The other big budget breaker people talk about is consulting for application and software integration. Companies have been throwing bodies at the “IT complexity problem” and new application development projects at a rate that they do not like.
  • I spoke to a taxi driver in Las Vegas who had a junior college education and was going to a technical school to become a Python programmer. His brother, also a taxi driver, had done it and got a 1099 gig in a big insurance company doing IT work. He said it’s a gold mine because nothing works together anymore. Great human story, but when you get around to hiring taxi drivers to patch IT together, something’s gotta change.
  • While the numbers say AWS is growing gangbusters, every single one of the middle market customers I listened to this month talked about the sticker shock they experienced when they saw the actual costs after they got going. The transition cost bubble may not be a bubble for a lot of people.
  • Universally, test cycles on new projects are too long and people want ways to make all these new technologies work together better without hiring more people or consultants to do programming and API work.

This is obviously non-scientific. We are doing some survey work that I hope to share soon. What are your thoughts?