Every now and then there are discussion about the future of Software Testing. Lately I’ve seen a post from Lanette Cream (The future of testing is wide open) and Jeff Goldsmith from SauceLabs (The Future of Testing). Trends are around Agile Testing, Automation, Continuous testing, Crowd Testing and last but not least: Cloud Testing. It matches the results of the poll I’ve created a while back. If you’re passionate about Software Testing, you’re following all these trends and probably implement some of the new innovation within your team. I know I did while building my organization at Experian Decision Analytics:
- We’ve invested in Selenium to run our functional and some performance testing for customers.
- Built various homegrown framework to run continuous automated testing.
- Start using Amazon EC2 to run scalability testing for some of our web application.
- Adapt our testing process to accommodate for SCRUM.
These initiatives made a huge difference in the quality of our products and I think changed the way QA was perceived within the broader Experian Organization. Best of all, I really think it made a difference for the people working for me who had an opportunity to work with the latest innovation in their field. I’ve always tried to keep people excited about their job and Software Testing in general and one way to do it is open up their horizon and make sure they have ample opportunity to make a difference. This said, influencing Software Testing within an organization is one thing. Influencing the whole industry is a complete different ball game. If the company you work for is not a pure player, it might be a bit tough, frustrating and much slower than what you’d want, especially if you’re very passionate and want to make a real difference! I’m one of these people who wants to make a difference during their career. Not only to the company I’m working for but also to the broader industry.
That’s reason number #1 I’m joining SOASTA: To have an opportunity to revolutionize Software Testing (or at least part of it ! )
During my various interaction with customers while working at IBM and Experian, I’ve always been struck by the challenges encountered by Performance Engineering team:
- Challenge to reproduce like-to-like customer scenario.
- Challenge to simulate realistic and flexible loads at a reasonable cost.
- Challenge to deploy and maintain a load grid/lab.
- Challenge to generate a geographically spread load.
- And for me the biggest challenge of all: How in hell can I make sense of all the data received during tests (we’re talking Terabytes of performance data received from all application and hardware layers !) and provide valuable aggregated information allowing me to take actions. Oh and by the way, I’d like these information to be available in real-time during production performance test. Yup, real-time analytic on a live environment! Testing in a lab, while important for extracting some type of results, cannot answer questions about production performance or capacity with a high degree of accuracy and confidence.
All these challenges come from the following observation: The most common test tool in use today for web application testing was never designed to tackle these type of challenges. You have a tool today used by thousand of customers which is, from technology to licensing, completely inapt to answer requirement for web sites with a significant presence. Yes, I’m talking about LoadRunner of course. It makes customers believe that testing with a few thousand users is good enough in today’s world. It makes customer believe that analyzing production performance data, 20 minutes after the test is good enough. It makes customers believe that we’re still in a world of expensive Enterprise License Agreements when the reality is Software-as-a-Service and On-Demand. LoadRunner was relevant 15 years ago, it’s not anymore. As discussed in a previous post, I was forced to splurge a significant amount of money to get my organization some LoadRunner licenses for the sole reason to be credible in front of our customers. That’s all they know today and if you’re not using LoadRunner, you don’t do Performance Testing. Period. It made me sick frankly. LoadRunner is not a bad tool all together but it doesn’t meet the challenges of increasing complexity and can’t scale up to the real-world traffic levels applications are seeing and it’s definitely not cost effective !
That’s reason number #2 I’m joining SOASTA: To have an opportunity to offer to all customers in the world a real alternative to Loadrunner. I want to help them get the most out of their web applications which is sometime their only source of revenue. I want to help them provide their customers the best online user experience. I want to provide them confidence about their online operation. And when they come up with a critical online offer, I want them to avoid the AT&T syndrome !
I’ve been working within Engineering teams all my career and I think I have a reasonable understanding on how to build good software But building the software is only one part of the whole Software Business. There is a large side of that business that I’d never been directly involved in: Marketing, Sales, Market Research, developing channels, finding resellers, building partnership with ISV, building a brand, intelligence gathering on competition, generate leads, handling customer account etc. If one day I want to run a business (either mine or someone’s else) I better start getting myself involved in these activities! Do I have the required experience? Hell no! But you’ve got to start somewhere. And as Steve Blank said “Nothing I couldn’t fix. I took the job!”. I believe I make up my lack of experience by enough business acumen, domain expertise and a whole lot of passion! And a real will to help customers with what I think is the best for them.
That’s reason number #3 I’m joining SOASTA in a Business Development role: Broaden my experience and prepare myself to run a business during the next cycle of my career.
Last but not the least. The Startup part. I’ve started my professional career in the US in 1998. Right in the middle of the Internet bubble. Startup were everywhere and something struck me the most: Constant innovation and business agility like never before! I was working for IBM at the time and frankly I was a bit jealous. I was learning quite a bit and will never regret that time! It brought me a very solid foundation in engineering and management. But I’ve always been intrigued by startups: How can so few people build such innovative products/services? How can they adapt so quickly to a very dynamic economic environment? When you work for a large company for a while, at some point you wish it could go faster. You’re looking for snap decision, agility at all level of the organization, dynamic budgeting, making daily impact instead of spending a long time influencing a slow moving organization, bottom-up approach where nothing happen unless you make it happen (versus demand-driven initiatives). Working in large company can become very comfortable and you always have to challenge yourself to get yourself outside of your comfort zone. Working in a startup always put yourself outside of this comfort zone because you have to do so much, so fast, sometime outside of your domain of expertise (large company provides very comfortable silos). At this stage of my career, this is what I’m looking for!
That’s reason number #4 I’m joining SOASTA: To work in a very dynamic environment, get out daily out of my comfort zone to experience things I’ve never done before !
Last week, I’ve joined SOASTA as Vice President Business Development Europe for all the above reasons. They’re game changers, a growing startup, they have a strong customer base and best of all, they’ve realized that LoadRunner is no longer relevant! I’m returning from a trip at SOASTA HQ in Mountain View where I had a hell of a time with a FANTASTIC team! I’m going to have the time of my life !
I will of course continue to maintain this blog! Maybe broadening the topics, maybe not What do you think?