By now you have heard the news that Neotys has been acquired by Tricentis. To those of us in the performance engineering industry, this should come as no surprise. It will be welcome news for larger Enterprise accounts, and especially for those companies who rely heavily on packaged applications like SAP, Oracle, and Salesforce.
In late July 2020, Tricentis issued a press release indicating a significant expansion in the partner relationship with SAP. This announcement was big news for a few reasons:
- This was the biggest announcement from SAP concerning a testing vendor since 2003, when Mercury Interactive unveiled the “Optane” solution for SAP.
- For the first time since the early 2000’s Mercury/HPE/Micro Focus is no longer held as the default commercial testing vendor of choice for SAP implementations.
- This news comes at a time when Micro Focus customers are reeling from licensing audits. There has been serious heartburn about this for the past 24 months, and many long-time customers in the Fortune 2000 have decided to drop their maintenance renewals and look for answers elsewhere.
It’s always been obvious that Tricentis has had the original Mercury suite squarely in its sites – no matter what company the products are currently licensed through. Those familiar with Tricentis know its beginnings as a functional automation solution (Tosca), which goes head to head with Micro Focus Unified Functional Test (UFT). The merger with QA Symphony added a Quality Center alternative, and the acquisition of Flood.io was to complete the holy trinity of QA management and automation with a performance testing solution.
When I read the SAP announcement, I immediately had questions about it and posted on LinkedIn:
What struck me in the announcement was the way Tricentis solutions were packaged for SAP®, specifically performance:
“SAP Load Testing by Tricentis – Optimizes the SAP user experience with scalable, on-demand performance testing for SAP Fiori® and modern cloud applications, such as SAP® SuccessFactors® and SAP® Ariba® solutions.”
This seems to indicate that SAP Load Testing is a full-featured solution – but only for “modern cloud applications”? During the sales cycle, what happens when customers ask the sales engineers how they will handle testing legacy projects? Once the awkward silence becomes intolerable, the response would be to have a solutions partner handle it from there. They will advise to use another product (i.e. LoadRunner or Neoload) to accomplish just the parts that Flood doesn’t. You see how this isn’t a great thing for Tricentis to virtually invite competitors to their clients? If I am a client, I’m then asking the solutions partner, “If I have to use another product for SAP GUI, then why don’t I just use that product for my web stuff too?” At which point the solution partner shrugs and plays dumb, not biting the hand that feeds it.
The problem with Flood
THE MAJOR blindspot with Flood is the lack of protocol support outside of web/HTTP. For those of us who actually do the performance testing, we know Flood is good for what it does. However, it has huge gaps in functionality that Enterprise customers with legacy software still need. These are protocols that would support upgrading. There is too much reliance on third party scripting components, and a lack of comprehensive analysis. It is considered by many to be a developer centric solution that is less useful to the traditional performance engineer. While JMeter/Gatling/Element (based on Microsoft Playwright as of version 2) scripts at scale can certainly handle more modern web-based products from SAP, there is still a lot of testing being done (and needing to be done) on SAP GUI. These open source third party scripts fall short. Other than LoadRunner, there is only one other game in town for SAP GUI…
In 2018, Neotys released version 6.5 of their product with SAP GUI support. This was a major step towards getting on the radar of the Enterprise customer. At the same time, they also released a tool that would allow LoadRunner scripts to be converted into Neoload scripts – enabling an easy switch. In version 7.1 of Neoload, the Citrix protocol was added (which I wrote about back in December of 2019). This was another major step for Neoload to be considered “Enterprise ready”. Since then, there has been uptick of interest in the product by long-time Micro Focus customers, as well as those struggling to scale open source point tool solutions.
This acquisition doesn’t mean Tricentis is finished with Flood. In fact, CEO Sandeep Johri indicated they are continuing to develop the product based on the latest blog which features the acquisition announcement. Johri states, “There are some exciting Flood developments in the works right now; we look forward to sharing details in the upcoming months.” So there will be two performance testing solutions within Tricentis going forward, each aimed at a different role or responsibility.
The problem with Neotys
While Neoload is a very good performance testing product with greater protocol breadth than others (besides LoadRunner), Neotys only has one flagship product. For Application Performance Monitoring they have a tight relationship with Dynatrace, and integrate with others. Neotys is known for being integration friendly with many products to meet additional requirements from their customers. Not all of the integrations are officially part of the product. Because they have been so integration happy, it has caused a bit of a support headache. I’ve had to use some of the more unofficial ones and hunt down the original author to address an issue that could not be handled by Neotys support. It feels like open source at that point.
The bigger issue is that Neoload isn’t part of a larger testing suite that addresses more than just performance. There have always been cycles in the IT industry between “best of breed” versus single-vendor solutions. Many times, it simply comes down to who is in charge of the decision and their philosophy at the time. What has worked for leadership in the past, or the relationships they have rules the day. In my experience, the larger the company, the more apt they are to choose a single-vendor solution to reduce complexity and paperwork for annual maintenance agreements. One vendor to pay, and one vendor to sue if things go wrong. It doesn’t have to check all of the boxes as long as it checks the main ones for now. Because of the typical mindset of the C-level in larger companies, I have always believed being a point tool to be a negative for Neotys. Expanding the protocols is a great first step, but unless it is part of a larger cohesive ecosystem that gets into the areas of test management and test automation of functional, performance, and security – it would always be an uphill battle to get a seat at the big boy Enterprise table.
By acquiring Neotys, Tricentis solves most of the issues hat I have listed above. Tricentis can support SAP GUI (and virtually anything else with the Citrix capability as a catch all protocol), and expand the user roles for both products. Neoload compliments the testing trifecta better than Flood. This reduces the initial resistance from customers looking for the all-in-one QA and DevOps testing solution. This enables (and probably emboldens) Tricentis to further chip-away at the Micro Focus market share as more capabilities are added as a result of combining the efforts of the two companies. In this particular case I see it as a win/win for both companies. It’s a win for SAP too.
Neotys has been on a “slow and steady” growth trajectory for 10 years. It’s safe, but up until recently hasn’t raised an interest in large companies. Yes, having 600 customers is awesome, but their biggest competitor has 10X that number. On the other hand, Tricentis has an extremely aggressive growth plan, and we can see from their annual earnings reported over the last few years the positive effects. Neotys needs to be aggressive right now, as their product is just beginning to mature.
This doesn’t remove all of the gaps for those considering a switch from LoadRunner. There are still key areas that Tricentis will need to work fast to address:
- TruClient – there is currently no real TruClient alternative, or way to convert TruClient scripts easily to another tool. Neotys attempts to work around this by supporting Selenium scripts. Tricentis Flood offers Browser Level Users (BLU) – which is a variant on the headless Chrome software now based on Playwright. Neither of these is a really a great TruClient replacement. Until they can provide one, those companies with a heavy investment in TruClient script libraries will not move away from LoadRunner. If the functional automation capabilities from Tosca could be stripped down to a small footprint and executed as multiple instances on the same machine (integrated into a Neoload generator) – you might have something.
- Analysis – LoadRunner still offers the best and most comprehensive built-in test analysis. This is an area where Neoload is ahead of Flood, but there is still a LOT of room for improvement before it’s considered a LoadRunner killer.
- Diagnostic/Profiling Information – There is still a couple of blindspots where deeper code and database issues hide. Getting inside the JVM or the .NET CLR, or looking at poor performance SQL calls still require additional tools. While the Micro Focus Diagnostics products are way long in the tooth, they still have this capability. Dynatrace adds this capability for Neotys, or a third party integration like DBMarlin for database profiling under load.
- APM – Again, Dynatrace provides an Application Performance Monitoring solution by integrating very well with Neoload. I’ll be the first to vocalize it now: A great next move is the merging of Dynatrace and Tricentis to provide these additional capabilities for an all-in-one SDLC solution – especially around performance.
- Service Virtualization – It is a very important part of implementing DevOps to shift testing left and actively test even when some back end code isn’t ready. While Tricentis has addressed this for Tosca, there is nothing with regards to performance testing.
A key element often overlooked is product support. Mercury made this of prime importance. They excelled in their ability to communicate clearly, effectively, and efficiently. They helped customers solve technical issues like they were fighting a common enemy in the heat of battle at war. Defects were the enemy! When the product itself was the problem, they took ownership and produced a fix quickly. It was an integral part of customer delight, and that is why customers loved them. Their effectiveness is one of the main reasons customers could justify the annual support and maintenance fees. In the final days of Mercury and into the HP transition, support suffered greatly. HP promised to fix it several times – even from the stage at HP Universe in front of God and everybody. And they never did. It continues to be a proverbial thorn, and it is one of the reasons some have come to despise the Micro Focus.
If Tricentis can make support a primary focus for customer satisfaction and continue to invest in ways to make it easy for customers to help themselves while using their products, this will go a long way in securing their dominance and ensure greater success. And even if support is great at Tricentis today, can they scale it? The support and maintenance ecosystem for Micro Focus is currently larger than the total revenue of Tricentis. If market share is taken and Tricentis continues to grow this fast, support must scale with it or they will suffer the same fate. I’m not stating that this is easy by any means. I am stating that it is important, and it needs to be front of mind for people like Grigori Melnik (Chief Product Officer at Tricentis). It should keep them up at night if people start complaining about support. It is the first indicator of trouble. These days it isn’t just about what comments are in the support tickets. Check Twitter for mentions of the company name or hashtags. You’ll find enough vitriol there to motivate any product owner, even when you are doing a pretty good job. People are willing to pay for support and maintenance when they perceive value. But they will skewer you if they feel you are just taking their money with nothing in return. The question is whether Tricentis can learn from the mistakes of HP/HPE/MF?
Another area to pay attention to is the partner strategy that Tricentis lays out after the dust settles from the acquisition. Mercury made some very good moves with large system integrators like Accenture, Deloitte, etc in their early days. They had technical integrations with Oracle, SAP, PeopleSoft, Seibel, and others. They were technology partners with Sun Microsystems, Ariba, Broadcom, etc. While Neotys has dedicated resources to address the partner front and build technical integrations, it hasn’t jettisoned Neoload to the top of the tool pile. I’ve seen more articles on JMeter in the last several years than anything else when it comes to packaged applications, and that isn’t because it’s better. Solution and Technical partners recommend products that are solid and trustworthy and make their implementations easier. Can Tricentis make great relationships with partners and reap a revenue windfall like Mercury did? The SAP partnership was their first opportunity, so we should watch this closely to see how it goes longer term.
What should we expect now?
I have studied moves in the performance testing market for over twenty years. Even though Gartner no longer separates out the performance testing tools with their own space (which I feel is a BIG mistake), I’ve taken it upon myself to research them and look at the SWOT analysis of the major players. This acquisition is a major change in landscape, and should be a huge wake up call for Micro Focus. The wolf is at the door. It leaves only a few remaining players with any kind of market footprint (i.e. Smartbear) outside of JMeter/Blazemeter or a completely feature deprived open source github repo somewhere maintained by a guy named Fred. That doesn’t mean taking market share is going to be easy. Just going by the annual revenue numbers released by Tricentis and the percentage of increase over the past couple of years, let’s say they do over $150 million in 2021. They are still competing with a $3 Billion dollar Micro Focus giant, and just the Mercury/HPE business is bigger than all of Tricentis.
If Tricentis is able to deliver for SAP on the large implementation and integration projects that are happening, they stand to gain a lot of market share. This decision to acquire the capability to performance test SAP GUI instead of building it was a no-brainer. After Tricentis purchased Flood.io, I did not see a lot of push behind it from the parent company, and it seemed to have been floundering around on its own. Last year there were several updates and improvements, but it never looked like a first class citizen in the shadow of the almighty Tosca. Flood lost a few key players last year; Co-founder Tim Koopmans and the socially visible evangelist Nicole van der Hoeven (now at K6). Hopefully, there will be more funding so that Neoload can expand as the Enterprise solution it should be, and innovate faster.
This innovation thing is something that Micro Focus really needs – like a shot of adrenalin. Either Micro Focus will force itself to engage and make innovation a key objective, or Tricentis will be successful in eating their lunch and runaway with the performance testing market. What do you think will happen? I’m sure there are a lot of people who will be watching Tricentis more closely from now on. Will a dark horse or new upstart jump in to take the remaining market space? Octoperf has been hard at work improving their product, and Gatling just went into a BETA for their new SaaS platform. Will these features be enough to lure anyone in the Enterprise away? It will be interesting to watch this all play out in the market over the next 12 to 24 months, that is for sure.