Lol, from what I have seen in the field, the 'tax' is used more as a persuasion not to use Hikvision than to make money, i.e., instead of paying Genetec the 'tax' put that money to buying a non-Hikvision camera.
Also, generally, Hikvision and Genetec are increasingly not playing in the same segments anyway.
I don't know the whole process but there is some level of checking / validating and many manufacturers do cooperate with IHS.
How do you ensure, e.g., that Genetec is accurately reporting how much VMS software revenue they generate in APAC (as an example) strikes me as a hard problem. Many / most of the manufacturers will report honestly but it is not clear how IHS can check / disprove false numbers.
This is one of the reasons we have never attempted to do such market numbers reporting as it depends on the goodwill and honesty of the manufacturers.
The issue here, again, is the regionality of your survey. It is not representative of the environment outside of the US where Genetec is far weaker and not as well established.
Without going in macro detail - if these surveys were polled as "US/Canada" and "all other regions", you would see a very different picture. Whilst it is representative of your subscriber base, this in itself is not representative of the global picture. My concern here is that IPVM seeks to use a global stage based upon a heavily biased US perspective. In presenting results of surveys it would be more accurate to use a recognised survey weighting rather than a core of US based subscribers.
In response to the IHS survey, having provide face to face info to IMS previously, I can tell you what an unqualified farce the process is. There is nothing subjective in the analysis and every question is led to achieve the response sought and that can be sold. As John says, I don't know the whole process either - but I have been interviewed by IMS so I know how this part of the data collection works - and its complete garbage. Every time I stated a response that was negative to their preferred one - they ignored it. Where I said limited growth they used that to sat growth. A maybe became a yes and a no wasn't recorded. The person asking the questions knew nothing of the product, manufacturer or industry and just sought to collect anything they could use to substantiate an over-inflated claim.
IMS/IHS statistics are rarely, if ever, held to account. Headlines are based on the blatantly obvious and then synthetics figures are thrown in to substantiate of them. Various start-ups then use this as fodder to sell to the markets and investors so they is a complete end to end industry based upon fresh air.
The issues I see for Genetec (and Milestone) is they may become the victim of their own success. If they keep charging high fees to extremely large enterprise customers year after year, how long will it be before some of those enterprise customers just decide to roll their own VMS and dump the fees (and Genetec) permanently? Sure there's a dev cost and an going maintenance cost of writing your own software, but somewhere there is a point where that becomes cheaper than paying someone else a huge fee every year. And with an all IP solution, there is really no hardware to deal with, just software. Some of their customer base are more than capable of doing that and I wonder how many have already tried, or did deploy their own solution.
before some of those enterprise customers just decide to roll their own VMS and dump the fees (and Genetec) permanently
Super low probability. Genetec (and other high end VMSes) are putting in hundreds of man-years into these products. Maybe a few of the tech companies but practically no city, airport, retailer, etc. is to going to try to do something like that. You might counter with "Yes but they don't need all of those features" which may be true but that would encourage them to switch to a cheaper VMS that is 'good enough' rather than developing their own.
I agree, this is a possibility, but highly unlikely. Many underestimate the effort that has gone into developing the leading VMS platforms. A customer would need to spend a ton of time and resources replicating a VMS. Regardless of the size of the company; time and resources are finite. Any rationale company would focus it's resources on its core competency and use the best of breed VMS for security. A scary exception could be Google but even if they did it I think most would be wary by now.
For argument's sake, let's say Genetec makes weird choices and has outdated tech. And let's say in a perfect world, someone else with new tech and 'normal' choices can do it in tens of man-years instead of hundreds.
It would still be a bad decision for most enterprise to build their own VMS. Maybe NYC hires you and you write them a VMS but I think most enterprises would spend far more money, time and headaches trying to put it together than just picking one of the dozens of off the shelf ones.
Super low probability. Genetec (and other high end VMSes) are putting in hundreds of man-years into these products. Maybe a few of the tech companies but practically no city, airport, retailer, etc. is to going to try to do something like that
There have been 2 major companies I've worked with that tried to do this, both were total clusterf*cks. One is now in it's 3rd MAJOR RFP in 5 years. But they've got $$$$$$$$ to burn and a revolving door of leadership
The other spent millions in man hours building theirs, including trying to write analytics. In an even bigger laugh they tried to market/sell this to other Retail competitors, even getting booth space at ISC one year. That went over like a lead balloon.
They have so much invested they can't change, but their entire organization is suffering. Given their low, low margins, they have zero cash to get out out the quagmire as Loss Prevention is typically the last to be funded. It would have cost them less than 1/3 of what they've spent overall to have simply bought an enterprise product up front and stuck to doing what they did well
I will say I have known a few organizations that have considered or tried to do this, and I have never heard of it going well. It does not mean it is not possible that there are successful ones, but I do not believe it is a real worry for large VMS manufacturers. They would pivot their sales models far before it damaged them significantly.
A more likely (but still probably unlikely) scenario would be for a VMS company with a decent product but low market share, out of desperation or genius (or both!), try and give away or even open-source their software and becoming a pure consultation and services company on that product. And some large end users choose to come aboard that model.
In additional to John’s comment, large enterprise customers are not only looking for the expertise of a company but also the responsibility of the company.
Selecting a solid leader in an industry also provides political insulation from making a wrong decision. If a company picks a leading/bleeding edge company, even if their product is more advanced, it’s a risk. Smaller companies are vulnerable to growing too quickly, losing their core technical personnel, challenges with supporting customers, or being bought out.
Granted this can happen to any company in any industry but, as an example, an individual is less likely to be called out for signing a contract with IBM’s data center services vs. Cervalis data center.
With 300+ votes, Genetec leads Milestone 52% to 35%. Interestingly, amongst integrators, the split is much closer with Genetec edging out Milestone 46 to 40; however, amongst manufacturers, it's 53% to 33% in favor of Genetec.
Time and Time again it has been proven that going with the biggest doesn't always guarantee success in of itself. There are plenty of manufacturers who are not only as capable but more responsive. And, really, when you think about it, who in our business, (other than goliaths) wants to go out and promote a product produced by a company who's business strategy specifically spells out "The gutting of mid and low-end players". We don't consider ourselves to be "low end players" but we are a mid-sized business who takes great pride in what we do and the level of service we provide. Not sure that economic discrimination of dealers is the smart move in the long run.
To clarify, Genetec is not trying to gut the mid to low-end, Genetec was saying that since that part was getting gutted that they were better off focusing on the high end. The gutting was done by Hikua. Also, Genetec is alluding to manufacturers there, not integrators.
I can think of at least one underlying reason why larger organizations might want to try to do this. No matter how big a VMS company gets to be, it is very common for large scale (enterprise) end users to find something inherently missing in the software feature-set that they truly need and still do not have. - features that even the best and the biggest VMS's are not including. Having some degree of familiarity of the typical software development process, I can see why this might be frustrating for some of these end users. We have some customers that actually have some pretty good feature-set ideas but are told it will never make it past the first cut of SD (Software Development) - which is to see if it is being requested on a large or national scale by many other users. That being a common requirement for most feature development practices by most VMS's, this rule seems to contemplate that a great feature is only great if alot of users say they need it. I do not agree with the premise that a good feature idea is only a good feature idea if alot of people thought of it and needed it simultaneously. Doesnt or shouldnt matter if its a good or great idea.
But at the end of the day a VMS mfr is not going to spend money to develop even a GREAT idea if there is only one user asking for it. And that frustrates alot of enterprise users and some (albeit few) even go so far as to decide to do their own thing to get what they want. So.........
Why are we, as an industry, still not giving alot of folks what they want? Or at least listening to them much more in terms of their needs?
The overall concept of your post has several flaws, but there is one really big one that I will address here...
Every one of these manufacturers has a professional services department. With few exceptions, they can and will write just about any feature set into their product that you want...for a price. At the end of the day, just because something is a good or even great idea doesn't mean that it's worthy of the engineering/dev/qa time to build it into the core functionality of a product. In fact, I would argue that this type of all-encompassing approach is what has killed or severely hampered some products over time -- namely Lenel. When you spread engineering resources so thin, you end up with flaws and issues and a lack of scalability or flexibility.
Bottom line, if these big enterprises want these additional features, they can pay professional services to develop them. Yes, it will cost a kings ransom, but I can guarantee that it would be substantially less expensive than trying to grow their own.
With enough money, you can do anything - if you've got a lot of money, you can simply buy a VMS company, and hope for the best.
I'd like to make a few comments on buying bespoke features from the pro services teams.
There's an assumption that developers within the VMS organization are better equipped to implement the feature you're looking for. This is not always the case; sometimes the company will simply act as a middle man between you and an independent contractor who has no prior knowledge of the VMS. You're paying the pimp so to speak, and someone else is doing the work. While you do get the privilege of yelling at a rep for the company, you're not going to force a solution out of them. The pimp who sold you the integration left the company 6 months ago, and the lady of the night overdosed last night, so all you get is a bunch of apologies, fairytales and requests for log files.
The VMS might be aging and filled with hacks and strange quirks, this makes the development of a feature unfeasible on the platform. There might be feature-gaps, or a function is undocumented and not officially supported. Often you'll find that employees at the VMS vendor are not aware of the functionality. Ultimately, you might end up paying 3x the cost for an integration because the VMS platform is a pile of outdated spaghetti code, and once you've paid up, you're even more entrenched with the vendor (who's loving the deal).
But, by far, the biggest risk is API changes
When I do integration work, the contract stipulates that the integration is made for version X.Y.Z of product A. If the client switches to another version, they may remove or change features such that the integration breaks. The cost to upgrade the integration can be as small as a single line of code, but it may also completely destroy the integration and require a complete rewrite of the integration. Some companies are great at maintaining a stable interface, others are absolute dogs - it comes down to cultural differences of the companies I think.
One of the things that can make it substantially cheaper to get bespoke work done is on the "idiot-proofing". One example could be that you need to manually restart a service or avoid certain characters - those things would be unacceptable in a retail product, but MIGHT be tolerable in a custom integration, and could potentially make things cheaper.
I don't think it would make sense for a sausage manufacturer to enter the VMS software realm, but there certainly are situations where a completely bespoke recorder/client might make sense. I see a lot of companies that buy expensive VMS systems that provide a terrible UX for the relatively narrow use case they have.
You make some good points, but most of them are what I would consider "common sense"...which we all know are not always so common.
Of course not every professional services team for every VMS manufacturer is top notch, but there are a ton of ways you can protect yourself. I would think/hope that anyone who is making such a request within any large enterprise would be familiar enough to understand that or self-aware enough to hire someone to make sure they understand it and do the right thing. At the end of the day, if you're asking Lenel to write some custom features into your instance of Prism, you deserve everything you get out of that deal.
What (I thought it would be obvious that) I am referring to are companies that use the "enterprise" platforms that are clearly some of the best-in-class products -- I'll use only two here because they are in the title of the thread, Genetec and Milestone, but there are others out there. While I'm not saying that every single interaction with their proservs is going to be amazing and fantastic and top notch, if you work with them to assemble a proper scope of work with all parties having an understanding of what needs to happen, you're going to get a pretty good product out of it. If you don't do that front-end work, you probably won't get a very good product. Just like with virtually everything else -- garbage in, garbage out.
You point about API changes falls in the same category -- the best-in-class players do a REASONABLY good job of either preventing or at least mitigating the effects of "breaking" changes to their systems, and they'll work with you on getting it fixed either way. Yes, probably for a fee.
Maybe I have just been lucky enough to never have any interaction with end-users that are arrogant enough to think that they can build something in-house that can compete with some of these other products out there, at least at a large scale. I have no doubt that someone could home-brew a basic system with a few random features, but for any enterprise-class product that has a full feature set, there is less than a 0% chance that any company is going to be able to home-brew their own comparable platform. It's just not possible.
That said, I suppose that when even executives/leadership at Boeing take the "bUt CoDe iS cOde" approach, I'm sure that means there are leaders at other companies dumb enough to try to build their own enterprise-class VMS.....so I suppose your point is valid. That doesn't mean it's in any way reasonable that a company would attempt it, but I'm sure that you (and others) are correct that some idiots will try it...
I've never met anyone who wanted to do it either, and I'd try to talk them out of it if they suggested such a thing. But I do think that a lot of users would be better of with a bespoke system vs. a mangled/modified "enterprise" system.
E.g. there's hardly a need for an "enterprise" system to record interrogations. It's a super trivial task and is probably doable with wowza, ffmpeg and node-red. It might even be a better and simpler solution as you'll get stand-alone mp4 files that trivially be copied to another media.
It just seems to me that a lot of times, users are being sold a jackhammer when they need a pair of pliers under the guise of the jackhammer being "enterprise".
Nah. The post was contemplating why some enterprise users try to write their own VMS and I said at the core, their reason is that they are not getting what they need from the VMS industry. The overall concept of that base theory is sound - I have personally and specifically experienced this on numerous occasions in my career. I agree that not specifically calling out in my post that desired feature sets not already present can be purchased for a dollar value from any VMS company was an oversight on my part, but it was not the intent of my post in the first place. There can be many reasons why some large enterprise users attempt to write their own VMS's. It is only logical that one of the reasons they do so is because they find their applications to be unique in more than one regard and they have not been able to find even 1 major player who has enough of their desired features included. Whether the industry has them, doesnt have them, or makes the customer pay for them is irrelevant to my original point. If they ultimately find out they will have to pay for the features they are looking for because no one in the industry has yet to address them, then I too would just make the leap and say, hey if im gonna pay for software development time to get what I want, I may as well own the whole dam process and control the entire piece and make it 100% of what i want. In both scenarios - your approach AND the big end user who is creating their own VMS - the end user is paying for software development time. If thats the case, I too would want to own ALL the SD time. That way I know for sure Im going to get everything I want. Thats why I think they attempt to do it (for better or for worse is another matter and another conversation) and i believe that is both a reasonable assumption to make and a sound theory to put forth for consideration.