I would love to be proved wrong on this, but most of the startups that are focused on:
- Observability
- Security / Compliance
Seem like they're plays by the founders to get acquired by bigger firms (either for acqui-hire or just to augment their own "observability and/or security" offering).
I would be very curious to see a company that really make it big and remain independent in this area. It seems to me like DataDog has been the only successful company in this area, with some other honorable mentions (e.g. honeycomb.io, sentry, sensu etc.)
This seems to me because after connectivity (“solved” with service mesh) getting proper insights at scale on observability is quite hard. Security is and will always remain har as one mistake can have large consequences.
I once served as a PM for a startup in the observability space.
Monitoring is a brutal industry, as engineers love to build monitoring products. There is a long tail of hundreds of monitoring startups catering to every wim that are backed by only a small handful of engineers.
On the other hand, integrating a monitoring product into a company large enough to pay for the product is extraordinarily expensive. The DevOps team might want to have end-end visibility, but they can't add the library support required to make the deal work. Supporting one companies specific libraries is unlikely to pay off for other companies.
When it comes to distributed tracing, the value proposition depends on convincing every dev team in the company that this is the best solution for monitoring.
Larger companies have an easier time "cross-selling" these features, and can sometimes win deals by simply having the capability even if its never used.
I mean, few startups in any area make it big and remain independent. It's probably just that startups in this area are more visible to you because their products are ones you might use.
I think you're right -- if only because it's probably the most viable path to growth.
If your value prop is 'we're stable and you can trust us', and your buyers are 'large enterprises that don't like to take risks', your biggest impediment if you have a great product is that you aren't part of a company like ServiceNow.
What's the saying? "The battle between every startup and incumbent comes down to whether the startup gets distribution before the incumbent gets innovation."
Even harder when you need enterprise proof-points...
It boggles me that ServiceNow is as popular as it is. The software is dreadful to use. We try as hard as possible to avoid anyone ever having to use it through slack bots, data exports, etc.
And this paragraph says nothing (what company isn't already digital?):
Companies are betting on going digital in order to thrive in the 21st century, but the transition is often challenging to navigate," said Pablo Stern, SVP & GM, IT Workflow Products, ServiceNow. "With Lightstep, ServiceNow will transform how software solutions are delivered to customers. This will ultimately make it easier for customers to innovate quickly. Now they'll be able to build and operate their software faster than ever before and take the new era of work head on with confidence.
Yes! Close would attempt to change the status of a ticket from Open to Closed, but often fail along the way with a cryptic error about the RADstack being 99% full.
Exit would close the window, and also break your workflow if you had HPSM open in two browser tabs
For a very long time, HP Service Manager didn't really have a functioning web client at all. Customizing and distributing the thick client to end users was just part of the fun that came with managing that platform.
As competitors like ServiceNow came onto the scene, they tried to create a better web client, but at least when I was still close to that product, they never quite got there. My two final HPSM projects were:
1) Building a mobile / blackberry compatible web client from scratch (HTTP/SOAP) so management types could see/approve change requests on the go
2) Migrating to ServiceNow :P
It would not surprise me at all if they had some kind of ActiveX thing going on with their web client...
Oh wow. I still remember having to create change records in HP Service Center. You had to have split screen, with all text (change description, list of impacted CI’s etc.) besides it in notepad. Then it was quick hand-eye coordination and quick CTRL-C - CTRL-V. Too slow? You’d hit the timeout on the last screen and could start over!
Agreed. This is a particularly petty grievance but it boggles the mind that the landing page for ordinary users is some backend ticket management view and the thing users want is the cushy ticket requesting page with the view of their active tickets and so on which is unfortunately buried behind some unmemorable path and which is not linked to from the obscenely noisy landing page (or any of its network of pages that most users don't care about at all). If you're smart, you create a bookmark and move on with life, but I keep treating it like it's the last time I'll need that stupid link and of course that's entirely my fault. But still, designing software that requires individual users to manage bookmarks is insane.
So yeah, that was my petty grievance--no doubt I or the administrators could make this more pleasant. But I've never heard a single soul have anything favorable to say about SNOW. I don't know if SNOW's success is attributable to an unusually impoverished competitive landscape or if there are hidden factors (it appeals to IT's innate enmity toward the rest of the organization?), but it is surprising that SNOW survives at all.
Having worked at a company with a ServiceNow implementation, I think I'd summarize my experience as: ServiceNow can be transformative when implemented by a competent team, or utterly useless when not managed well or left as-is upon deployment.
Orgs get in trouble by trying to solve internal process inefficiencies and related people problems by throwing ServiceNow at it. The "digital" version of a shitty process is still a shitty process.
> But I've never heard a single soul have anything favorable to say about SNOW
I've been to the conferences, and there are plenty of folks who love the technology stack, and can do powerful things with it. But I suspect that most people experience the product as an end-user, and that experience is 100% dependent on the team that owns/implements the platform. There's a big difference between "We have a center of excellence for automating manual processes, and ServiceNow is part of that solution" and "We brought ServiceNow in to solve world hunger" .... ("is this thing on??...")
Reminds me of SharePoint. Some people swear by it, but my impression is that it's something that incompetent managers buy and then have to shell out later for "SharePoint consultants" to come in and actually make it useful for your organization (and whether or not it is useful depends on the quality of said consultants) and the best you can hope for is a solution that is 80% as nice (from a user perspective, not an IT perspective) and 2000+% the cost of GSuite.
"Going digital" is a continuous process, and while I hear what you're saying, SaaS vendors like ServiceNow have been pushing "digitizing processes" hard in the pandemic era, because most companies still have a long backlog of manual/paper processes left to "digitize".
To understand the popularity of ServiceNow, one need only look at the legacy products it displaced: BMC Remedy, HP Service Manager, CA Service Desk, etc. ServiceNow may not be sexy by modern standards, but gives large enterprises a level of flexibility and agility that was never possible with the legacy vendors.
I've had some realizations recently around this. The quality of the software is in my opinion not a huge factor for the success of the company. Look at car companies like Ford, Toyota, VW. Compared to a Ferrari, Porsche, Bentley they are crap, but yet they are still relatively successful companies in their space. Software is sold to managers. How they make their purchasing decisions is the biggest factor, which comes down to how good the salesperson they interact with is, and their decision making framework. Even if you could make a better platform than service now, you still probably couldn't convince the decision maker to use your software instead. Things like how much they trust the company will be around long term, the support offered, the list of "features", personal relationships are usually more important. A company is more than just the overall relative quality of the product.
> Look at car companies like Ford, Toyota, VW. Compared to a Ferrari, Porsche, Bentley they are crap, but yet they are still relatively successful companies in their space.
This analogy doesn't work here IMO. Indeed, in many respects like reliability, a Ford, Toyota or VW is actually much better than a Ferrari. They're just very different brands: one set are mass market brands and the other ones are exclusive luxury brands, and they have different features that appeal to each.
But Ford, Toyota, VW cars are certainly not "crap", and e.g. Toyota is renowned for the quality of its vehicles.
Yes, it's not a perfect comparison... But if you were to ask someone who regularly drives (or is driven in) A Bentley whether a Toyota RAV4 is crap they will probably say yes. Why? Because what's important to them is the luxury, and the bells and whistles of their Bentley. Compare this to software where they largely have the same "features", but one platform is easier to use than the other. Is the software that is easier to use "luxury software"?
It’s easy to forget this if you work in tech, but there are many, many enterprises still stuck with spreadsheets and legacy software even worse than ServiceNow.
Think about regional banks, governments, utilities, oil & gas, manufacturers, field engineering services, healthcare systems, ... These aren’t exactly customers who can just use Slack instead.
I've been re-implementing Remedy functionality for a telecom operator around year 2000. They decided to go with custom software because Remedy was horrible. I wonder what kind of horribility it is now, 20 years later.
And we still have to use ITSM systems that look and feel almost exactly the same like 20 years ago. How it's possible, the ideas behind these tools are not exactly rocket science, simple ticket tracking with some standardized data structure and relationships - nothing that you couldn't put together in a month. I wonder why companies even buy these tools, probably you could have much better system developed for you and your specific needs for 10% of what it costs to buy certified ITSM system.
I remember using ServiceNow in the finance industry many years ago. It was a fairly comprehensive dumpster fire back then. This above paragraph does not bode well for future improvements.
I'm not at all a fan of ServiceNow, but it's primary use is for huge organizations, up to and including entire government departments with hundreds of distinct contractors and client agencies. There is no practical way to use Slack bots to manage ticketing and work tracking when you have tens of thousands of users and Slack is not approved for any kind of controlled data (PII, PHI, CUI, Classified, Proprietary) and doesn't offer a self-hosted version.
service now has horrible defaults. access is so restricted you can’t see anyone’s name and no one can see your ticket but you. you can submit something and have it not move and have no idea who to talk to. then your communication tools become part of the service now workflow
We use Jira for developer work tracking. And use bots and humans to avoid developers ever having to visit ServiceNow during incidents and post mortems.
ServiceNow is a classic Enterprise software play that I like to call "old, but new".
You take the same old software, with the same old processes, the same old salespeople and sales approach, and you make a new one that is a little shinier and (crucially) freshly made. Because you used your same old connections and investors to fund the development, you have a much better product than whatever cruft HP Service Manager, CA Service Desk or BMC Remedy have been pushing, simply because you haven't been fiddling with a product originally written in C++ in 1998 for 20 years.
Furthermore, your solution is comprehensive enough that it covers most of the existing functionality of the old products in the space, and it ticks all of the same compliance checkboxes too. So there's very little career risk associated to buying it. You even took the customer to the same old restaurant you used to take him when he was buying HP software from you!
Another popular example of this strategy is Zoom. What is Zoom? GoToMeeting/Webex, but new.
On the lower end of the market, there is more entertainment, because not everyone that had legacy products want to pay through the nose for the pleasure of dealing with an army of Service Now suits. This is where companies like Freshworks, Atlassian and ours (InvGate) come in.
Source: founder of InvGate, mid-market player in the same space
If we end-up in a similar discussion after 10\20 years, your comment would still be valid because Chatting, Operations software are part of very basic needs of any org. So, at that time, your comment also would be - "old, but new" :-)
I agree that it's better, but that's the point. It's better in the same way that Service Now is better than HP Service Manager: It's essentially the same product, but started from scratch 10-20 years later.
There is value to that, and going all the way to IPO by succeeding against second system syndrome for 10 years straight is no small feat.
This is as opposed to other companies that actually change the way things are done (i.e. GMail is not "Outlook, but new").
I wholeheartedly agree with you with the exception of Zoom. We started to Zoom over 5 years ago and it was shockingly better than everything else we had used and tried (almost of them). The great thing is a lot of their competition have slightly improved their game as well which rarely happens in Enterprise software. Companies usually continue to use crappy enterprise products despite much much better startup alternatives out there.
Like I said to the other commenter, I agree that Zoom was better than the alternatives. I'm just saying it's the same product as Webex. They didn't come up with a new concept or new technology or a new business model. They just made a higher quality wrench. Which is totally fine, and is clearly worth a lot of money. People need wrenches.
I don't know about the main people being power users. When I worked at a company that used it, virtually all of Engineering, Support, and Product were constantly using ServiceNow's horrendous UI to deal with customers and internal IT workflows.
Like others said, we tried hard to not have to interact directly with SN, adding scripts and integrations into other systems.
It boggles me that ServiceNow succeeds even though their most ubiquitous product, IT Service Management, is also their worst. There are less half-baked products besides the ticketing system.
For example, the IT Operations Management suite includes tools to collect specifications about devices without installing an agent, aggregate alerts to better assess outage impacts, and automatically generate topology maps by analyzing network traffic. It's a wholly different product that justifies some weird design decisions in the Incident/Change system, but it can help explain where Lightstep falls in the ecosystem.
Source: ServiceNow ITOM consultant who has seen customers fare better when they look beyond Incident/Problem/Change/Catalog
Usual suspects are Freshservice, Jira Service Desk and our (very) own InvGate Service Desk (demo instance link below because how can I not pitch my own product??).
Yes, if someone can enlighten me - what is so great about this software, by looking at their product's growth and popularity i was expecting something extraordinary. Then i had to use it and it was pretty boring, unintuitive, unsophisticated CRUD framework for tracking stuff and it wasn't clear at all where is the value it provides over thousands of other applications of such kind.
The big plus of ServiceNow is that it’s slightly less terrible than the competition.
There simply is no good product in this space, at least not for enterprise clients with 1000’s of users and even more regulatory and internal requirements.
To me the best thing is that it is decently quick compared to other systems that I have used. Querying a table directly in SQL? Instant. Querying that same table in Remedy? 30 seconds. I'm sure the people behind Remedy have a great reason for it, but I do not care.
Servicenow occasionally does a spinnywheel but 99% of the time I have loaded the table I need in less than 5 seconds.
Plus:
- I can open it in multiple tabs. Not the shitty in-app tabs with unclear labels that take forever to switch, actual browser tabs. Other solutions give you a session collision if you open two tabs. In SNOW you can middle click anything.
- The search is decent.
- I dont have to click a goddamn 'form fill' button on every single field. If your form designer isn't braindead you can tab you way through the form without needing a mouse.
- I can make my own templates for common tasks and use them right away without invoking a manager. I can put my favorite templates on a toolbar instead of having to click into a sub-page.
None of these things are big asks but they are things that, as a user, the competition is chronically bad at.
Maybe it depends on the deployment, but my SNOW experience was the polar opposite.
It was dogshit slow, search was useless and painful to use. Most things required a mouse. Tickets got lost. The UI looks like 2006 and the UX is out of a Microsoft text book.
The fact it can be used across tabs isn't really an achievement, but if your bar is so low...
In my experience with SNOW middle clicking on anything brings me to the page I want, but it removes the navigation bar from the side which makes it almost less than useless. And not to mention the weird thing where it kicks you out of a ticket after you put it in.
Where I work ServiceNow and JIRA are complementary. We manage operation and support with sevicenow and manage development with JIRA.
Maybe they will eventually drop JIRA for the ServiceNow agile development module but they (the deployment team that I'm not part of) have many modules to deploy before evaluating that.
If I had to sync tickets I would also wish for a Jira only workplace as I mostly do projects that require development and planning but I am almost sure that the peoples who mostly do customers and employees support would have the opposite opinion.
OTRS is not really a competitor. It's a crusty open source solution whose only advantage over ServiceNow is that it doesn't cost upwards of $1000/user.
It's not really in the same ballpark. The decision was certainly financially motivated.
Yes, LiteStep was the first time I felt really in control of my computer. Virtual desktops, lighter shell, etc. For me it was the gateway drug that got me to eventually move to the Linux desktop.
My org was just stuck with Service Now recently when someone convinced management it was better than Zendesk, got the contract signed and then resigned when the integration became difficult. Wish I could leave negative feedback on his LinkedIn profile, but the guy even removed mention that he worked with my org and claims to have been self employed the last 3 years.
acquisition might have been their strategy all long, but I can't see ServiceNow doing anything but strangling what was a cool tool - though maybe in the best case some observability will carry over to huge enterprises
I don't get it. What is the value add? Apparently, more tracing of workflows and exposing traditionally DevOps style metrics into workflows that everyone in the org can appreciate. I'm not convinced. Maybe lightstep is having issues with growth and sold out?
This seems likely. Went through a tracing RFP at a large company and people knew products like datadog, splunk, appdynamics, even honeycomb, but had never heard of lightstep. Even ex-googlers who knew bhs had never heard of lightstep. And when I was being recruited by lightstep in early 2020, before COVID, the process ended in a hiring freeze. I never got the impression they were thriving.
Corporate press releases [1] are a bit of an exception to some of HN's rules (such as "please submit original source") because they are suboptimal for curiosity [2]. Mostly they are bland boilerplate, and often are written to conceal as much as they reveal. So we tend to change them to the best third-party article that anyone can point us to.
80 comments
[ 3.4 ms ] story [ 152 ms ] thread- Observability
- Security / Compliance
Seem like they're plays by the founders to get acquired by bigger firms (either for acqui-hire or just to augment their own "observability and/or security" offering).
I would be very curious to see a company that really make it big and remain independent in this area. It seems to me like DataDog has been the only successful company in this area, with some other honorable mentions (e.g. honeycomb.io, sentry, sensu etc.)
Monitoring is a brutal industry, as engineers love to build monitoring products. There is a long tail of hundreds of monitoring startups catering to every wim that are backed by only a small handful of engineers.
On the other hand, integrating a monitoring product into a company large enough to pay for the product is extraordinarily expensive. The DevOps team might want to have end-end visibility, but they can't add the library support required to make the deal work. Supporting one companies specific libraries is unlikely to pay off for other companies.
When it comes to distributed tracing, the value proposition depends on convincing every dev team in the company that this is the best solution for monitoring.
Larger companies have an easier time "cross-selling" these features, and can sometimes win deals by simply having the capability even if its never used.
If your value prop is 'we're stable and you can trust us', and your buyers are 'large enterprises that don't like to take risks', your biggest impediment if you have a great product is that you aren't part of a company like ServiceNow.
What's the saying? "The battle between every startup and incumbent comes down to whether the startup gets distribution before the incumbent gets innovation."
Even harder when you need enterprise proof-points...
Proofpoint (PFPT) is a big security / compliance company. Market cap is about half that of Datadog.
Crowdstrike (CRWD) is another big one. Twice the size of datadog.
And this paragraph says nothing (what company isn't already digital?):
Companies are betting on going digital in order to thrive in the 21st century, but the transition is often challenging to navigate," said Pablo Stern, SVP & GM, IT Workflow Products, ServiceNow. "With Lightstep, ServiceNow will transform how software solutions are delivered to customers. This will ultimately make it easier for customers to innovate quickly. Now they'll be able to build and operate their software faster than ever before and take the new era of work head on with confidence.
If I remember correctly, that POS had both close and exist on the same screen performing different functions.
Exit would close the window, and also break your workflow if you had HPSM open in two browser tabs
As competitors like ServiceNow came onto the scene, they tried to create a better web client, but at least when I was still close to that product, they never quite got there. My two final HPSM projects were:
1) Building a mobile / blackberry compatible web client from scratch (HTTP/SOAP) so management types could see/approve change requests on the go
2) Migrating to ServiceNow :P
It would not surprise me at all if they had some kind of ActiveX thing going on with their web client...
So yeah, that was my petty grievance--no doubt I or the administrators could make this more pleasant. But I've never heard a single soul have anything favorable to say about SNOW. I don't know if SNOW's success is attributable to an unusually impoverished competitive landscape or if there are hidden factors (it appeals to IT's innate enmity toward the rest of the organization?), but it is surprising that SNOW survives at all.
Orgs get in trouble by trying to solve internal process inefficiencies and related people problems by throwing ServiceNow at it. The "digital" version of a shitty process is still a shitty process.
> But I've never heard a single soul have anything favorable to say about SNOW
I've been to the conferences, and there are plenty of folks who love the technology stack, and can do powerful things with it. But I suspect that most people experience the product as an end-user, and that experience is 100% dependent on the team that owns/implements the platform. There's a big difference between "We have a center of excellence for automating manual processes, and ServiceNow is part of that solution" and "We brought ServiceNow in to solve world hunger" .... ("is this thing on??...")
To understand the popularity of ServiceNow, one need only look at the legacy products it displaced: BMC Remedy, HP Service Manager, CA Service Desk, etc. ServiceNow may not be sexy by modern standards, but gives large enterprises a level of flexibility and agility that was never possible with the legacy vendors.
This analogy doesn't work here IMO. Indeed, in many respects like reliability, a Ford, Toyota or VW is actually much better than a Ferrari. They're just very different brands: one set are mass market brands and the other ones are exclusive luxury brands, and they have different features that appeal to each.
But Ford, Toyota, VW cars are certainly not "crap", and e.g. Toyota is renowned for the quality of its vehicles.
Think about regional banks, governments, utilities, oil & gas, manufacturers, field engineering services, healthcare systems, ... These aren’t exactly customers who can just use Slack instead.
Licensing cost seems to be way worse, creating a situation where users can only see their own tickets and not those of co-workers.
If your org is not that large or if it can't allocate a semi dedicated team, other products will probably have better defaults.
You take the same old software, with the same old processes, the same old salespeople and sales approach, and you make a new one that is a little shinier and (crucially) freshly made. Because you used your same old connections and investors to fund the development, you have a much better product than whatever cruft HP Service Manager, CA Service Desk or BMC Remedy have been pushing, simply because you haven't been fiddling with a product originally written in C++ in 1998 for 20 years.
Furthermore, your solution is comprehensive enough that it covers most of the existing functionality of the old products in the space, and it ticks all of the same compliance checkboxes too. So there's very little career risk associated to buying it. You even took the customer to the same old restaurant you used to take him when he was buying HP software from you!
Another popular example of this strategy is Zoom. What is Zoom? GoToMeeting/Webex, but new.
On the lower end of the market, there is more entertainment, because not everyone that had legacy products want to pay through the nose for the pleasure of dealing with an army of Service Now suits. This is where companies like Freshworks, Atlassian and ours (InvGate) come in.
Source: founder of InvGate, mid-market player in the same space
If we end-up in a similar discussion after 10\20 years, your comment would still be valid because Chatting, Operations software are part of very basic needs of any org. So, at that time, your comment also would be - "old, but new" :-)
It's a cycle and it goes on :-)
> Another popular example of this strategy is Zoom. What is Zoom? GoToMeeting/Webex, but new
Zoom is drastically better than WebEx and GoTo*. It actually works cross-platform and has much less hassle.
There is value to that, and going all the way to IPO by succeeding against second system syndrome for 10 years straight is no small feat.
This is as opposed to other companies that actually change the way things are done (i.e. GMail is not "Outlook, but new").
ServiceNow covers a wide array of our business needs without having to buy into a number of different SaaS platforms.
From what I've seen, it's much more feature complete and advanced than the competition. (Specifically, procurement)
Is it a PITA to use? Yes. But it forces best practices and the main people who interact with it will be power users anyways.
Like others said, we tried hard to not have to interact directly with SN, adding scripts and integrations into other systems.
Sold to Execs and Managers, not users.
For example, the IT Operations Management suite includes tools to collect specifications about devices without installing an agent, aggregate alerts to better assess outage impacts, and automatically generate topology maps by analyzing network traffic. It's a wholly different product that justifies some weird design decisions in the Incident/Change system, but it can help explain where Lightstep falls in the ecosystem.
Source: ServiceNow ITOM consultant who has seen customers fare better when they look beyond Incident/Problem/Change/Catalog
https://webdemo.cloud.invgate.net/
The people buying it aren't the same people that have to use it.
There simply is no good product in this space, at least not for enterprise clients with 1000’s of users and even more regulatory and internal requirements.
Servicenow occasionally does a spinnywheel but 99% of the time I have loaded the table I need in less than 5 seconds.
Plus:
- I can open it in multiple tabs. Not the shitty in-app tabs with unclear labels that take forever to switch, actual browser tabs. Other solutions give you a session collision if you open two tabs. In SNOW you can middle click anything.
- The search is decent.
- I dont have to click a goddamn 'form fill' button on every single field. If your form designer isn't braindead you can tab you way through the form without needing a mouse.
- I can make my own templates for common tasks and use them right away without invoking a manager. I can put my favorite templates on a toolbar instead of having to click into a sub-page.
None of these things are big asks but they are things that, as a user, the competition is chronically bad at.
It was dogshit slow, search was useless and painful to use. Most things required a mouse. Tickets got lost. The UI looks like 2006 and the UX is out of a Microsoft text book.
The fact it can be used across tabs isn't really an achievement, but if your bar is so low...
From their front page.. Appears you go back in time without Lightstep.
I have not looked into the specifics, but have a hard time believing that ServiceNow should be special in that regard.
Also, the way items are related is so far behind JIRA and TFS, that there is no competition.
If ServiceNow is the future, i prefer living in the past.
Maybe they will eventually drop JIRA for the ServiceNow agile development module but they (the deployment team that I'm not part of) have many modules to deploy before evaluating that.
If I had to sync tickets I would also wish for a Jira only workplace as I mostly do projects that require development and planning but I am almost sure that the peoples who mostly do customers and employees support would have the opposite opinion.
It's not really in the same ballpark. The decision was certainly financially motivated.
Yikes. That's their slogan? They're not accomplishing this very well.
ADP - payroll software
Slack - chat client
ServiceNow - helpdesk software
Corporate press releases [1] are a bit of an exception to some of HN's rules (such as "please submit original source") because they are suboptimal for curiosity [2]. Mostly they are bland boilerplate, and often are written to conceal as much as they reveal. So we tend to change them to the best third-party article that anyone can point us to.
[1] https://hn.algolia.com/?dateRange=all&page=0&prefix=true&sor...
[2] https://hn.algolia.com/?dateRange=all&page=0&prefix=true&sor...