3 Replies Latest reply on Apr 2, 2014 4:36 PM by extrands

    What's taking so long?

    extrands

      Before I start my rant- I want to say that I love SolarWinds products, and am an ardent advocate for anyone considering network management software to take a hard look at Orion.  It's good stuff and generally makes my job easier. New features are being added all the time to make a great product even better.  SolarWinds solicits comment and feedback and I have personally seen requests that I have made incorporated into the product.

       

      Soapbox time.  Anybody listening?

      /rant on

       

      I'm wondering what it takes to get a feature request incorporated.

      It doesn't seem to be time related, as of the top 10 NPM feature requests, all of which were submitted in 2012 (its 2014 for the chronologically challenged), only 1 is tagged with the coveted "what we are working on tag".

      It doesn't seem to be popularity related, as the top up voted request has been there for 636 days and hasn't been implemented.  mavturner made a  comment  last year (Jan 20,2013) that because of the high response it was being moved to the top of the list. Still- 15 months later- nothing.  The top of that list must be really something.

      It doesn't seem to be related to shame- because rants like this pop up from time to time in looking over the requests and still nothing happens.

       

      So, what does it take?  Are our requests only seriously taken into account when they just happen to coincide with SolarWinds own internal feature enhancement list, and credit can be taken for "listening to our customers"?  How, exactly are enhancements decided on in the first place?  Customers clamoring for them doesn't seem to be the impetus you would be led to believe- so where do they come from? Why do customer requests not receive more consideration?  I understand the process to a point.  You have Patryck, a developer, who can only work on a certain number of things.  So you have an internally generated enhancement, and a customer generated enhancement.  Which one wins?  I know which one does- but I don't understand why.  It can't be a coin toss, law of averages says we should win a few. Breaking out Occam's Razor, the simplest explanation is true- internal enhancements are preferred and the customers requests can wait- cynically called- "we've already got their money".  We are told it might make it in the next release- but you can't hold us to that. 

       

      How about this for a feature request.  I request that SolarWinds PM's review the customer submitted feature requests on a periodic basis- say every month or two- and rather than letting them stick around for ridiculous amounts of time (636 days!!!), just close the darn things out, tell us the truth- you have no intention of including it, and we can all get on with our day.  The way it is, you are blowing sunshine up our collective posteriors. I realize the difficulty in a public company making commitments and potentially not meeting them, but we need some realistic time-frames.

       

      /rant off

       

      So- perhaps useful responses from the PM's would be giving us some better insight into the selection process, helping us understand what makes it, what doesn't and why.  Give us a better idea on the timelines involved.  You still have a great product, good customer focus and service, but this is an area you can definitely improve.

        • Re: What's taking so long?
          chris.lapoint

          Hi extrands, we do appreciate candid feedback, so I'm really glad you took to the time to raise the visibility on this issue, which is at least partly a lack of context (our fault for sure) that I'll try to correct now.  As you're well aware, a sizable portion of the top feature requests in ideation are related to the Orion alerting engine. It's really not surprising given the fundamental nature of that capability in NPM, SAM, and all our Orion platform based products.  And as we thought about how to tackle all of these one-off feature requests, it became clear to us early last year that we'd only deliver fleeting happiness for everyone by continuing to address them one by one. The right answer, as daunting as it might have sounded at the time, was to take on alerting in the same way we had taken on web reporting.  We needed do right by our customers and product and revamp the alerting engine, move it fully to the web, and make it dazzle you with its simplicity and ease of use, all while retaining all the goodness of the Advanced Alert Manager you have loved to love or loved to hate.  This IS what we're working on.  When Rob Hock, PM for the Orion platform returns from Cisco Live Australia, I'll make sure he loops you into those conversations.  It's still early, but I think you'll be really excited about where we're heading!

           

          With respect to the other feature requests and generally how we prioritize,  I'd like to clarify that only a very small fraction of our roadmap is driven by what you would consider internally-generated enhancements and most of those are related to technical back-end issues that we believe will ultimately affect customers or community integration.  The overwhelming majority of feature requests come from customers, whether they be prospects, existing customers, or non-customers.  We have to listen to all three and as you've noted above, the prioritization exercise is never a fun one.  We'd love to give everyone everything they want all the time, but if we could that, my team and I would likely be out of a job ;-).

           

          I'm going to have Francois Caron, who runs our NetMan and Orion platform product teams, chime in as well on what we're working on.  As he'll outline in his response, we're definitely listening, so please keep asking.

          1 of 1 people found this helpful
          • Re: What's taking so long?
            fcaron

            Hi extrand,

             

            Thanks again for your candid feedback and your simple question: anybody listening?

            Yes we are listening, the 100's of questions that we answer every month on thwack prove it, but it's not really the point: listening without acting won't do any good.

             

            Take the most voted NPM Ideas and let's see how much we are acting on them:

            1.Unmanage reason (like shutdown tracker in Windows 2008)                      
            2.Silence Alerts While Still Monitoring                                                             
            3.If...Then...Else statements for Alert Manager                                               
            4.Undeleting an element (node/interface/volume etc.)                                  
            5.Mandatory Custom Properties
            6.SNMP Trap & Syslog Rules Overhaul                                                                    
            7.Granular Node Management Rights                                                              
            8.Node Outage Duration
            9.Test/Development Licenses
            10.Network atlas , link color, description reflecting events,alarms, utilisation
            11.Custom Properties for Groups                                                                        
            12.List layer 3 information for interfaces, including VLAN interfaces
            13One alerting engine
            14.Multiple IP Address polling support for a single node
            15.Orion Next Generation User Interface (UI)

             

            • 3 have been delivered (or have improved) already (#10, #12, #15)
            • 2 are being worked on and exposed in the "What Are We Working On" post series (#6, #13).
            • The majority of others relate to Alerts and as Chris alluded to above, we know that Alerts in general is probably the primary use case for Orion users.
              As you know, we can't speak too much about what's coming before a certain stage, but about a handful of these Alert-related enhancements are being actively worked on.
              Again, we know the Alert module in general needs attention and I'm confident that at least a few of these 5 will be delivered in the next releases of NPM and Orion-based products.
              We know it seems like a long time, but in the case of Alerts, this time is just a reflection of the attention we are putting behind our rework of this critical module, and not a reflection of SolarWinds ignoring the input of our users.

             

            So, I'm confident to say that we have, or are acting on half of the above list, or more.

             

            Another way to look at things is to look at the last NPM version released, NPM 10.7, where we highlighted the features that were actually coming from your Ideas.
            It's about half of the release being directly driven by "Ideas".


            Also, don't forget that we take ideas from different sources than the feature request/Idea forum: the 1000's of other thwack blog posts, support cases, 1:1 conversations with users, usability interviews and research... all these contribute to the input we get from our users and prospects, that we need to prioritize.

             

            OK, one more. You are not the first one to mention the "SolarWinds own internal feature enhancement list"; some even suggested or implied a "secret agenda" or strategy that we might have, in relation or not with vendors and some partnership we would have...
            Let me be clear about the "SolarWinds strategy": it's pretty simple: we listen to our customers and we deliver as much as can of what they need NOW.

            Sure, we also have a mid-long term strategy and thoughts, about what the future will be and what our portfolio should become, and this has an impact on our roadmap, which might not always seem obvious and driven by the immediate community needs.
            What company does not do that?
            But unlike many competitors – the big-4 for ex. and many others - we are very conscious that in our industry, the future can change very rapidly, and we are humble about our ability to predict for sure what IT management will be in a few years. The result is that we don’t let this distract us more than what’s reasonable, from delivering what you guys need today.
            I've followed many competitors’ roadmap (and worked for a few of them) and my sincere opinion is that basically nobody touches SolarWind's ability to listen and deliver what the community needs.

            We are not perfect, but by listening to passionate reactions from users like you, we'll continue to improve.

             

            Alright, almost done. 
            When we decided to introduce the "Ideation" voting system in thwack, a while ago now, we were conscious of the risk: by making things more transparent, we were potentially making it more obvious when high demand features were not delivered fast enough.
            We knew it could sometimes generate frustration, but after a short internal debate, everybody agreed that this was a "good risk" to take, because we owed this transparency to our huge and loyal community of users.

             

            I’ll let Rob, who looks after NPM and Core, elaborate more if he wants to, but in the meantime, let us know if you want to talk.

            Rob and I would love to have a 1:1 exchange with you on this, or any other topic related to our products.

             

            Thank you for your interest and continuous support!

            1 of 1 people found this helpful
            • Re: What's taking so long?
              extrands

              Thanks chris.lapoint & fcaron,

                   I think we all appreciate the feedback, and we do all know that you do definitely implement some of our ideas- something I assure you we do not take for granted.  Where we get frustrated is when popular ideas languish for long periods of time without PM comment, or PM comments that don't really say much.

               

              Understanding that you can't implement them all, some are actually contradictory of one another, we do want to see some evidence of interest and progress in our feature requests.  2 years, frankly is simply too long, regardless of your development process.

               

              We understand in the wonderful world of litigation that you must be cautious about statements regarding features in upcoming releases, but there certainly has to be a way to keep us better informed and process the requests in less time than it takes to go to Mars and return.

               

              In all seriousness- what you are telling us is "I understand what you want today, we'll have it for you in a year- or two". Every single feature request is made because we, your customers, have a need that we are trying to fill for our own customers.  We pay for the software, we pay for the maintenance, and we rightly believe that when we identify something that is popular (as evidenced by thwack voting), and when we are told something is "on the list" that we should see it in a reasonable amount of time. I define reasonable as months, and not years.

               

              Again, as I've said before, I still think you have the best product out there, and I know you are working on making it even better.  You do respond to our requests and add features that are important to us.  Keep up (and speed up ) the good work.