 Okay, good afternoon, everyone. I hope everyone's had a fun and entertaining DrupalCon. We're at an interesting spot in time. So welcome. My name is Anne Stafanik. This is Vanessa Turk. Hello. And we're here today to talk about Drupal support, specifically on how to stop the bleeding. So just a quick round of introductions. My name's Anne. I founded a company called Canopy Studios. I've been working in Drupal since 2007 or 4-6 back in the day. And I have spent a lot of time doing support in my career. I worked at an agency in Canada for three or four years where they already had a support program that we built out. And then I worked at two other agencies helping build their support program. And now I have my own agency with a team of about 30 people. And we offer professional Drupal support, and we have a dedicated team and a dedicated project manager. So I'm really excited to share some stories today. And this is Vanessa. Hello. I met Anne around the same time as we started in Drupal. So we've actually been Drupal friends for a super long time now. Took a little bit of a different path. I've worked in project management and training clients and really got to love information architecture and trying to find ways of making things easier for people when they were putting together their Drupal site. So at this time, I'm starting to focus on UX strategy with most of my time spent on research and information architecture. Okay. So much like we do all of our client services, we actually first wanted to start off with some questions. So a quick show of hands. How many people in this room currently offer support? Wonderful. And how many of you of those that offer support offer support to projects built by other developers? Hmm. I see some, like, maybe. And how many people in this room are actually looking for Drupal support? Zero. Okay. Good. We just know how to target our words here. Okay. Excellent. We have a bit of a smaller room here today. And what I'd like to do is maybe if we could shout out very quickly some of your challenges that you're currently facing. We're going to take 10. So maybe first one, randomness. Okay. And then balancing your professional services with your support work. Selling the cost of the support. Any other challenges or curiosities? Okay. Inheriting someone's code base. What are your current challenges in your Drupal support offering? Pardon me. Can you repeat that one more time? Okay. Low value request from Haig. And then the one in the back. General chaos. Okay. All right. When the back there? Oh, team motivation. Okay. And one over here? How much time? Okay. Okay. How much time we spend in support? We're just so design support design support. Okay. Okay. Excellent. Excellent. Right. And that's very common, isn't it? When you can't see the code, you really sometimes are unclear. One more. One more. Yeah. Oh, great one. Yeah. Trying to sneak in development into the support queue. Something I've seen a lot. So we're going to try to tackle some of those questions during our presentation. It'd be a little bit agile here today. And if you do have any follow-up questions, we're going to leave some time and space for QA. And then you're welcome to come and chat with us after as well. So most clients, whether you're supporting your own clients or supporting clients from what I call the Wild West of Drupal, we often take in the first stage is really bringing those support clients in and we do this first triage. Just like you're in an ER room, you're in an emergency, you're trying to figure out what's going on, you're looking at what's happening. So today, we're going to talk a little bit about just a little touch on what support actually looks like and then we're going to go into how to make data-driven and value-based decisions for your clients. And I also wanted to say we're going to introduce a client as a case study. We left them out of our slides because they came to us when we look at the types of supports that we often get when someone comes and says, I need Drupal support. We have a whole bunch of different types of things that they're asking for. And believe it or not, one that's very common to us is I just need a themer. I just need a themer. They built my whole backend. I just need a themer. I have these comps. Can you help me get it finished? And this client came to us with that same need. I just need a themer. They're a fairly large support client, so they're a little bit of an anomaly, but they came to us as a hot mess. And the first thing we did is we took an assessment of where they were. They had a whole bunch of stuff done, and there's a whole bunch of things where support can actually fall. So today, well, you know, we can talk about how to actually set up service contracts and how to do audits. We're specifically going to really talk about how to deliver value through this process, and we're going to talk a little bit in depth of how to actually create data to make data-driven decisions. Okay. So a new support client. Most often, if you're supporting your own site, they're usually coming, they like you. They want to stick around. You've done good work, and they're going to continue to work with you after launch. Those that come from the Wild West of Drupal, they're often quite wild. More than anything else, they lack trust. I call this developer burn, and it's not developer burn out, but it's actually, they've been burned by some type of development situation. Their developer might have gone missing. The developer, you know, the team might have been, you know, uncommunicative, and they couldn't find them. And maybe that their project went way over budget without any communication or understanding. They may have stretched timelines. Generally, when people come into support, they lack trust, and it's your job to build that trust back up. And I highly recommend that when you're taking on a new client in support, before you even look at the code, and before you even kind of dive deep into the issues, you talk to the human being that's dealing with these challenges, and you ask about their current situation and how they got here, and why they're so frustrated. Some people end up really hating Drupal after they've been through their Drupal project, and they come and say, oh, just let's just build this in something else, because this is a hot mess. And sometimes we need to do a rebuild. But really when we're looking at helping clients, we're trying to figure out their pain points, and building trust often in client services comes down to if a client has had troubles with their developer going missing in action and not actually communicating, you call that client every day until they're tired of hearing your voice. And I'm serious. A lot of us seriously just work a lot on email. If we've got a client that's just come to us, it's totally freaking out because their developers has left them. Give them your cell phone. Let them call you. They will call you. And if you call them, and they will start like, okay, I'm good. You don't need to call me anymore. I know you're there. Another thing, if the client's always very worried about budget, they often come with like, okay, I have this task list, but I only have five grand and what are we going to do? We really want to help clients make good decisions. And sometimes it's just taking one bite at a time with clients that are really pained from doing some big scale development and they have no money left. Because often when they hit our support queue, nine times out of ten, they're tight on funds because they've just been in some type of fire. So let's now, I mean, support stuff is a really interesting place to work in because I do find a lot of this is about building trust and also creating value. And I do strongly believe, like I really like this quote from Albert Einstein is, strive not to be a success, but to rather be a value. And when you can help your clients and really create value for them, they're there for the long term. So while a lot of people set up support retainers and they expire and they just do, you know, module updates and security patches, we really look at our support offering to offer more than just your maintenance. So there's a couple of key questions we're going to ask and answer during this presentation and one is what's your turnaround time in the issue queue? Support is time sensitive. We're dealing with production. This is not development where we can stall. This is real, someone's real live site. And saying that your turnaround times are really important. Saying that when everything's ASAP, when everything's the same priority, it becomes really hard to prioritize and nothing is important. So here is like really important to do expectation management. We used to try the aspect where we had the development team doing the support and that doesn't quite work. A, support can be sometimes very small so that developer gets onto a new project and now they don't have time for support at all. And B, supports are a bit wild and when it comes up and there's a fire and you actually do need to action but that developer that was supporting it before might be just heads down with their earphones getting their shit done, right? It's hard to pull them out of that and have them have that switching cost to get those fires done. So to have a dedicated support team is really awesome to be able to have those resources available. So in managing expectations around the turnaround time, we recommend something that's fairly industry standard of this hierarchy of chaos. We call it the hierarchy of chaos because we're always trying to bring calm to the issue queue. And when we're looking at this, we're looking at four different levels of trying your support tickets. Ideally, you kind of let the client classify them and while the first one's very obvious and the last one's very obvious, those two in the middle are very muddy. And this is where you're going to have some good client service to be able to identify what this issue is and is it critically impacting the business and help managing expectations with a good service level agreement so a client knows if they submit a low impact ticket that might be done in a week versus if it's a critical impact, white screen of death, sites down, you know, that gets a one hour turnaround time. Typically support requests are two to three hours maximum. Anything that has that four hour threshold is now what we call micro development. Micro development can go all the way to 50, 60 hours, but as soon as you get over really 20 to 40 hours, it really is a true development project because we really need to schedule time, resources, do proper QA and deployment. So these two areas, they're very interesting to kind of work with your clients. It's a bit of trial and error. It's figuring out what works, what doesn't, and then helping clients understand that if they submit everything at P1, all of a sudden nothing is P1. So we really have to help clients understand. Some of the other things to bring calm into the issue queue and it's really like what you want to do is to not be in a fiery mode because that, you know, first off is be responsive. If someone submits a ticket or an email or a phone call, acknowledge it, whether it's even your system that's sending an automated, thank you for your request, we'll be back in touch with you. Or if it's a project manager saying, hey, we received your request, we've put it in an issue queue. It's okay that you don't have the solution, it's totally okay, you just have to acknowledge it. It goes a long way. Another thing that we really strive to do is develop intelligent audits. And it doesn't just mean like running the site audit module or any of those, it's actually creating a template that's standardized that helps you ask strategic questions. And we also try to like think about our client's business. It's not just about the pixels or the modules. It's like what are you building? What are you trying to do? How are you trying to get there? What are your performance measures? How do you know you've been successful? Who are your users? What do they want to do? And who are your administrators? And asking really intelligent questions helps you extract information so that when they start submitting tickets, you can reference that information and say, oh, but wait, sometimes support can be death by 1,000 tickets. And if you're just simply a ticket taker and you don't actually think about the bigger picture, the site can actually become challenged very quickly. So intelligent audits overcommunication as we talked about, and really it's under promise over deliver, which often I try not to set expectations as when tickets are going to get done. We say, hey, we're going to resolve this as soon as possible. So and then if it's taking a while, if there's some challenges updating, hi, we're still working on your ticket. Hi, we've run into some information we need from you. Can you please get that over? And we have all types of audits. Vanessa has a good analogy on kind of like, you know, cleaning out the garage. I'd love this story. Yeah, I love talking to my clients in metaphors because often we don't share the same language. Whatever they're business in, I'm not a subject matter expert in. And often they are not, they're not super technical. So we try to kind of reach and bridge the gap through some metaphors. And sometimes when they just, they get so excited about everything they have to do, they don't really understand how long things can take when things are messy. So it's like cleaning out the garage. Like all you need to do is the keep and the throw away pile and you'd be done by dinner time. Three days later you're still sorting through baby pictures. And it can be a lot like that with a long list of items because some of the items are with the contractor's story, things like, for example, if I have a house and I have a new appliance and I just need somebody to come in and cut a hole in the wall and stick it in another socket. I just need to get someone in there to just carve it and splice some wires and do that. And that's not something that a responsible contractor would do without asking some serious questions of what kind of appliance and what kind of wiring and what kind of load we're talking about. And this is what we're talking about with audits is stepping back and saying, okay, what are we trying to do here? I'm not just going to install this or move this over without asking some questions. And so spending the time on the task is not the same thing as estimating a task or estimating a solution. People will come to you with symptoms. People will come to you with ideas. And often they've already moved into the how instead of the what. So what we need to do is to spend enough time in the what to understand the difference between a symptom and a solution. So I don't know if we want to talk briefly a little bit about the, yeah, if you want to introduce the use. So our lovely client that came to us was just needing a theme or ended up having three developers working on it full time for three months. What they thought was 80% done was more like 10%. And it was a very challenged situation because they had designs that were not built by a person who understands the web very well. And they themselves came from a background where they like to push some pixels. So we went into the mode of helping them as best that we could. And they had all of these like, okay, we have this user registration form. And it was to take in caregivers. And caregivers were going to be found by care seekers and we were going to have this matching situation and search and a whole bunch of different, like there's dashboards and messages and a lot of social activity. And it was very challenged because they were saying, we don't really care about mobile. We really wanted to do desktop. We want it to be perfect on desktop. We want all the links to be this. We want this to be here. And we really fought Drupal a lot to help accomplish some of their design goals. And even though that we said, hey, based on our experience, we think that if you really want to get live, let's build this. Let's get it live. Let's test it. No, no, no, we just need this to be perfect. We can't launch. And literally while they were so crunched for time, they delayed their project in extra eight weeks to get the pixels perfect. And as much as we give our professional advice, sometimes we kind of have to let the client make a bit of a mess so they can understand. We're just like, you kind of have to let a teenager do their thing so they learn, right? Much like with client services, we tried to do our best to inform them and they went through the process. And in their form, they had 57 required fields. The original design was very hard to implement. So we negotiated a new design and we were able to implement it in a way that was much easier. And we were still like, hey, guys, we still think this is too much. No, no, no, we have to get this information. We have to get this information. Were you having phone interviews with these people? Yes, but we still need all the fields required. And what happened was we launched and we found that almost 70% of their traffic was coming from mobile users and they were having one heck of a time filling out this form. Literally, they had spent, you know, tons of time, tons of money, and they spent so much time making this pixel perfect for desktop saying, no, we know that mobile, they aren't. You know, they really didn't do the research. And sometimes when clients, they come, you know, you go through this process and sometimes when you get to production, you have to be ready for this fire that's going to come up because they're finally going to see that it's not working. So this is where we looped in Vanessa and we had her come in. So one of the things that we first think of when we think of data is Google Analytics. And it is helpful because it's going to tell us things like, yes, we do have mobile users and you can certainly set up all kinds of events and funnels and things to give you some more information. But I think we forget sometimes that we can design our own studies that will give us qualitative and quantitative information based on some questions that we're asking the clients a little bit more about. So it can be as simple as, and this is what happened, while we were having the UX kickoff call, I said, hey, is there anybody you can call that you were meaning to get on the phone and try to figure out their form? And we literally had the client call one of these applicants and had them on the phone with us. She said, oh, our team's just listening in because they're trying to make the website better. And it was astonishing how confused the user was, how she thought she had submitted the form, how she didn't realize some of the fields had gone missing. And that was enough of data, just that conversation, to really have that gravity we needed to be able to recommend further UX work. And what came out of that was a study where we observed using a tool, the users on both desktop and mobile, but with a focus on mobile, and actually watched videos, screened videos of actual users filling in the form. There's some privacy tools that can be implemented, if that's a question that you're thinking of right now. But it really is, it's mind-bogglingly helpful to actually see people getting confused, and it's quite painful for the client. And it really does generate, and one of the questions was, how do you deal with people who don't understand the value, who don't want to pay the cost? If you get them to do a little math of all the people that are given up on that form, how much money is each user costing them by dropping the form there's some money that they can spend on improvements? Yeah, I think I already kind of went into this. It was really interesting to get live, and that we knew, all the developers, we were like, we knew this was going to happen. And we don't want to say to the client, well, we told you so, but data speaks a lot. And if you are with a new support client who has a lot of trust issues, put analytics on their site if they don't already have it, create some good events, put them on the different parts, and watch the site. And part of your audit can actually just be watching how users are interacting by using Google Analytics. Google has a ton of free videos and training materials out there to teach you how to properly set that up. If you are working with commerce, this is extremely powerful because you can actually track a lot on the e-commerce side using analytics and get really rich data back to figure out, oh, okay, how do we encourage something? Or they see the shipping cost and we are losing them on shipping, so maybe we need to send them a coupon for free shipping within 24 hours. And the stuff you can do now with Google AdWords and remarketing type stuff, there is really no excuse to not use Google Analytics in support because trust me, when a client has trust issues and you show up with a bunch of numbers to show them and you show them real data as to why their stuff isn't working, but they're all yours. And when you start making strategic recommendations about other things, they go, oh, yeah, they believe in our business, they're going to help us. And that's a really big win. Yeah. What am I mean by gorilla testing? It was like that phone call. Can I just listen in? Can I listen in here from actual users? Can I just have a quick look at your site and think critically through it? Sometimes we offer fixes without actually looking to see from beginning to end the workflow that we're fixing. It brings to mind a story of when I was asked to check the Google Analytics for why people may be choosing to call in online catering orders rather than use the online ordering system. And I couldn't really figure anything out in analytics, but when I looked at the site, one of their hero sliders said $20 off if you call in your order. So I'm a big fan of talking to real people that are using the site. Sometimes that can be difficult to define who that is. With an internet it's pretty clear. With a site where they've already got some users that they're friendly with, that they may be able to contact, that can be useful. But where to test users actually come from? When to use actual site users may be something where there's an intranet or there are people who are power users that you can see that there's different roles that they're already using the site and you can try to get them to walk you through how they're using the site rather than just for asking for insight and improvements, listen for the cues that they are explaining to you. Sites that may have a particular demographic, a site for pharmacists then it would be useful to, if you can, if you have to be looking for test users it might be useful to find actual pharmacists. With the caregivers site we use people who are actual applicants because we were able to make those recordings of real life interactions. There's obviously a lot of third party testing tools and they can be handy when you're dealing with a site that has a wide variety of intended users and a wide audience. They'll let you do a little bit of demographic tweaking but really if it's something that is meant for the general public that's when that can be really helpful. So I think the core of our presentation is really trying to talk about ways that we have found to get a little bit more value into the offerings that we have in support. So before we're saying yes to something, we really want to understand why they're asking for it. Have they had the suggestion from somebody? Have they read an article? Did, you know, something, somebody whispering in their ear that this is a possible solution for them. And I don't want to challenge clients, I just want to gently lead them through to help me understand. So it's not a difficult conversation but I use phrases like alright, so just walk me through how this would work if I'm arriving here and you have this new upload tool and you want to attach files, what kind of files are we talking about and what would happen if they exceeded the file limit and just kind of trying to talk about different ways if things might go wrong if they weren't using the tool properly. So to me it's yes there are emergency situations but it's quite rare that some of the questions that people are asking are true P1s. A lot of the time you're hovering in that P1, P2 area and so there's a question about prioritization. If you're able to talk critically about how the feature that they're requesting or the fix that they're requesting actually impacts their business and maybe some possible work arounds or trying out different ways to think of alternate solutions. I always think it's better to have a couple different ideas and bounce these ideas off than to just be a note taker and a hired hand to do these things. So do we make them mad by doing that? Do we upset them by doing that? Is it about choosing the right clients? I was going to ask you to talk about that briefly. So I think there was also a question about the randomness of it and I do find like there is you are in triage like they are bleeding, it's a mess, we don't even know what's going on, no band-aid is going to stop this so it is taking that list and not dismissing it really is taking it a part of your audit process and when you go through and you look at the code and you look at the file structure and you see what they're doing for their hosting and you do all this, you also take into consideration their requests and you take a look at that and sometimes there will be some discovery work, you know to it and really when you're working with those clients they're going to be intense especially at the first one, so it's a little bit of choosing the right clients, it is gently explaining as the rationale as to why maybe things were built this way because they maybe not understand that why things are breaking because they're not coders, they have no clue what's underneath the hood. It comes back to their perception of what's going on so a lot of it is really an education it's in good documentation and we find like the end of our audit the deliverable is a set of actionable plan and actionable steps for implementation and it's not all to be done upfront. Sometimes we front load the first part of the engagement and we say okay we're going to block 40 hours and just burn through this because this is really important and we now understand why and we understand how things go and sometimes clients want to phase that out over a couple of months so we help prioritize and create a backlog of tickets. Yeah Exactly, supports hard supports really hard and it's also very hard on developers it's it's hard to find developers that want to be in support so you got to make it fun and wild and you know we have this thing like we don't launch any sites or do anything on Fridays you know it's freedom Fridays we really try to instill a culture of like if the issue is clear chill out, you know go for a walk hang out, just have your phone near you for an emergency and to really like you know pay your support developers well because they have a tough job and you know help them by giving them a line an escalation path so they're not just always like front line they can sometimes okay I'm just going to help out or also with team motivation I find that support developers they have so much time in the queue before they get burnt out and they just want to do something new they just want a new project they're tired of banging their heads against the issue queue they really want something new so to honor that and to help them maybe cycle out into the development project or if there's one of those micro development projects that come up make sure that they have proper time allocated so they can really work on it and really supporting them and you know a huge thank you goes a long way and to help to praise them in front of the clients also goes a long way because so often that our you know clients are really good at letting you know when they're mad right and they don't always know or they don't always express their gratitude so by you thanking your team in front of your clients they go oh yeah yeah thanks guys right so it's really important and it's important to have you know people that are not afraid of being a little cowboy in support you know sometimes the stuff's really interesting you know and it's also healthy for you as an agency or a person providing to have healthy boundaries we do our audits not only to understand the code base but to also take a litmus test on how crazy is the code and how crazy is the client and the developer has the ultimate say if they don't want to work on this stuff they can go we can get rid of the client and way rather get rid of the client than try to hold on to some chaoticness and we do have this rule if they come in like we had this one wordpress client that was a whoo it just stressed out a support queue and you have to rebuild or you have to go away and they went what you have to rebuild or you have to go away and they were like oh oh well like I don't know who else is going to support this because we're professionals and this is hard and it's been hard on our team and so they said okay we'll do a rebuild and we said okay we'll bandaid and then it comes in negotiation because still while we're in this rebuild I want to launch this new blog you're rebuilding your site so again it goes back to that intelligent question of why? why are you doing this? does the user care is this going to impact your business in a way that's beneficial to spend the money on something that's hemorrhaging it's awful and if that's the case then we'll help you support it but we need some more realistic timelines from you and helping negotiate timelines and the issue queues really helpful when you have some more intense clients so let's get a little bit more and this is something that we have a balance of e-commerce clients nonprofit clients, professional businesses higher ed and we find that the underlying current from all of that is helping clients make good decisions we don't always know we've I mean we don't know anything about trampolines then you have a trampoline store and you learn a lot about trampolines and the question always comes back again is who are the users and how are they what is the success factors things that are your best friend Google analytics, we can't stress that enough another thing is implement, test, fix test often keep taking small bites run it, roll it out, see how it goes iterate another one is that your email marketing tools MailChimp has a wonderful analytics and you can layer that with mandrill for your transactional emails you can see so much data out of that so we highly recommend using a professional service like MailChimp and you can also get a lot of feedback from social media are people sharing stuff are they commenting managing your social media channels is a great way to actually gain feedback from your users because you unfortunately especially in like when you have like a brick and mortar business, you're most often going to hear complaints and you're going to hear the good stuff right so we're supplying to you know if you have a Google like a business and they have a Google places and someone leaves a comment being able to immediately comment and help those clients so let's let's oh and another thing is a lot of your clients might have analytics but I bet you that maybe maybe 5 10% actually look at them so encourage your clients to also use Google analytics to see what's going on it'll open their eyes okay so why don't we just quickly talk about some tactics for making data driven decisions and not all support clients have the budget or the bandwidth for this type of stuff but it is stuff that you can take in little bite size pieces and just do a little bit every time and we have a situation where we have a retained support agreement where they sign up for 6 or 12 months and they have a block of hours and we chose to do the model of roll over hours not to have them expire and the reason that we chose this is that yes while we're in the business of making money we're more in the business for helping our clients and giving them value and sometimes clients they get busy on something and they don't use their hours so we roll them over and that allows us to take on when we don't have a client that has a lot of money but we have these roll over hours about 6 months into the engagement if they're actually truly quiet because most clients really do use most of their hours but if we do have those hours on bank we can start doing some of these richer opportunities and assistance so this is Vanessa's area here I wanted to talk a little bit more about what I mean by data sometimes it really is very qualitative because people have trouble putting numbers on things larger companies will have marketing numbers but it's really difficult to actually even put a goal together we'd like to reduce our bounce rate or we'd like to increase subscriptions and we don't know by how much and over what period of time and those are the things that really make it a more realistic goal I think we've all heard about smart goals and why they're better than vague ones so I wanted to talk about if we don't really have a lot of hard numbers what might be some ways of gathering qualitative data one of the things that I notice is that clients struggle really hard to come up with ideas of how to make their site easier to use and they come up with ideas that maybe not are not the most realistic or they don't make a lot of sense but they're trying to solve a problem of make my site easier to use and so they'll come up with they'll bring in a symptom you know people are leaving this page or they're not able to find certain pages and then they'll kind of make suggestions and put a button here and highlight this and I have a client that have highlighted a lot of things and the entire site is now highlighted so we have to think of another tactic and it might just very well be redesigned so one of the things that happens a lot is that sites are designed with this sort of in and out kind of navigation where they go down into a page and there's no way for users to move horizontally across information and what I mean by horizontally across information is you'll know it when it's done well sites that you get lost in you're reading an article and there's another one that's really interesting and this is on the same topic and then there's a slide show over here and you just kind of keep going sideways and where it's not done well is when you have a single article nothing of any kind of pathway sideways you have to keep going back to the home page and reorient yourself so I do find that a lot of clients will bring in that sort of site and come up with all sorts of crazy ideas to try to solve that problem so we really need to work with them to understand when they say I really want the menu to be bigger that they don't really mean that yeah they're trying to achieve something so it's for examining a channel if they've got certain language in a brochure and brick and mortar and social media make sure there's consistency across the various channels so users don't get disoriented as they're moving through the website to social media and maybe coming into the store and having a look around so I'm just touching on some of this stuff and then with the horizontal content flow if you can see in analytics that they're not visiting very many pages and the client is reporting that they're not really carrying out some of those conversions they're just not happening then we're looking for a way to provide complimentary content and maybe a better visual focus of how to move through those pages and we really often people just want people to get lost on the site because perhaps they've got some ads or something that they want to provide that keeps users on the site but sometimes it's about just getting people as quickly as possible to a registration or to complete a certain task and those would be treated a little bit differently but that's something you can talk through some tactics on that and really leveraging the various multi-disciplinary members of your team for research, for design for the optimal kind of tools to use on mobile because there's definitely ways to make things easier in technology making sure that your fields are the right fields so looking at that flow horizontally through the site is a big one for me I'd also like to talk them through basic scenarios for the site because there's usually more than one and there's usually more than one type of audience and type of user that we're dealing with so users will come to the site, they will I think these are a little bit backwards yeah, that's where I wanted to start with the site, if they've never been to the site before it's a journey of discovery they're just starting to understand and comprehend the offerings, they're realizing that they perhaps need something that is offered on the site product, service, something like that and then they begin to go through a period of consideration people that are in a consideration mindset are going to need different content in a different pathway through the site than people who are in discovery or power users who may just be in that retention phase coming back to the site on a frequent basis so if somebody if a client is requesting something that is only for one or perhaps ruins another one of these pathways through the site, this might really affect the success of that site so I like to test it through these different lenses to make sure that we're not that we're not screwing something up for another type of user so it's really identifying those needs of the different types of audiences and pathways through the site understanding that they're going to be needing to understand the site offerings and what their needs could be when they're at the site and then there comes usually a period of choosing between isolating and kind of focusing themselves down to if it's an e-commerce site a few products that they would like to choose and there's that refinement that happens that is often a place where people drop off because of various barriers so keeping that in mind and bringing up the idea that there's X, Y axis of people who may be familiar with the site but not familiar with the content versus all the way over to people who are familiar with the content and not with the site and designing for different kinds of navigation versus browsing versus keyword search something that I explain to my clients is that people don't always approach the site the way they approach the site specifically because they've probably seen it so many times that they're not seeing it with fresh eyes and so that it's always good to get somebody on there who's going to be testing it out. Nielsen says there's a lot of people who are familiar with the site and I do believe that because in the past we've had very few people come in and have a look and right away you're starting to see patterns of things that they're starting to bring up so some other ideas for benchmarking with the client is to talk about where they think users may be having some problems once you've talked to the client about that sometimes they have people who are asked either about the site or the products and services and it's surprising to me when the web people haven't talked to the customer service people to find out the questions that their clients are asking them about their products on the site so trying to get some of those clarity levels anxiety levels, distraction levels doing those interviews and maybe it's not an exact science giving them some ratings 3 out of 5 people seem to have that some of the site elements created anxiety for them even just to have that is going to be better when you come back with some fixes you can see what's gone up and what's gone down and where you've made an impact and I want to make sure we have time for questions too so what do I mean by gorilla UX I've brought it up before this is just a simple getting on the phone with people sometimes I have a script sometimes I just kind of talk them through how they use the site how often do they visit what do they look for and it's really fascinating sometimes when they bring up completely different ideas than the client because bringing that to them because it doesn't come from me has huge value because it comes directly from a customer I am able to say this is what your customers are saying and I'm just a third party in the equation if you would like to solve this problem then we have some recommendations around it I'm just going to power through some of these so do you want to come back on here for this one I do find it really interesting that I think it's something like a user needs to visit your site 9.7 times before they are going to take action so there's really importance of understanding how to get them back to your site and in support while we're maybe not doing digital marketing we're really trying to really help our clients make their organization and their business better so we're taking a new approach to support where yes we answer tickets yes we'll get you out of where you are yes we'll help you know why your bullet points are not showing up as bullet points and we'll help with all the module updates yes we're going to do that for sure but we really want to go beyond that with this new model we're really taking in we're looking at the issue we're looking at what the symptom is and then we're trying to make a benchmark as to what that would look like if it was improved find the solution make the tasks and actually when you make the task to find what the acceptance criteria is super important in support to test make sure you have a development and a staging environment and make sure you test and when you do deploy it make sure that the support team is around to support just in case for some reason something goes wrong so we are thinking more even though we do have like a dedicated support team we do leverage the larger team to be able to make good decisions I know that sometimes we have a client that comes to us and says I just need to make my site look better on mobile and I'm like well what does that actually mean and sometimes the first line of defense is okay let's do some crazy mockups and let's just get it to look nice on mobile as the bandaid but then the longer term solution is let's do some user studies let's go dive deep let's do some testing and actually design for mobile but sometimes we do need to take that first response get that bandaid on and then really find the finer solution it is meaning that you when you're creating this culture of offering support in your organization it does have to touch everyone even though you do have a team you need to have people that are able to help out maybe not all the time but it's helpful sometimes when you do get into that nasty problem it's just really tricky sometimes one of your senior devs is going to look at it with fresh eyes and be like oh this is the issue and the guy goes oh yeah that's awesome thanks and then they can power on encourage your user experience specialist to get involved and if you have some of those banked hours or you can layer that on as an additional service offering to your support it's absolutely valuable because they all of a sudden open their eyes to be oh it's not just about my module updates and removing those red errors it's really about creating an awesome site do use proper work tools I mean it was really funny how many of our clients come to us in support and they're doing everything on production with no version control it blows my mind how many sites are out there with stuff that's like with pantheon like that's not even an excuse anymore right they don't you can plug and play on almost on your own and again QA is really really really important this is people's live site so they're way hoppier they're way more excited they're way more stressed out so making sure that you do good QA that's a good point for the unrealistic expectations for unrealistic expectations we have a lot of clients that come in and they go here's my 20 tickets so are they done we just got them okay well maybe tomorrow tomorrow morning are they done did you get it done can we look at it can we look at it it's like come on guys let's give us some time and again it goes back to prioritizing it goes back to setting expectations we find it extremely difficult to estimate in support especially if it's not our own code so we tell our clients we're not going to estimate your tickets we get one in our queue that we know is going to take more than half a day the support developer is going to raise their hand and then we're going to figure that out and it's often this is going to take four to six hours or you know we let them know that even though we do know your site there are some stuff that are going to be unknowns okay so let's I'm just going to quickly read through the list here and make sure we have all kind of our general questions and then we'll take some more questions we got the randomness it's really about prioritizing and helping clients see the value in your support offering versus just being a helping hand and it's definitely been my experience that some people refuse to prioritize so that Moscow version of prioritization or trying different tricks to get them to arrange things in order can help but yeah there's definitely people who after doing that still say that everything's the same level of importance where I think scoping really helps is carving out what you would do in the first week or the second week the cost of support subscription and also we also recommend we have two different types of support tools we have a one that's email driven we use teamwork as our main ticketing tool and we have the ability for the clients to create sprints and they can put tickets in an order and they can drag and drop them and then we also have use teamwork desk so they can send an email if they're not you know if they're just simply answer my request that allows us to have a good that goes to all the support team members there's no perfect tool by any stretch of the imagination inheriting code from others is number one do an audit number two just tell them that everything is your best that you can possibly do really being honest in support goes a long way and we can go into different cost models I do highly believe having a good service level agreement that defines how you're going to work together what response times are we've had a very good success model with a block of hours so people feel like a tangible chunk of work that they're going to get low business value request from high value clients this is where let's say we have large clients and they're asking for hey can you enter my content that's okay I will do that if they really want me to enter their content and pay me a good dollars great sometimes they just really don't have the bandwidth and they need that help and I think that's where it comes to if you do have you know people think you can put juniors in support and honestly that's total bs if you put a junior in support they're going to suffer and they're not going to really know the solutions you know and I feel their pain like support can be hard so training a junior is not putting them in support training a junior is like bringing them into the fold of your organization and giving them good mentorship and you know we have some people that kind of do a bunch of development work and they're like I want to take a support vacation I just want to bang on tickets for a while and like you know just get I want to build an ad system in a day and release it like you know there's there's elements of where you can create this culture of fun within it too many sites and too many states maybe the person for that could stand up and speak up is the too many sites and too many states we expand on that a little bit yeah yeah I mean that's a tricky one balancing priorities is with with clients in different states is helping again manage expectations some of our clients in smaller blocks of time we use them at the end of the month so we actually keep we retain we hold on to their hours and if they have no emergencies we then help take care of some of their backlog tickets because if you go with the client that he maybe has five or ten hours a month and they go okay I want it we are new month I want to do all this and they burn through their ten hours in the first you know week and then they have a true emergency they can't really afford it because they've already spent their allocated time so really helping like help manage expectations set a monthly meeting do a review have a backlog and have know that your p2 p3 p4 ones are stuff that can be taken care of and blocking your team's time to facilitate that so team motivation I also find that with with with support sometimes people developers are asked to estimate on something that's not estimatable so something like this site is could be easier to use estimate please on how much time it will take to improve and and something like that is is is really a problem not really a solution and even if I had a solution it would probably be 12 items which then need tasks which then could be estimated so a little bit of sanity goes a long way for developers who sometimes have unrealistic things asked of them yeah having a project manager they can which too is really helpful that they can vent to and the project manager being like a little bit wild and have a little fun and send some youtube videos and you know that you know so maybe we should do open questions yeah sure do you think there's anything else on there that yeah do you maybe want to go to the mic so we can hear your question sure we'll repeat it right we have a base level there's a minimum 10 hour commitment to every month oh yeah the question was is how do you know how much time to allocate to a client and how do you estimate the number of hours they're going to need we usually say our base level is 10 you can go 20 30 40 50 and we can reevaluate in two or three months and adjust if it's too many or not enough we do monthly usually yeah and then we do have some people that come in and say I want it you know 20 hours a week from you guys and we and then we try to maximize that yeah okay yeah so the question is you have multiple versions you have multiple sets of hours some have rollover hours some of them you know it's quite chaotic in terms of you know having all of these different clients and smaller chunks we find that we do kind of two weekly management scrums it kind of looks at everything we use some tools like we really like using Trello and having just one you know one board that has all our support clients on it that we see who's assigned we see all of it in one it's really helpful and we have morning scrums every morning at 10 a.m. all the support developers get on scrum and they do a 15 minutes down of what are you working on today what did you work on yesterday where are you blocked and we resource probably only one or two to three days you know sometimes we're like okay the support developers also doing this new build because they want to do something inspirational so we know that they're down 20 to 30 hours a week and can't take on more than a few tickets a week so it does take a lot of like planning and talking to your team and having that like daily touch point with all of the support developers on one call helps them also ask questions and helps them you know okay I'm working on this oh I've you know I can help there and having you know the project manager has a lot on their plate you know so to try to keep them clear from other things it's really helpful I mean it's a very challenging role support is hard you know so any other questions yeah I mean support developers are almost like unicorns in a way they do or tend to be multi-faceted and I do find a lot of the people that you know we usually pair a front end and a back end person to each client and then they have backup developers and then we have those specialists that will come in on that ninja basis okay you need some performance work we're going to allocate 20 hours we're going to pull in Tim for that because he's our support he's our performance expert so I do find the specialists much do really well in the kind of the green field the big builds and then if you can help plan because most stuff that when you're engaging with a specialist it's usually not ASAP emergency let's do it you can actually say okay in a week's time we're going to look at doing all of your performance optimization and sometimes it's relying on those those you know your crew just to be able to have a channel to ask the specialist questions and then and then that developer gets to learn what's more specific and unique and kind of you know oh I'm going to learn all about high performance and I'm going to rely on Tim and Steve and they're going to give me advice and they're going to actually review my work and they're engaging because they get to learn new things okay we probably have time for one more question yeah oh yeah no they always have to choose a 10 hour month package which includes security updates clients never just they just they will say oh I just want security updates in module and then they're going to like oh I broke it I'm going to build a blog can we do something it's like well then you're constantly you know say okay well you need 10 more hours or 15 more hours so we do a minimum 10 hours which includes quarterly maintenance where we go through our sites on a quarterly basis and we update the modules and we do the security patches unless we have something that comes out that needs to be done right now then we sit you know essentially we let all our clients know hi there's been a release we need to take care of the next update to hold all of our sites get updated fortunately with tools like Pantheon all of our Pantheon sites is like click click click click click click click and they're done and the rest we cycle through but usually with security updates and modules they usually can be done quarterly and it's really hard just to sell that I mean it's easy to sell that but then the client wants more clients are never simple if I were so simple then I don't know so support wouldn't be around wonderful so ooh there's my rocket ship oh yeah see there's a rocket ship so I want to thank you guys so much support is about 50% of our business so we've done a lot of it so please come and talk to us we're with 508 we're also giving away free audits we're going to give away a bunch of those so I have some cards up front if you guys want to grab them and you just go to our website and you can enter there so thank you guys so much