Bummer! This is just a preview. You need to be signed in with a Pro account to view the entire video.
How to Moz Lingo: Cross-Team Communication When Crisis Hits14:42 with Carin Overturf
Mozzy does not always mean bright and shiny. Sometimes things go south, and it's these times when good communication across all teams, technical and not-so-technical, is critical. Carin brings the tactics she's learned about effective crisis management after surviving a few storms as a technical manager on the Mozscape team.
[MozCon] [Carin Overtuf - How to Moz Lingo: Cross-Team Communication When Crisis Hits] 0:00 [? music ?] 0:04 [Carin Overtuf] Hey, everybody. I'm going to switch it up a little bit. 0:08 Those were some exciting talks this morning, but now we're going to switch to crisis, 0:10 which can be a little more exciting, I guess, maybe. 0:14 Anyway, I just recently switched to the product team. 0:18 I've previously been the engineering manager for big data, 0:21 and I was a project manager for them as well for the past 2 years. 0:24 Big data works on Fresh Web Explorer, Just-Discovered Links, Mozscape Index like Phil was talking about. 0:27 But before coming to Moz, I was actually a developer and a manager at AT&T Wireless 0:33 for a front end rich media team. 0:38 So I've worked across a lot of teams for the past 6 years, 0:40 and I've learned a lot about communication. 0:42 One of the biggest things I've learned is that making sure that you're keeping good communication 0:45 across teams in the difficult times is really, really important and really hard. 0:50 Cross-team communication can be tough even when times are good. 0:55 Engineering, product marketing, we all kind of talk different languages. 1:00 Engineers, they really love the numbers, they love the facts. 1:04 Product marketing, they love words, they like describing things, they like talking. 1:06 And when you're in meetings with the engineers, they can get a little frustrated sometimes 1:11 because they just want the facts, they want to go build the cool stuff, they just want to get it done. 1:14 So when things are burning down and things are going to hell, 1:18 how do you keep up that good communication across all your teams? 1:21 Today I want to share with you guys some of the lessons I've learned along the way. 1:24 First of all, I think we all try and practice good etiquette in our day-to-day lives, 1:28 especially when it comes to email. 1:33 Some of the things just in general I like to keep in mind 1:35 is to make sure that you're using your subject line as a headline. 1:37 You want to set the expectation for people when they're going to open up an email. 1:41 Like, is this urgent? Is there a deadline associated with whatever you're emailing them? 1:45 Let them know, like, do they need to review something by the end of the week, 1:49 or is it something that they can put off, like unicorn cats or something like that 1:51 that they can read whenever they want. 1:55 But make sure that you're setting that expectation in the headline 1:57 so that that person who has to read that email knows exactly what's expected of them. 1:59 Another one to keep in mind—and this one I have a really hard time with— 2:03 you want to keep your emails on topic as much as possible. 2:06 Sometimes you get so busy, you've got a thousand things going through your head, 2:09 you want to just dump all those thoughts through an email, save yourself time, get done with it, 2:12 but that sucks for that person who has to open up that email. 2:16 I'm a big fan of the getting things done methodology, 2:19 so my inbox is like my to-do list, 2:21 and I'm going to go through all my email. 2:23 If I can do it in 10 seconds, I get it done. Otherwise it goes on my to-do list. 2:25 If I get an email of 5 unrelated topics that I'm going to have to synthesize through, 2:29 it's going to take me a while, I definitely put that on my procrastination list 2:34 and I usually don't get back to it for a while. 2:38 So just keep in mind if you want a quick response from somebody, 2:40 try and keep your emails on topic as much as possible. 2:42 And the third thing to always be aware of is your tone, which I'm really bad at this sometimes. 2:46 I love sarcasm, but it doesn't always translate that well with email and instant message. 2:49 I think we've all probably learned that at some point. 2:54 So when things are going to hell, that's when our communication skills really get tested. 2:56 You can pair code, you can test your software as much as you want, 3:01 but inevitably, you're going to end up with that stupid, obvious bug 3:04 that ends up in production that nobody sees until you have 20,000 paying customers having access to it. 3:08 So how you as a company deal with these types of situations 3:14 is what's going to set the good companies apart from the exceptional companies. 3:18 And I want to share with you guys today 6 of the lessons that I try and keep in mind 3:22 in these types of situations. 3:26 Number 1. Everyone Makes Mistakes. 3:28 It's going to happen. 3:30 When mistakes happen, pointing fingers and placing blame is not going to get you to solutions any faster. 3:33 Having a culture that values transparency is a huge advantage in these types of situations. 3:38 It creates a really good feedback loop so that your engineers and your team can jump in, 3:44 start investigating, and start trying to solve the problem without that fear 3:49 of blame or getting called out for something. 3:52 I know a lot of you guys have been our PRO customers for a while. 3:54 You might remember the crawler outage that we had about 2 years ago 3:58 where our crawl diagnostics, we lost a whole bunch of campaign data. 4:02 I was a project manager at the time along with Thomas and Bryce. 4:06 We were hanging out on a Friday night, like 7:00 at night. 4:10 All of a sudden we see all these alarms going off. 4:14 We jump online and we notice that your crawl diagnostic data 4:16 in some of the PRO app campaigns was just disappearing. 4:20 So we jump on and we investigate, and it didn't take very long to figure out 4:23 that there was a misconfiguration in the way that we had set up the database 4:27 to store all that campaign data. 4:30 We had set the database up in the cloud on Amazon Web Services 4:32 on these machines called Spot Instances, 4:37 and they're a really inexpensive, great way to run machines in the cloud 4:39 unless you're storing important data on them. 4:44 They're really cheap when demand is low, 4:47 but when demand gets high, you get outbid for them and the machines just disappear. 4:50 So we were running into the Christmas, the holiday season, 4:54 and we started getting outbid for the machines, 4:56 but we didn't have the proper monitoring on the machines, 4:58 and we shouldn't have had the data on the machines in the first place. 5:01 So we figured this out pretty quickly. We knew what was going on. 5:04 And me being the project manager, it was my job to jump online 5:06 and let everybody know what was going on. 5:08 So I had to email our help team, our community team 5:10 as well as the managers, the E-team and Rand, 5:14 and let everybody know what was happening. 5:16 And instead of the E-team jumping in and being like, "What the hell?" 5:19 "How could you guys do this? How could this happen?" 5:22 which would be a totally fair reaction, their first reaction was, "What's the customer impact?" 5:24 "How can we get this fixed as soon as possible? And how can we communicate this to our customers?" 5:29 And so since we don't have that culture of blame and finger-pointing, 5:34 that allowed our engineering team to jump online, try and fix this, 5:37 and it gave Bryce the freedom to get this blog post up so we could tell everybody what was going on. 5:40 I think he ended up getting this up at 2:00 a.m. on Saturday morning. It was a rough night. 5:45 Number 2. Have a dedicated email distribution just already set up. 5:50 You want to include the people who are going to be directly responsible 5:55 for maintaining your quality products and communication on this distribution. 5:59 Don't include your entire engineering team, don't include your entire marketing team 6:02 if your marketing team is that big. 6:06 A lot of times they're going to see these emails come through 6:08 and they're going to get really used to ignoring them if it's not affecting their team. 6:10 But as soon as you know what team is involved, loop them in right away. 6:13 And if you're not sending them a bunch of extraneous emails, they'll see these kinds of emails come through, 6:17 they'll know they need to jump into action, and they'll know to prioritize these emails. 6:20 Number 3. Clearly define what constitutes an emergency. 6:25 Making sure that everybody has a clear definition of what an emergency is 6:29 will create that urgency that you need when these types of situations come up. 6:33 And for you guys, what defines an emergency is going to be different, 6:37 but for us, we settled on this. 6:40 If our community or our help team sees 5 emails or 5 tweets come through in 24 hours about the same issue, 6:42 they know to write an email out to the engineering team, be like, 6:47 "Hey guys, this is what's happening. Can you guys jump in and investigate what's going on." 6:49 On the other hand, if our engineers start to see any operational issues 6:53 or something going on with one of our features, 6:57 they know to jump online, send an email to our community and help teams that they can tweet, 7:00 and let the help team know that, "Hey, you're probably going to see an influx of tickets." 7:03 "We're investigating the issue. Here's how long it's going to take until it gets fixed." 7:07 But the most important thing to remember is to hold that definition up across all teams. 7:11 You want to make sure that every team has the same definition of what that emergency means. 7:16 That way, any time anyone on any team sees these emails come through, 7:21 they know to prioritize them above all the other hundreds of emails that they have in their inbox. 7:25 Lindsay Wassell, she's one of our associates on the East Coast, and I don't think she even knows this, 7:30 but when I was an engineering manager for big data, 7:34 I would wake up every morning, make my coffee, 8:00 or 9:00 in the morning, 7:37 and go through my emails looking for emails from Lindsay. 7:40 Since she's on social media 3 hours before the rest of us, 7:43 she knows if Open Site Explorer is down, she knows if PRO Dashboard is down. 7:48 I'd scroll through, I'd find out what was going on, 7:51 I could prioritize my day and know if I'm going to have to call an engineer 7:53 at 8:00 or 9:00 in the morning to wake him up. 7:56 You want to make sure it's really an emergency if you're going to wake an engineer up that early in the morning. 7:59 They're not going to be very happy with you. That's happened. 8:03 And number 4—this is probably the most important thing— 8:07 clear and concise email is really, really crucial. 8:10 If I got an email like this—and I have—in my inbox, 8:13 it's going to take me forever to synthesize through all this information. 8:16 I can't clearly tell what the problem is, I don't know who is affected, 8:19 I don't have any clues to start investigating, 8:23 like what teams do I need to initiate, who needs to start investigating this stuff? 8:25 It's going to take me a long time to read through this email 8:28 and actually figure out what's going on. 8:30 But looking at an email like this, you can skim it, you can see it really quickly, 8:32 stuff sort of pops out at you. 8:36 One of the things I like to keep in mind is to make sure that you put your conclusion first. 8:38 You want to make sure that the first thing a person reads when they open this email 8:43 is the most important information. 8:46 I got this great piece of advice from an old manager, Kate Matsudaira. 8:48 She always said to keep that conclusion to 2 or 3 sentences 8:52 because as a manager, you're always running from meeting to meeting 8:55 trying to prioritize your email on your iPhone, 8:58 and if you can see that conclusion in the preview screen on your iPhone, 9:00 you're going to know to prioritize. 9:04 So when I was in a million meetings all day long, I'd be scrolling through my email 9:06 trying to figure out, "Okay, an emergency is coming through." 9:10 "Is it my team that needs to get involved in this or not? Can I follow up with it later?" 9:12 So keep that in mind. 9:16 Another thing is to take advantage of bold, italics, underlining, color. 9:18 Don't go overboard because it won't work, 9:23 but just make sure that it's really easy to skim these emails, 9:25 that somebody doesn't have to read this word for word to know exactly what's going on. 9:28 You want the most important information to sort of pop out at people. 9:31 This is the most important, important, important thing. 9:36 This is all your engineers care about. 9:40 All the engineers want to know is how do you reproduce the problem. 9:42 And I can't stress enough how important it is if you're going to write any sort of emergency email, 9:45 make sure that your engineers know how can they reproduce that problem in real time. 9:49 The only way that they're going to be able to investigate this 9:54 is if they know who's affected, they see screenshots of it. 9:57 They need that step-by-step way to get into the user, 10:02 be able to reproduce that problem in real time, and begin to investigate what's going on. 10:06 And the last thing to keep in mind is make sure to call action items. 10:11 Don't be afraid to ask a specific team to jump in and help you 10:15 or maybe even someone specifically on a team. 10:19 We do this all the time with Erica and Jen. 10:21 When we know something is going on we'll be like, "Hey, this is happening." 10:23 "Can you tweet out to the community and let everybody know." 10:25 But don't be afraid to call people out. 10:28 If you just send out an email and say, "Hey, this is what's going on," 10:31 people might think to jump in or might think somebody else is going to jump in and take over. 10:34 So don't be afraid to just call it out, take that question away from them, and let them know. 10:37 They'll be happy to jump in and help you, probably. 10:42 Number 5. Set realistic expectations both internally and externally. 10:45 You want to provide a really open communication internally with your teams 10:51 as well as externally with your customers. 10:56 Even if it's bad news, your customers are going to be able to deal with bad news better than no news 10:58 or overly optimistic deadlines. 11:04 In the case of the crawler outage, this was a really, really shitty situation. 11:06 We lost a bunch of crawl data on these machines, 11:10 and we knew it was going to be at least a week before you guys were going to have crawl data back in the app, 11:13 and we knew it was going to be weeks before we could replay all of your historical crawl data. 11:17 We had to completely rebuild this entire database. 11:21 And so Jen Lopez would try and get on as often as possible 11:24 and update Bryce's blog to let everybody know what was going on. 11:27 It was really bad news, but all we could do really was just be honest with everybody 11:30 and let them know what was going on. 11:34 And some people appreciated this. 11:36 Jonathon Colman, who is really active in the community, transparency, 11:38 he appreciated it. 11:41 Shit happens. People understand that. 11:43 But then some people are going to be upset, and that's to be expected. 11:46 These people are paying you for a service. 11:50 They're expecting the service of you. 11:52 And when you have problems like this, you're affecting their day-to-day life. 11:54 So making sure that you're empathetic and that you're keeping open communication to your customers. 11:57 You want your customers to know this, and you want your teams to be aware of this as well. 12:03 But make sure that you're being empathetic. 12:06 We messed up with this. 12:09 In the case of the crawler outage, we promised a fix and we missed that deadline. 12:10 But not only did we miss the deadline, we didn't communicate to our customers 12:16 that we were going to miss that deadline, and so they were upset. And it's really fair. 12:19 So in these types of situations you get really stressed out, things are going crazy, 12:23 you get really busy, but just remember that the communication and empathy is really important. 12:27 And empathy to your team as well. 12:32 This is affecting your customers, but it's also affecting your engineers. 12:34 In the case of the crawler outage, we had a couple engineers that were up for 7 days straight 12:38 trying to rebuild this database. 12:41 So just keep in mind this is your customers, this is your team. 12:43 Empathy and honesty is the best thing you can do. 12:46 And finally, take time to follow up with a postmortem. 12:49 Unfortunately, a lot of times postmortems can be used as a way to sort of cover your butt 12:53 or maybe place blame, sadly, but they can actually be a really good learning tool 12:57 and it's a really great way to build confidence back within your team. 13:01 Postmortems are super simple. You only need 3 things. 13:04 You just need to let everybody know what exactly happened, 13:07 get everybody on the same page so everybody knows exactly what the situation was, 13:09 and let them know what you're going to do in the future to make sure this never happens again. 13:13 With that, number 3 on the bottom there, make sure that you're including those action items. 13:19 Make sure that people can see what steps you're taking to make sure that this doesn't happen again. 13:23 Some things that I like to try and keep in mind when I'm writing postmortems, 13:28 like I said, make sure that it's actionable and educational. 13:32 Otherwise it can look like a way to place blame. 13:36 Also try and use "we's" as much as possible. 13:39 In these types of situations, it's rarely ever one person's fault. 13:41 And even if it is, you definitely don't want to call them out. 13:45 They're going to feel awful enough as it is. 13:49 So as a manager, you want to try and accept the blame as much as possible, 13:51 but if you can't, make sure that you're using "we's" in the communication about the problem. 13:54 But make sure that you call out everybody that jumped in and helped out with the situation. 13:58 This is a really great way that you can build positivity back into a really negative situation. 14:02 How can you guys take these lessons back and apply these back at your company and in your teams? 14:07 First off, start small. 14:12 If you don't already, try and encourage transparency in your teams. 14:14 Make sure that your team knows it's okay to make mistakes. 14:18 As long as you're honest about it and you communicate. 14:20 Everyone is going to make mistakes. 14:22 And 2, practice that communication. 14:24 Writing an email is hard and it sucks and it takes a lot of time. 14:26 So practice that during the good times. 14:30 Practice communicating with your customers during the good times, 14:31 and it's going to be a lot easier and a lot more natural when times get tough. 14:34 Thank you. [applause and cheering] 14:39
You need to sign up for Treehouse in order to download course files.Sign up