Janis Zech: Hello, and welcome to another episode of the RevOps Lab podcast. I'm here with a repeat guest, Mallory Lee. Hey, Mallory.

Mallory Lee: Hi. How are you doing?

Janis Zech: Yeah. Doing very well. I think since our last podcast episode about aligning teams through a pipeline council, which I really highly recommend listening to. I think it was, like, almost a year ago. We actually met in person, and we're back here again. So, yeah, welcome back.

Mallory Lee: Yeah. Thank you for having me.

Janis Zech: Yeah. I mean, so you changed jobs, I think, in between. You had PhoneBurner now. I think a lot of folks already know you. But for the people who don't know you, who are you? What do you do?

Mallory Lee: Yeah. So I like to joke that I am a mom and I have three boys and no hobbies. That's not exactly true. So I've been working in RevOps for, you know, five or six years now, but I've been in marketing operations prior to that and B2B marketing for about fifteen years overall. And so I'm getting to that point now where I'm stacking up a lot of experience in the tech world and love what I do. I just I love working. I know that that is a very dorky thing to say, but I do really enjoy it. And I live in Indiana in the Midwest here in the States with my family. Work at PhoneBurner. I lead revenue operations for us. And it's been a different experience. PhoneBurner is smaller, and we are profitable and bootstrapped, and that's different than a lot of the bigger VC funded companies I've worked at in the past. So it's been a really cool experience to get in there and learn.

Janis Zech: Yeah. Awesome. Well, good to have you back. Today's topic is all about automating CRM updates for reps so that they can focus on selling. I think there's a stat from Salesforce that says, like, you know, typical sales reps spend seventy percent on nonselling activities. I don't a hundred percent buy that metric. I think it's, you know, like, it's always a question of how you look at it. But, certainly, I think updating the CRM has been around forever. I think there were attempts to put this into comp. There were attempts to force it down the throat. I think there were a lot of basically things that all actually didn't really work in my mind. So, I mean, maybe to kick off, right, like, what are the ripple effects of poor deal hygiene in your experience?

Mallory Lee: I've been there same as you. You have your carrot where you're trying to incentivize the right things getting into the CRM and then you have your stick. And I remember when I was starting my career, the phrase that you always heard was, oh, if it's not in Salesforce, it doesn't exist. And it was kind of true because if you wanted to be paid commission correctly, you needed to have the deal in there to get paid. But everything up until that point was kind of like all bets are off. And, you know, reps, they get in there, they do their data entry when they have time. And if they are selling a lot, they don't have a lot of time. Right? And we want them selling a lot. So what I've seen come out of that is either some information is totally missing altogether or it's just stale. And when you come into, you know, maybe a meeting like the pipeline council where you're getting a group together or a forecast meeting and you're looking at what is in the pipeline. How are we doing? Are we gonna hit our number? What's the status of this? You know, you kind of talk through it as a team and you start to see where things break down because maybe you express a concern about a deal that is stale. And the rep will say, no. I met with them yesterday, and it's fine. You know, it's on track. And so then there's this mismatch between, you know, what the data in the CRM tells you and what the rep tells you, like the boots on the ground, real time answer, the real time confidence. I've seen it work the other way as well, where you've got a nice pipeline and a deal has been lost, but it's not updated as lost yet. You go to that meeting and like half the people in the room know that that deal is actually lost and then the other half don't know. It's just, it's awkward. You know? And you never wanna be the person who's out of the loop because what you're doing is like making investment decisions, making marketing decisions, doing things based on this reality that's not real. And so it ends up creating a big gap of trust, and people don't trust what they see in the reports and in the dashboards. And so then it's kind of like, why do we have them if no one trusts them?

Janis Zech: Yeah. I mean, as you know, this is a topic dear to our heart given that, you know, we actually solve this with Weflow. But, like, I think we see this so much. Right? Like, I call it bloated pipelines. And mid quarter, you know, deals getting pushed out to the next quarter. And I think I spent around, like, thirty minutes talking to Robert, the ex CRO from Camunda about, like, when to create an opportunity. Right? What is the opportunity? And then, you know, like, what are the stage entry exit criteria? You know, how do you actually derive visibility for everyone to know what's going on and then, you know, to go into the forecast meetings, into pipeline council meetings to basically trust your pipeline metrics like stage conversion, win rates, sales cycle length. Right? And these are all the things that basically are influenced. And we have so many customers that come to us and say, oh, we need to improve our forecasting. We need to improve our forecast accuracy. And, you know, then you peel back the onion and you see that data is missing, visibility is missing, understanding of what's going on is missing. And so it's not about then creating a roll up forecast or something. It's really fixing the root cause from a first principles perspective. Right? And I think you just outlined this beautifully. I think we've all been there. Right? Like, it's obviously, with the ability now that LLMs provide, there's a big change. Right? So, I mean, you mentioned you're actually trying to fully automate data entry for your reps. So I'm curious. Right? Like, what are the different areas you're thinking of? How you're approaching it? What is working well? What is not working well? Maybe we can just break that down.

Mallory Lee: Yeah. I'll start with a brief story just to kinda double down on what you're saying too. When I first joined a company, this would have been twenty twenty maybe. So, essentially, I was brought on to the team to begin RevOps to help with this forecasting question. That was the biggest question that the team had. And as I got in there and started looking at things, what I discovered is that sales was using stages and probabilities that, you know, are tightly coupled together. And it was possible to create a forecast from that information. The problem was that no one believed it. And so finance was doing something totally separate. They were exporting data from Salesforce, looking at how much revenue in pipeline was getting created each month, and then just using weighted averages to forecast out how much of it would close. And it was, you know, a predictive model before the times of having products that really did that. And so the sales team would communicate some kind of forecast and finance would just ignore it and create their own. And sometimes finance was more accurate than the sales team. And so, you know, there's just so much data and there's so many ways to try to get in there and understand it better. But what I wanted to bring into that organization and what I really believe is important is empowering your salesperson to be able to say, what is the category of this deal? Is it winnable? Is it likely? Is it a long shot? Or is it committed? I don't really want them to have to worry about too much else. I want the rep to be able to tell me this is committed and this is the date, or this is the close date, but it's a long shot. And I want their judgment to be applied to the likelihood of winning versus, you know, procedurally what stage are you in? What's your percentage level of assuredness that you're getting this key requirement before you move to the next stage? It kind of puts the trust back in the rep's hands to say, I trust you to tell me if you're gonna win this or not and let the robots take care of everything else. Because at the end of the day, the systems, the LLMs that we have available to us now, they are very good at understanding down to the deal and down to the rep, are we on track? Are we mentioning the right things on the call? Let's analyze the transcript. Did we talk about a proposal? Did we get pushed back on price? Where did they counter? What are the blockers from a legal standpoint that are in our way? All of that is so much easier to dive into and dissect and understand now than it was previously. And so my goal, at least at PhoneBurner currently, is that the rep tells me which line items are they selling. I want them to have the line items on the deal, and we're using HubSpot for this. So opportunity, if you're in Salesforce world, I want them to tell me the close date to the best of their knowledge, and I want them to tell me the forecast category. Are we committing this or not? Everything else, my goal is to really automate for them.

Janis Zech: Okay. So this is essentially the amount, right, which comes from the line items. It's the close date and the forecast category. So activities, emails, meetings, you automate. And then contacts that are being added to email threads or meetings that don't exist in HubSpot are fully automated. Can you properly map those activities in HubSpot? Like, do they map well to the deals?

Mallory Lee: They do. The mapping has been pretty strong. So a couple of ways we're doing that. Our reps use Superhuman for email, and they have a pretty nice HubSpot integration with a click of a button. It will send the person into HubSpot for them. And what we're using for meetings is something called Fathom. Fathom video, I think is the right way to say it. And so that's integrated with the HubSpot meeting functionality, and we've had good success with that. It brings in the activity. What we're doing on the Fathom side is we have a custom sort of output or summary from the call where we've instructed it to look for certain things. And the summary is automatically created when the call is done, and then each section of that summary is sent back into HubSpot in a field. So I can see on the deal the information from the call, and it's getting put on the deal automatically after that call is had.

Janis Zech: Yeah. I mean, that sounds very familiar. We do this for Salesforce. So we capture the emails, meetings, the attachments. We then, you know, identify which contacts don't exist in Salesforce. You can restrict that to, for example, only existing accounts. This is fully automated but with mapping controls out of Gmail and Outlook. And in Salesforce, I think the extra complexity often is the mapping because the contact to opportunity is mapped via the opportunity contact roles, which often are not well populated. I think HubSpot has a different data structure. Right? Is that also your experience? I mean, you've been on both sides. I know you now use HubSpot, but you've been a long term Salesforce user.

Mallory Lee: That is different in HubSpot. It's a little different. It's a little bit easier to automate some of those associations from the contacts to the deal. It's also a little bit easier at PhoneBurner because we're typically working with more just one buyer. And so I do think that this depends a lot on your business and the differences that you're gonna experience in your sales cycle. So if you have a longer sales cycle, a more enterprise level deal flow, I think you would run into a couple of those same situations in HubSpot. But I can already think of some ways that I would still automate that, you know, to my advantage. Like anyone we had a meeting with, associate the contact to the deal. Those things are pretty easy to get in place with workflows.

Janis Zech: Yeah. A hundred percent. A hundred percent. So fully automated activity capture. And then I think the other piece is, obviously, most of the folks spend a lot of time in meetings. Right? So meeting recordings, transcriptions, AI meeting notes, and then also structuring the transcript against maybe your qualification methodologies. Like, what are you — I mean, there's different tools out there for doing that, but that's the other piece you're doing. Right?

Mallory Lee: Yeah. Definitely. There's so many different methodologies now that people use to dissect these things. And we have kind of a blend of a few different processes. Our CEO, Chris, has been in sales forever, and he's got some really nice approaches to, I think, assessing a deal and opportunity. So what we're doing is the SPICE framework. That's kind of the main gist. But then on top of that, we're really careful to try to understand what is the customer's North Star metric that they are using to define success. We really want to be able to get that from every call, every demo, and we tease that out of the system, and we make sure we uncover it. The rep can then apply judgment and make changes if they want to in the CRM. And then the other thing that he says that I love is, you know, why now or why not? And we have this built into our process as well. So as I'm working this deal, I wanna know why now. What is the compelling need they have? What is the event that is driving a decision making process? Why is this even a project? And then why not? If there's a reason that this is gonna go wrong, what is it? Let's identify that, keep ahead of it, work towards solving it proactively. And so I did a little test where I went to the Fathom recording of one of our meetings and I asked the AI, I said, why would this deal fall through? And it gave me a really nice answer. I was like, okay. Well, that's kinda cool. You know, you can use your judgment to understand that stuff, but you can also get this objective opinion from this third party robot that can, you know, give you help with that answer. And I think it takes a little bit of the onus off of the rep to try to say, oh, here's, you know, where I might have a weakness or here's something that I'm not thinking of. It just like separates the personal side of it a little bit better. And I think that that's how you get better answers to questions. Like, it's not your fault. We just wanna know what are the risks on the deal and how do we, you know, help you overcome them. And so I think it's easier for reps to highlight those risks when they don't have to, you know, feel personally judged by it.

Janis Zech: Yeah. I mean, I think the whole unstructured into structured data is obviously one of the strong elements of LLMs. I mean, a little anecdote when we built our conversation intelligence product, like, over a year ago, you know, we first launched it and we didn't have auto language detection. And so we had like, Philipp and I, we had a conversation. We recorded it. We were just testing it. And the conversation was in German, but the transcript language was English. So, basically, the transcript was unreadable. And the AI summary was on point. And we looked at that, and we were like, oh my god. Like, the LLM can basically read between languages. And then, you know, we started, you know, playing around with, you know, AI field updates. And, you know, obviously, the challenge there is you have, you know, many different field types. Right? So pick list, multi pick list, amount fields. And so what really, like, blew our mind was, like, we had a call, and then we looked at — we discussed the seat amount, we discussed the seat price, and it basically automatically calculated the amount value. Right? Which —

Mallory Lee: Oh, wow.

Janis Zech: I mean, we were, like, looking at this, you're like, oh, you know, and it obviously doesn't always work. Right? Because you don't always have that conversation or it's not always that clean. But multi pick list — I think what's super important there is, right, the flexibility for RevOps to be able to create different templates. So you can map your HubSpot or Salesforce fields to prompts. You can customize the prompts. Prompt quality plays a big role, right, if you have multi pick lists and you don't incorporate that into the prompts. It just doesn't really work. But if you do the right prompt engineering and you have the right setup and tool, it's really powerful. And I think in the end, what are good salespeople doing? And I don't want to put this down, but most of the digital signals live in emails, meetings, contacts they communicate with, and then conversations. And this is all digital now in most cases. And so the LLMs just unlock a completely different potential there.

Mallory Lee: Yeah. Absolutely. You know, the thing I haven't been able to sort out is the text messaging or iMessage. Right? In a lot of businesses that I've been a part of, one of the best indicators of success is if you're on a text message basis with your buyer, especially in enterprise, I think that becomes really important. But I don't think that you can easily integrate that or measure it. It's like the one remaining black box that you just have to have some janky checkbox like, are you texting with them? Yes or no?

Janis Zech: Yeah. I mean, that's a great point. We haven't found a solution. We have customers that come up, and then you're like, okay, basically your text message is basically also a private place. Right? So how do you separate? Basically, how do you identify? Do you let people give access? A lot of things in Europe happen via WhatsApp, for example. Right? In the US, maybe more iMessage. That's really difficult. Like, so they still need to do some mental work there, or are you asking them to do it? Or how?

Mallory Lee: Yeah. At PhoneBurner, I haven't really made it a big part of our process. It was more so at previous companies. And it was something that I would hear a lot when I would be inspecting the pipeline. And I would mention, oh, you haven't talked to this person since last week. You know? How's the communication going? And they're like, oh, it's great. It's all on my phone. It's all on text. And it drives me nuts. I'm like, how do I get into that? You know? But I think that there's also people who are doing a lot more communication on Slack now. I don't know if that will become something that's a little bit easier for us to do. But I was working with a vendor and we were gonna start a POC, a proof of concept together. And they said, okay, can we make it Slack official? And I said, sure. And so that was like their signal. Like, okay. If you're willing to create a Slack room where we talk to each other together, then this is going great. You know? So it's Slack official.

Janis Zech: I've never heard that term. I love it. But, yeah, I mean, obviously, I think being on Slack together is essentially another way of having a close communication loop. Right? So Slack, LinkedIn, SMS, iMessage, WhatsApp, all places that are harder to automate, I would say, as a summary. So if you haven't found a solution there, no worries. But if you found a solution, you might be onto a good business. Right? Because I think there's a lot of folks who need that more. Please let us know.

Mallory Lee: Yeah. Tell us.

Janis Zech: But there's a Slack community for RevOps folks. Mallory actually helped initiate called RevOps Chat. So check it out and ping us there. That would be awesome.

Mallory Lee: Yeah. Would definitely love to hear from you all.

Janis Zech: Okay. So is there anything else you think needs to be automated to get to good deal hygiene?

Mallory Lee: Good question. I mean, prior to having a deal and having deal hygiene, we also have this concept of leads coming in, and we're assessing them. Right? We're trying to figure out, is this a deal or is it not? And so there is a process also of pushing contacts forward through, you know, the early pipeline, I think, to discern if the deal is there or not. So I've tried to automate a few of those things. For example, in PhoneBurner, when we have calls using, like, the PhoneBurner software, they are marking the outcome of the call. And if the outcome of the call is, you know, one of five things, I'm then automating that contact status back in HubSpot to kind of move through that customer lifecycle, if you will, to then show us how many of our contacts are engaged, how many of them are dead, how many are we working, but we haven't engaged them yet. So I'm trying to automate that part too, because in the past, I've had to do a lot of coaching around how to use the statuses effectively. And that's just another place where you can start to doubt your own data if people aren't consistently using the same definition for, you know, dead or whatever that stage is. So we're doing a little bit of automation there. We automate a few things on pipeline generation because we have a PLG process where the sales reps are not involved. So we have multiple go to market motions that are getting supported. So for the product led, the opportunities or the deals are getting created and closed won without anyone intervening at all. And on the sales side, we are giving the rep full control over when they create the deal. So we have an automated deal creation, but more so where are the contacts, where do they stand in this process. And then if a deal is opened, we'll go and update the contact status to reflect that we have an open opportunity. So some of those things are coming together pretty well. We are not currently automating stages. I think that that could be an area of opportunity for us in the future. You know, because we have this simplified process, it's very simple on purpose. We want to be easy to sell, easy to buy from, fast process. We have a very high velocity sales motion. And so our goal is to keep everything as simple as we can. So we're doing many things all in one place in HubSpot. So we're using their proposal and quote functionality. So, again, I can automate it. If a quote's created, go change the stage to proposal. We can do those things. And so I could see that being on the future horizon. And I do think that that's different from the judgment that you apply to an opportunity to say, I'm gonna win this, or I might not win this. And I think that stages need to be treated as very black and white true false. This is, you know, this is what's happening on the ground. We're in discovery or we are in the proposal or we are in negotiation. I don't like when people mix up their confidence level with their stage labels because I think those should be two separate things.

Janis Zech: Yeah. A hundred percent. I mean, I think so many thoughts here. I mean, it's something I think a lot about, but, like, the stage is essentially your process. The forecast category is your judgment. And then typically, you combine, you know, a technical forecast. Right? Might have — could be weighted forecast. Ideally, update the stage probabilities according to your different, you know, like, realities or dynamically weight is, I think, a better way. Right? Like, last sixty, last ninety, last a hundred and eighty, last three sixty days. That's, I think, really a quick win. Then you obviously have the bottom up forecast. You have to, like, you know, kind of the judgment from the field, and then you typically have, like, some AI prediction, ML prediction, technical forecast. I think run multiple of those. Right? Like, I think it's always recommended. And I think what you just said with top of funnel, I think that's another — that could be another episode. Right? I think there's a lot happening on the top of funnel side with, you know, where should top of funnel actually spend time? Right? Like, whether that's, you know, targeted account list, signals, but a lot happening there. Curious, like, I mean, what we see a lot is also that pipeline management is not just a new logo thing. It's an expansion. It's a renewal thing. Existing accounts play a bigger role now in driving, you know, like, efficiencies and growth. It has always been important, but I think with the move towards, like, more efficient growth, it's become maybe more focused. I assume the same principles apply there, automated activity capture, like conversation intelligence to get full visibility. Is that correct?

Mallory Lee: Yeah. For our business specifically, a lot of our add ons or upgrades are done via self serve. And so we don't need quite as much time to be taken in the CRM for expansion. But for our larger customers and our larger accounts on the enterprise side, we definitely have some renewal deals and the same thing happening on the renewal side of the house and some of the bigger expansions into like a totally separate business unit, if you will. So same thing would apply there. We're also doing the same sort of activity capture for onboarding and getting the customer up and running. All of those meetings are being captured and recorded in the same way. I haven't done much yet to kind of mine the intelligence out of that. So for example, I could have a whole different set of instructions for the LLM to go understand how did they set up. You know? They're making these decisions live on the call with our solution architect. Sometimes they're gonna change their strategy during the implementation on the fly. How can we, you know, take note of exactly what got implemented and not make, you know, the onboarding team go back and type all that out. So there's just like the sky is the limit on ways that you can go use that information.

Janis Zech: Yeah. A hundred percent. I mean, you just need a flexible setup to have different, like, kind of templates for different teams, and here you go. Right? Like, typically, the onboarding or, you know, CS teams have very different kind of qualifications or things they do. Right? So the AI summary or field updates, they're just different. They might also end up at a different object in the CRM. Right? So just need that flexibility, but it's absolutely possible these days. And I think most tools support that as well. I mean, one maybe one final closing note. Right? We started with the ripple effects. Right? And, like, let's assume we have this — you know, everything is automated. What I also really like, by the way, is the automation of, like, you open an opportunity, close an opportunity, and self-service. Right? Like, that's pretty awesome. I haven't heard that a lot. But then, right, you tie this all back to signals to understand, like, leading indicators of deal health. Right? Would you say doing this, the conversations in the pipeline, how some of them have gotten better or the visibility has gotten better? What are the outcomes if you do this well?

Mallory Lee: I think it accelerates your conversations if you're doing it well because you remove the step of checking first if everything is correct. You know what? I see ten deals in your pipeline. Is that right? Yes or no? Instead of debating it, you can just say, what's the most important deal? AI is telling us that this one is the most important deal. Do you agree? It gives you this freedom to sort of, like, interrogate what the AI is telling you and then to get the rep's opinion on it and then to talk about the actual merits of the opportunity versus just trying to get your arms around how many opportunities do we even have. So I think it allows you to spend time on things that matter a lot more. I will be the first to admit that there are times where I still question things. You know? If HubSpot is telling me that a deal is stale or overdue, I think, oh, did that activity get captured correctly? You know, did we have a breakdown in the process of capturing something? And so that is always in the back of my mind. And I always try to approach it with that level of curiosity. I'm not gonna go to the rep and say, you're seven days late on following up with so and so. I'm gonna ask them like, hey, did HubSpot miss something here? And sometimes it does. It just misses something. Or your connector gets shut off or your Chrome needs to update. Right? And so it's not foolproof, but hopefully it gives you a few things that you can skip over, build trust, understand, like, what's in front of you, and removes some of the, like, subjectivity from the conversation.

Janis Zech: Yep. Awesome. Awesome. I mean, super interesting. Thanks for sharing all this. Maybe a final question for you, and I actually didn't prep you on this, so you're on the spot now. What's a book, research report, or also awesome content creator you would recommend for RevOps folks to read or follow?

Mallory Lee: Great question. Well, I already told you I have no hobbies, I don't read much. So I will be vulnerable and admit that. But what I noticed is that the team at Fullcast just launched a book, and I have been interested to check that one out. I haven't read it yet, but I know that is something that's been on my radar. Also from, like, a podcast perspective, obviously, I like your guys' podcasts a lot. I think that there are so many resources out there. And since we've started RevOps Chat now, my hope is that those will be a little bit easier to come by instead of trying to go out and find them on my own. I need the community to just send me things like, Mallory, you need to make time for this.

Janis Zech: Yeah. I mean, I'm curious how that's gonna continue. I think we're off to a good start. Lots of folks joining. And yeah. I mean, I think we're gonna launch a little city chapter team challenge next week. You know? City chapter growth by capita to make it fair. That's something like this. It's not final. You know? Like, I still need to discuss it with Daniel and, you know, the founding member group, but it sounds like it could be cool to, you know, do that because I think meeting with local folks in real life is just super fun, and it's actually fairly easy. We've done a bunch of meetups, you know, last year in London, Berlin, Munich, and I think we wanna actually continue that. I know Harris is putting one on for Amsterdam. And if you have a local community, you can just basically ping them and say, hey, look, let's meet for after work drinks on Thursday in this bar, and we just hang out for two, three hours. And people go out even without RevOps folks. But it's been really nice. Right? Like the meetups, we had people join six p.m., leave ten p.m., and you have to do nothing because folks just come together and it's like a therapy session.

Mallory Lee: Yeah. You guys do a way better job at that across the pond than what we do. I don't know what it is. Maybe it's like the fact that a lot of us live in the suburbs, and so you have to get out in your car and drive twenty minutes to the bar. I don't know what it is, but I'm always really jealous of all the great meetups that you guys have.

Janis Zech: I think tell your three kids, you're back in two weeks, and we'll do a little road tour through Europe.

Mallory Lee: Hey. I am all for it. Better yet, I'll bring them along and they can be — yeah. I'll bring my two along and, you know, they can play and we just, you know, have it be unwind.

Janis Zech: Look, Mallory, thank you so much. This was awesome. Really appreciate it, and wish you a great day.

Mallory Lee: Yeah. You too. Thank you.