#5: Why your best signals go cold before anyone acts (and what to implement)

Hey everybody we are back with episode five of the context layer tapes. Super excited about this episode. Today we're gonna be talking about why your best signals go cold before anybody acts and what you should actually be implementing. So we're gonna get into some of the nitty gritty details and breakdown where AI is ultimately going wrong and again what you can do to fix that. I'm Florence Twia the GTM engineer in residence here at Zoom Info and alongside me is my co-host John Lloyd. Welcome back John. Hey hey good to be here. What's what's going on with you this weekend for and I know you have a vacation coming up any big plans. As you indeed I'm actually flying out later tonight. Got a good buddy's wedding in Portugal. It's a big one two over a hundred people which is ambitious but it's good it's gonna be good. Oh my god. Euro Summers. Enjoy. Thank you so much. You're gonna head up Spain as well and it's in a Portugal so big week ahead for me. What about yourself? All right yeah well my daughter is turning one soon so we're doing birthday planning. I asked my wife what the theme was and she just said the theme is birthday so I think I think we can roll with that. If it sounds pretty you'll hold to me. Sounds like you're working with a lot. I like it. Yeah excited to dive in. All right John so let's get into it. So today we're going to focus on one data point. Again we interviewed 50 GTM leaders at some of the largest corporations in the US and one of the questions was around real time signals. So interesting data point 56% of senior GTM leaders told us the same thing needs fixing in their setup. So their AI tools don't have access to real time signals and I think this is a very important topic to talk about because when you think about an evaluation cycle for most companies and whether selecting vendors you only have a limited a limited amount of time to figure out exactly when they are going through that process and make sure that you're able to get in that evaluation especially if you're working with like mid-market enterprise organization. So I think this is the right framing right now for a few reasons. AI workflows depend pretty heavily on current and up-to-date buyer context obviously and then stale or even slightly inaccurate signals or data do make AI act on the wrong thing many times and it really does come to you the need for real time context right. What are your thoughts there? I'm when I saw this data point I was honestly a bit surprised that this number was so high because signals are becoming a bit of a commodity if not they have been for a little while. So I think the big differentiator as you called out is a recency and the ability to get a signal that's actually actionable not something that is months or a year old. So I think like that's how I'm contextualizing this data point it's that like what we're really talking about like the key phrase is real time right not just signals. So I was surprised that this was so high but you know I think it's it's hard to know exactly like who is the one responsible for you know researching an intent topic or visiting your website. A lot of times these signals are at the company level versus the person level but you know they fail to go to go deeper than that. So I think that's part of the problem too. Yeah that's definitely part of the problem and I think naturally a lot of signals do have a pretty fast half life or a decay period essentially and it's super important to act on real time signals and it completely changes your conversion rates again as somebody that's built out these plays and orchestrated a lot of these real time signal based workflows. So I can give you a number of different examples just for my own experience but one I've talked about this in the past web site signals so if somebody's actually on your website especially on high intent pages like the pricing page you get a massive increase in conversions if you're able to act upon those within 20 minutes. So if something's top of mind for prospect and you're able to catch them with a good contextual message while they're still there the chances of them actually booking a meeting at least for my data and experience is 2 to 3x higher. So that's one, another play that we ran significantly in the past and I know we do this at Zoom info 2 is the idea of social signals. So you know monitoring various social platforms see increasing gauging with content whether it's from your corporate account or on topics that are related to the product resolution that you sell. We've all been burned by those messages where it's like hey I saw you like to post about you know x, y and z we actually sell the solution for x, y and z should we get on a call. What I've noticed with a lot of these like social based plays is that usually the orchestration and the amount of time it takes for a rep to actually get that signal then act on it sometimes is days at a time and you have to remember like even thinking about myself and schooling on LinkedIn this morning is like do I even remember post like during gauged lift from 2 hours and that is no. We all have short-term memory loss. I'm not included. Well we just have so much information that's being fed to us at a date level so that's why like at a site out of mind and you need to act on it as quickly as possible. Another example job changes we talked about this in the past two I had someone reach out to me these literally on a cold call they thought I was at my previous company which I now haven't worked out in six months. So that's not obviously the rep needs to do a double check there and it is partly on the rep but I think for me it's more of a systematic issue where data is very still and then naturally you're losing the trust of a buyer before you even start because you don't have the right data at the fingertips of the rep. Yeah job changes are tough because that person that new CFO who's coming in who's going to control the budget that you need to get you know and harness for your sale I mean they're going to be rare to go once they come in so you can't you can't reach out to that company seven months later and assume hey the CFO's you know they're still getting the budget together they're you know still deciding what to spend on in the new year like no that's going to be one of the first things that they are acting on when they join a company especially if it's around like the start of their their fiscal year so so yeah a recency means everything when it comes to job change signals you know intent and and web visits 100% and I think that the last thing I'll say about that you know the timing aspect of it as someone that's bought software and has been in sales for over a decade now yeah I think one thing we don't consider enough of is the evaluation timeline so sure a buying cycle especially at a large organization might be 12 even 18 months you know so at the extremes but when you think about the amount of time that most companies select are doing vendor selection it's a pretty thinnest window of time where deciding who they're going to look at right so you usually have two to four weeks max in my opinion again there's always anomalies but there's that two to four week period where a company is deciding that hey like these are the vendors we're actually going to evaluate so you have this small window of time where if you don't act on real-time signals and intent in that window you're likely going to be too late and you have to think about it right like evaluating software is not really part of anybody's job description it's like an added thing that you're usually doing on top of everything else so people only have so much bring capacity to do these things so if you don't catch them in that window that's a problem yeah so we're thinking about two ticking clocks right it's the the the deprecation of the data the signal and the deprecation or the you know the the ticking clock of the procurement cycle as you said so it's like if both of those things don't align well then you know too little too late you're you're out of luck that's exactly right so John let's actually go through this we have a pretty interesting format for you today in the podcast so we're going to go through what it requires to to actually set up a play like this and then we'll actually walk you through a specific play on real-time signals and how we think about the orchestration of it what was there anything else you'd like to add there John I don't think so I think like you said there needs to be a setup the context you know a coordinated and connected account plan for whether it's an a or an sr a bdr and a m like plays shouldn't be run it's not a play if it's a it's a one-to-one you know outreach or even one-to-few plays need to be done at scale so the setup which you know I think our CEO Henry Schuck has posted about on LinkedIn before the setup really needs to be a scalable and connected plan that includes the signals that we're talking about so the signals when I say signal I'm talking about you know job changes like hiring mergers and acquisitions website behavior intent signals even contacts from recent you know calls and emails like if you can get all of those third-party signals can Connected to first-party data, like whether or not there was a closed-lost opportunity within the past few months. How much the person, if they are a customer and you're going for a cross-sell, upsell opportunity, how much they're currently spending with the company, whether or not their account type is prospect or customer even, like basic stuff like that. The first-party data and the third-party data need to be meshed with your AI tool. That allows you, whether you're on the rev-up side, the sales-ops, or whether you're an AE yourself, that allows you to build an account plan that includes all of these signals together across your book of business. That allows you to scale and save time and understand where and when these accounts are becoming actionable. Obviously, automation is huge here, setting up a workflow to give you an alert, or give your employees, your BDRs, your SDRs, an alert when that trigger happens, when that person visited the website, when someone filled out a form, when someone liked to post on LinkedIn, or engaged with you in any way, as you said, the recency is really important. Setting up a workflow, having that trigger set up is really important. I think that's the actual play here. What's the trigger that activates your account plan? Maybe let's go into the play a little bit. I think you probably have some examples from your past lives as a sales manager, a sales leader, where you've probably activated this. I'm a huge believer in building account plans. This is not only for AE's, by the way, I think some of the best SDRs that I've ever managed, especially when we're talking about mid-market and enterprise accounts, where those are very high value, and you should be spending some time on it. When you spend time upfront sharpening that axe and collecting the right data, you have such a higher opportunity to convert at a higher clip. I still remember to this day my best SDR that I've ever had, I'll even shout them out, Harry Sims. One thing that he fundamentally did different is before he started prospecting, he had a book of business in every single quarter, what he would do is you'd spend multiple days actually really thinking through accounts, building the account plans. We had a specific AI agent that obviously did a bunch of the preliminary research, but I think there's still the human aspect of it that takes time, where you need to connect the dots. How does that actually match the solution that you have? How are you going to think about writing the message, what are the frameworks that you're going to use? But you can imagine if you didn't have an agent to actually collect all the first party data, the third party signals, and all this data was all scattered, that would easily be 30% of your day throughout the whole quarter would just be trying to find all this information. Having one context layer, a pane of glass, where all this exists, and being able to fetch that data, and then allowing a human to contextualize it is such a game changer. The reason why it works, the context already exists, it's there, but usually you'd have to go to five different places, maybe seven places, and I remember we even had a slide when we would be presenting this to these types of solutions, whether it's Zoom in for a common room in the past, you'd have a slide. It's like reps have 20 tabs open, and it's just overflowing. At least, 20 is like on the low end of the spectrum. I've seen some desktops before that are just mind-boggling. So just to give you an idea of how powerful building these account plans are at scale, I think Harry had an insane conversion rate, so to net new accounts, I think he was booking like 15% of them over a quarter, which is literally unheard of. Most SDRs, usually if they're pretty decent, are three to four percent conversion, and I think my next best SDR at that point in time was at seven. So you're like three to five X conversions when you do this properly. So again, a lot of time was spent sharpening the acts up front, and I think we had a number of prompts and agents built that looked at multi-dimensional things. So one was like first-party data, so product usage, is this someone that's used our product in the past, are they on a free trial? If they are, what were they actually doing in that trial, and being able to summarize that? Yeah. Then you have third-party data, which again, most reps look at manually, which is job changes, technographics, job listings. And then we also have this category of second-party data, so especially a common room, because we work with a lot of developer tools. Something that was very important to us was open source repositories, so actually seeing what developers were building in a specific ecosystem. So you can imagine, we already have 20 tabs open now, if you're trying to extrapolate that to going into GitHub repos, that becomes very complicated. So having agents that can bring all of this into one place, it's extremely timely, you have everything you need to know, and you know the best point of entry. Again, led to three to five X better conversion. And this is why this topic of account plans in doing it properly, and timely is extremely important. All right, well, shout out to Harry, 15% is a pretty high bar that's intimidating. I have an example, too, that I want to speak about. So this is in terms of the play, right? So we're talking about powerful triggers that would kick off a workflow and get the right accounts and leads into, you know, the books of businesses of the SDR team. I was working with a client, a managed service provider based in Boston, and they were using intent signal based triggers. And they had an issue where the data just wasn't fresh enough for them. They wanted to go a little bit more recent, and they wanted to be a little bit more like cutting edge. They wanted to mix things up and not just use an intent signal as a trigger, meaning like when I say an intent signal, I'm talking about, you know, getting a ping from, you know, someone in your new business, a book of business that is just looking up managed service providers, or they're looking up cybersecurity, or they're, you know, they're researching other cybersecurity vendors or providers. They wanted to understand when that person was actually spiking across intent engagement, website engagement, and social engagement. So what they did, and, you know, this was partly built with Zoom Infos platform, partly on the client side, was we built an intent signal score and in market score that, you know, Zoom Info uses all the time for engagement, intent signal spikes, website spikes, they built a delta for that score, which basically flagged an account that had a huge uptick in their in market score. So if someone went from a zero to a 50, or a 50 to a 100, that told them that the person, you know, wasn't only looking up an intent signal, but they were just looking at their website, they were just engaging with them on social media, they were, they were doing potentially all three, like in one day. So that intent signal score spike that delta was the trigger. And we built a workflow that basically, you know, took that trigger and sent accounts and leads at those accounts into a Salesforce campaign for those SDRs or AE's to target. So that was, that's an interesting use case from my experience. And I saw like the call connector rate of those people working on the campaign, like Skyrocket, it was like, you know, 3% to like 12% that week when we were looking at the data and looks. So that's an example of just like an interesting trigger, a creative trigger that you can build and, you know, not just rely on one signal, but multiple. In this case, in the form of like a score change or a delta. So that was interesting. Yeah, that's super interesting. And just again, those to show the power of how important it is to have like real time data for sure. So I guess from there, John, the question becomes like what actually makes this all possible, right? And just from your perspective, how do you think about that? Yeah, we've talked in the past about, and we've seen the data firsthand, like the fact that folks are saying that they don't need better AI, they need better data, they need better infrastructure. So I think the infrastructure and the foundation needs to be there, right? Like we've talked about identity resolution, that's a huge factor. And first and third part of unification, which we just talked about, are the two biggest pillars here to prop up these plays. So identity resolution, we're talking about like, hey, when a signal fires, does it actually hit the right account? Are there, you know, tons of duplicate accounts? Is your hierarchy out of whack? So I think like one of the first things to check there is to make sure that identity resolution is in a good place. Pull your top 100 accounts and count the duplicates. Literally go into CRM, go into Salesforce, go into HubSpot, pull your top 100 accounts. accounts and look at the duplicates is the same company logged two or three times under slightly different names. Is there a high duplication rate? You know, because that means that records aren't resolving to one canonical entity. Maybe, maybe your sales reps have the ability to enter new accounts manually versus like just getting them through an enrichment workflow like Zoom Info's ring-lead solution. So I think like, and again, when I've seen the ability to enter stuff manually, I think obviously that gives sales reps with, you know, more flexibility, but it also creates a lot of noise and the margin of error is huge. So identity resolution is a big one. And then first and third party unification is the second one. So do outside signals, you know, reach the right record without anyone touching them. Are you able to connect first and third party signals to, you know, the same AI tool that brings them together. So the first thing to check there would be like, you know, pick one account and try to answer these questions in a single, singular screen. Like, what's the deal history as anyone at that account visited your website? And are they showing any third party intent or they researching topics related to your products or services? So if you can answer all of those questions as, you know, as in a single pane of glass or on the same, the same view, then I think, you know, maybe you're in a better, better spot to run a connected trigger-based sales play like the one we talked about. But if you had to open three different tools and do the math in your head, if you have, as Florence said, 20 different tabs open, maybe you have some work to do on the foundational side. What do you think about that foreign? What's your reaction to those two pillars? Yeah, I'm fully aligned of seeing this plan in real time. And I really like that idea of the exercise. And I think it's relatively easy to do in a post-mortem, especially if like you've lost the deals like where they're specific, go across all the different systems that you have if you don't have like a unified layer. And where there's specific signals or where there's, if you want to call them signs that you missed, throughout that deal that could have fundamentally changed the outcome of that. Right. You'll actually be surprised when you do that exercise because it's something that we even do here in our post-mortems when we lose deals. Just to see exactly like, if you had access to a specific piece of data in that time period, would the outcome have changed? And I'll give you a perfect example, because talking about, you know, first and third party, similar unification, I think earlier I talked a bit about the evaluation period and how much that matters in terms of vendor selection. So in one post-mortem that we did, one of the things or plays that we would typically run, especially when we were displacing platforms like Sixth Sense and Demand Base, we used a technographic signal to reach out to companies that we knew were essentially using one of these APM platforms that we felt we could displaced pretty easily. That signal actually was built in through a software or a vendor called Bill With. But what we realized was in that Bill With technology, you can actually see the moment or the specific month where Sixth Sense or Demand Base essentially was added onto an accounts website. Yeah. Right. So looking back at it, we realized that they ended up just renewing with Sixth Sense or Demand Base if we had looked at that third party signal that said, hey, they actually, the first time we saw Sixth Sense on their website was in November, that usually means that they're probably evaluating whether they're going to renew with them or turn with them three months before. So if we had acted on the real time signal and knew that we should reach out to them in like August, that conversation might have been extremely different because we would have caught them in the evaluation window. Yeah. Mr. Window. Exactly. Right. So there's these little things where if you actually go back in the post boredom and really try to analyze where you went wrong, having these this orchestration system, the single propane of glass where all this information would have been fed to you and then alerted the rep in real time could could change everything about a deal. Yeah. I love how you explain that. It goes back to the two ticking clocks or time bombs that we that we talked about of deprecation and missed opportunity. Yeah. Yeah. There's always always a ticking clock no matter who you're who you're reaching out to. Yeah. And you talked a bit about like what it takes to build this from like a technology standpoint. The other interesting thing because a lot of the times I'm thinking like, why don't people have this in place? Technology is one side of it. But what I've realized too, there's two other aspects, right? It's change management. So people in ownership. I think a lot of the times we don't actually know who like owns the system or who's going to be accountable for a specific single reaching a rep then making sure that an SLA is met in terms of like acting in that in that window of time. So if you do get a website visit, who is actually responsible to make sure that that rep does hit them up within 20 minutes? Is that an automated play? Is it a play where there is human intervention because you want more context? So one of my learnings around ownership and this actually came at common room because I was one of the main people that built out the orchestration engine is that you really have to figure out sometimes there's going to be one owner obviously the whole process. But there's a lot a lot of like many things that need to happen for orchestration to work. Who owns plays? Who owns, you know, tagging of specific whether it's like ICP or that there's a ton of different types of tags we can have. Who owns workflows? Who owns the actual agent building? Who owns the UI and interface of how this is actually being presented to rep. So when you don't have a clear understanding of that, it's very easy for this whole system over overall to fail. And that's just something that I've noticed from my own experience. I can back that up. The people ownership is critical. I mean, that's at the infrastructure level of of how critical it is. And it's often overlooked, making sure that the people who you're working with know what elements of the sales play process that they're owning is absolutely critical. Yeah, I fully agree. John, this was a good conversation. I think a question I'll ask you and then I'll challenge our listeners as well. If there's one single thing you'd want a GTM leader to do differently on Monday after hearing this call, what would you tell them to do? I'll go first. Going back to John's point, I think I really like the idea of doing that post-mortem exercise. There's a lot of information that you can look back on to figure out exactly. Not necessarily where you went wrong, but if you had access to a specific piece of information at a specific point in time throughout that sales cycle, is there something you would have done differently? And I think you will all be surprised after running that exercise and being able to move forward and actually change your system to make sure that you're focusing on being the action ability of that data is going to be absolutely key. And I think it can transform your GTM process. It's definitely transformed ours. Yeah, my answer to that would be look at the signals that you have access to. Pick your top signal and realistically answer the question of how quickly could you reach out to this person if that signal fired or how quickly could you get that signal to fire if someone triggered it? That would be my number one question. Yeah, I think that's going to be an eye-opening exercise for everyone. Yeah, I totally agree. John, it's been a pleasure on this Friday. To our listeners, I really appreciate you. It means a lot for you to listen to the context layer tapes. Please subscribe and share this with whoever own signals and routing in your GTM motion. Thank you so much. Have a wonderful rest of your day. Take care. We'll see you for episode six. See you, everybody.