Why CRM Still Fails Sellers And How AI Fixes It

Your CRM knows a lot about your customers, so why does it still feel like a place you go to do homework? We talk with SugarAI CEO David Roberts about the hard truth many sellers live every week: CRM has mostly evolved into a management tool, not a system that makes reps better in the moment that matters, right before a call, a meeting, or a critical follow-up.

We dig into why the “360-degree view of the customer” often backfires. More dashboards and more reports can create more overwhelm, especially when a rep is managing hundreds of accounts, dozens of deals, and a product catalog with thousands of SKUs. David lays out a more practical promise for AI in CRM: move from data straight to action. Instead of forcing sellers through a rigid process, AI can interpret unstructured sales work like emails and meetings, spot gaps like missing executive sponsors, and recommend the next best step that improves retention, expansion, or new business.

We also explore contextual intelligence and “precision selling” and why generic AI advice is easy to ignore. The best guidance comes from combining customer data with the real selling context: sales methodology, pitch decks, competitive battle cards, and objection handling. For manufacturing and distribution teams, connecting front-office CRM with back-office ERP data like inventory and shipping can surface reorder opportunities without sending reps into endless spreadsheets.

If you care about sales productivity, revenue team performance, AI CRM, and faster sales onboarding, you’ll get a lot from this conversation. Subscribe, share this with a sales leader or RevOps partner, and leave a review with the biggest “next step” you wish your CRM would tell you.

Transcript

Why Sellers Feel Stuck In CRM

SPEAKER_00 0:01

Hey everybody. Super excited today to be joined by Sugar AI CEO David Roberts to discuss how AI can finally make AI useful, not just by replacing sellers, but by making them dramatically better. David, how are you?

SPEAKER_01 0:16

I'm great, Evan.

SPEAKER_00 0:17

Thanks for having me. Well, thanks for being here as a longtime salesperson of 35 years. I've been steeped in CRM, which has for decades promised to uh help salespeople sell, but often does not deliver. We're going to dive into some really great topics. Uh, before that, uh, how would you describe uh sugar AI these days and uh your role in the company?

SPEAKER_01 0:42

Yeah. Well, I I'm here, I've I've joined and been here just less than two years, and we're really going through a pretty big change at Sugar. Um, the company started as an open source platform back 20 years ago, and we've really evolved through open source to more broad-based horizontal CRM. And today we're really purpose-built to try to do exactly what you've just described, and that is help sellers sell.

SPEAKER_00 1:04

Well, so it sounds like a very simple value proposition. Of course, there's a lot uh behind that. Let's set the scene. I mean, many reps, uh, sellers, individual reps, feel they spend more and more time entering data and living in the CRM without getting value in return. Where are we on the CM uh CRM uh continuum in terms of getting value? Have we fallen short of expectations over these last two decades? Well, I think so.

SPEAKER_01 1:36

I think um it's mostly been a management tool. Uh I started my career in CRM at Accenture working with FEBL systems back in the early 90s. And back then we talked about trying to help sellers sell um largely, if you ask a seller, and and Evan, you've been in the field for a long time. You block your Friday afternoons, you spend a couple of hours updating your CRM. So by Monday morning, when your managers look through your pipeline, um, if they feel like you've you've done what you need to do so that they can then be ready for their forecast

CRM As A Management Tool

SPEAKER_01 2:05

meeting midweek with the executive team. It's just this whole process of trying to shoehorn sellers into something that is process-oriented and structured. And that's not my experience on how the best sellers actually operate.

SPEAKER_00 2:18

Yeah, well said. So, what's the uh unique or valuable insight that a salesperson should be getting from a CRM system that most systems have not been able to deliver?

SPEAKER_01 2:30

Well, I think a really simple example of um maybe you're a seller who's got 150 different accounts or 200 accounts, and you're managing a pipeline of 20 or 30 deals, and you're working with a product catalog that might have 5,000 different parts or SKUs in it. Um, I think the thing we hope sellers get is go talk to this customer about this product. Specific direction. So um I've I'm guilty of this. We've talked about the 360-degree view of the customer for decades. I think that's a failed proposition. I don't think any seller could have all the information coming on from every single angle and figure out what they ought to go spend their time on based on all that information. I think it's just completely overwhelming. And I think this idea that we can provide dashboards to sellers that are going to help them sell better hasn't delivered.

SPEAKER_00 3:25

So enter AI and Gen AI, very exciting, of course. But how and why exactly does AI give CRM finally this ability to deliver that promise of actually helping sellers sell?

SPEAKER_01 3:38

Well, I think AI might might be perfectly suited for CRM. And I'll explain what I mean. Like if you look at operational systems, um, ERP systems, manufacturing systems, accounting systems,

Why Dashboards And 360 Views Fail

SPEAKER_01 3:50

they tend to be serving very process-driven business, very structured and workflow-driven business, things that are very data-driven, things that are what I call knowable or controllable. And I think selling is not that. I think selling is largely deal with unstructured data, Zoom calls, meetings, emails, unstructured data. It's not a linear process. Anyone, anyone who's carried a bag knows it doesn't just march along through the sales stages that your RevOps team built for you. Um, it's not knowable, it's not controllable. And AI is incredibly powerful at helping suggest the next step in the process without forcing a seller through a process or understanding insights from data that's inherently unstructured and making sense of that unstructured data. So I think AI is perfectly suited to go deliver on this promise.

SPEAKER_00 4:42

And speaking of AI, I mean data and AI intrinsically linked, and we have plenty of dashboards, we have reporting and reports, but you know, this business intelligence hasn't typically delivered enough. Why why is that?

SPEAKER_01 4:57

Well, I think it's just overwhelming. And I think it's assuming. So I I said this to someone the other day. If you if you look at the old way of doing this, you've got data, which gets presented in dashboards that hopefully sellers can draw insights from that allow them to determine something or make a decision that then get them to take an action, right? So it's this buildup and the assumption from a RevOps team or a BI team or a FPA team is that sellers are going to work their way through that stack and get to the action, right? From data to dashboards to insights to a decision to an action. Um, what AI can do is

AI That Jumps From Data To Action

SPEAKER_01 5:35

take you from data. Let's capture everything we know about that customer, back office, front office, everything we know, and go right up to action, right? Tell a seller go identify the executive sponsor for these five deals in your pipeline that don't seem to have an executive sponsor, right? Like guide them, um, point them in the right direction, tell them the next step to go take. Sellers are pulled in so many different directions by their boss, by their customer, by the market. Um, we now can say this is the thing you ought to go do that gives you the best chance of retaining business, cross-selling business, winning new business, whatever it ends up being.

SPEAKER_00 6:11

Brilliant. And you've talked about the need for contextual intelligence, not just artificial intelligence. What does that mean for the salesperson during a typical workday in practice?

SPEAKER_01 6:22

Well, I think AI is most powerful when it's got all the context. So it knows you're you're giving it like in an implementation in the early days was go define the process, identify the fields, do some data conversions. Today, an implementation includes uploading a sales methodology, including a standard presentation deck, including competitive battle cards or objection handling. We're loading content in as much as we're loading data in. And all that context is what makes AI more precise and what it's recommending to a seller. So we we call it precision selling. We think it's got to be precise. Salespeople are a little skeptical of anyone telling them what to go do. If you can't demonstrate that what you're guiding them towards is based on something valid or is going to result in a good result, um, they're going to give up on it fairly quickly. So it's got to be precise, and that context drives precision.

SPEAKER_00 7:17

Brilliant. And I understand the Sugar AI brings together customer information, sales data, ERP data, and more. How does combining that data, you know, give revenue teams, you know, a better perspective of customer and the account in general? Yeah, it just rounds out the context.

SPEAKER_01 7:35

So we have a we have a very specific strategy to combine front office and back office data together. If you're an account manager at a mid-market manufacturing company, like why wouldn't we show you in your workflow what inventory levels for your customers look like? And why wouldn't we let you create an opportunity when inventory levels drop so you can go engage with the customer

Context Creates Precision Selling

SPEAKER_01 7:55

about reordering or adding additional product to their um regular orders, right? Like there's so much data that we know about a customer in the back office, manufacturing data, shipping data, inventory data, supply chain data. Let's combine that into everything we know about a territory, an account, or an opportunity, and just give that seller as much as we can to guide them towards the right next action.

SPEAKER_00 8:19

Interesting.

unknown 8:20

Yeah.

SPEAKER_00 8:21

Throughout my career, I was told always by the sales management, you know, you need to be more strategic. You need to sell more strategically. It sounds like that unified view can really help with that when it comes to risk and growth and the whole perspective. I think so, but I don't think it's by overwhelming them.

SPEAKER_01 8:37

Like lots of times people are like, great, we'll send all the salespeople the inventory report. I don't know if you've ever opened up an inventory report, but like, gosh, that's not gonna like help me figure out where to go focus. Instead, let's take that inventory data, let's feed it into the AI models, and let's point the seller toward a reorder opportunity and not make them sort through a report.

SPEAKER_00 8:58

Yeah, reporting is interesting. Those terrible Excel spreadsheets that are, you know, 50 tabs long. It sounds uh useful for certain industries, maybe manufacturing distribution companies with lots of SKUs and catalogs and uh complex account relationships. Uh is that one of your focus areas that children's that's what we do best.

SPEAKER_01 9:19

We do best when there's complexity in the account base or in the distribution channel or in the product catalog because AI can cut through all that and guide people in the right direction. Like there's industries that I call, I mean, that's our ICP. I have industries I call RCP or related customer profiles. And as you work your way out in the bullseye, there's industries that are just not our customer profile. Um, we've got some software companies, we're a software company. We're not best suited to serve software companies. We're best suited to serve manufacturing and distribution companies.

SPEAKER_00 9:51

Brilliant. Uh I've heard you describe AI as the Iron Man

Mixing ERP Data Into Sales Work

SPEAKER_00 9:55

suit for sales teams. I've been fantasizing of a Jarvis-like environment myself with AI for some time. But what is it, what does that look like in practice, day to day today, even?

SPEAKER_01 10:07

Well, I don't think we're going to replace sellers. I think people buy from people. I've often said selling is the transference of belief. You actually get need to get the other party to believe that you can solve their problem, believe that you understand their challenges. Um, so sellers are always going to be vital to help drive growth within companies. What I want to do is just help them be stronger, faster, smarter, um, and still allow the parts that's human to be even more so. So I that's why I talk about it as the Iron Man suit.

SPEAKER_00 10:37

Very, very fun. Great, great analogy. Um, and we also focus in selling and sales on the top 10% of sellers, the superstars. Uh, I was never one of those, unfortunately, but you can't all be in the top one or five, five, or ten percent. You know, what about that middle 60%? You know, your average sellers helping them become superstars and improve, you know, their relationships with customers.

SPEAKER_01 11:03

Well, I think that's where the opportunity is. Um, if you sit down, you look at a bell curve of sales performance, the top 10 or 20% are gonna make club, they're gonna hit their number, they find a way every single year. Um you just need to help clear, clear brush out of their way and let them go faster. The bottom 10% is probably hopeless. You're not you're not gonna do much to get them to perform. So the value comes from taking the middle 60 or 70% and just moving them 10 degrees to the right and just getting them all to be a little better at what they do and and help them um help them perform better. And that's that's not just for the middle 60. One of

Iron Man Suit For The Middle 60%

SPEAKER_01 11:40

our customers tells me that it takes three years to onboard a new salesperson to get them effective. Wow, because they've got 10 literally 10,000 SKUs, and they're got a really complex distribution um model. Like we can get new sellers effective way more quickly by just pointing them in the right direction and not having them struggle by not understanding the catalog or understanding the channel very well.

SPEAKER_00 12:03

Great. And of course, it's still hard to hire salespeople. It used to be hard to hire engineers, developers. Now it's hard to hire salespeople. Go figure the human touch, judgment, experience, relationships aren't a commodity. So uh really interesting stuff. What are you excited about over the next weeks and months, quarters as we head into the end of the year? What's on your mind, your radar?

SPEAKER_01 12:23

Yeah, well, um, so we've delivered AI into our product in purpose-built ways. It's it's not what I call AI as a sidecar where you're just getting like, I don't know, advice or help or support. It's actually in the business workflow. Um, we've delivered some of that already. We have a big release in October where we bring a lot more of that to market. And it allows us to follow through on this idea of taking the 360-degree view of the customer and making it a 36-degree view or a 3.6 degree view and proving out that we can guide sellers towards the best opportunity. So seeing that come true, um, we're already, we have beta customers already demonstrating that this is really powerful. And I'm excited for um October when that comes to market, and we we see it in customers' hands.

SPEAKER_00 13:08

Well, very exciting times for sales sellers, and congratulations on all the success with Sugar AI onwards and upwards. Thank you so much, Evan. Really appreciate it. Yeah, thank you. And thanks everyone for listening and watching, and also check out our TV show, techimpact.tv on Fox Business and Bloomberg. Thanks, everyone. Thanks, David. Thank you.