From friction to flow: fixing the rep experience with AI

Julia Nimchinski:

And we are transitioning to our next demo session. Welcome to the show, James McArthur, VP of Product. at new.io. What a pleasure. How are you?

James McArthur:

I’m good, thank you so much, and also, thank you so much for the flexibility from yesterday, I greatly appreciate it. Yeah, I’m really excited to be here. As you mentioned, Jason Carther, VP of Product Advocacy and Customer Advocacy from New. I’m really excited to be here today. Thank you.

Julia Nimchinski:

We’re excited to host you. What’s the latest and greatest with new, and Where should we begin?

James McArthur:

Yeah, so, for everyone out there, New is a, Salesforce Native revenue architecture platform.

We are built… we are built with Agentic, and headless operations in mind, designed to work wherever your reps happen to be selling, whether that’s PLG motion, whether that’s selling through Salesforce or building out a platform anywhere else that you happen to be. The whole goal of NEW was to solve the problem B2C solved 20 years ago.

but for B2B sellers. And so, what I’m here to talk to you guys about today is something that we’ve released very recently, and I’m really excited about, because the entire world has been talking about kind of, how do you build Agentically? How do you create flows that sellers can use on a day-to-day basis every single day?

And new is… at the forefront of that when it comes to CPQ. So I’m gonna open this up. Yes, it’s in Salesforce, but then everything else I show you is built using our MCP tooling. We have lots and lots of different ways you can interact with this, whether that’s Slack.

Through… well, we have an external MCP, so wherever you actually want to hook it up. And so… I’m gonna go ahead, open up my chatbot.

And so, normally, when you go to create a CPQ, create a quote in a CPQ, the thing that first happens after you’re done rolling your eyes is the fact that you have to… go through, figure out what is the… what are the products I can sell? How much of them can I sell? What are the rules around it? And all of those are generally fairly opaque.

And then you have a system that has a hard time telling whether you’re actually selling a new business deal or not.

What I’ve built here is a guided selling process that I could build agnically, and I’ll show you that at the end of this, where I have not only… I’m not only automatically checking all of my active subscriptions, seeing if I have any net new opportunities. But I’m proposing the 3 bundles this rep can sell for this specific client.

I can filter this guided selling playbook to whether they’re Enterprise, SMB. I can have different rules for every single one of them. And so let’s say that… I want to sell… 125 standard licenses. Close it the third Thursday of next quarter. And, yeah. And so, thinking like a rep, We’re not… we don’t have… all the rules at our fingertips.

This isn’t something that we’re gonna do, but new is built on such a strong architecture. We’re built to work for… large B2B SaaS corporations, like OpenAI, where they have lots of different sellers doing all these different things, and it requires a ton of complexity, a ton of build.

And with new, I can very quickly and easily Using natural language, put in exactly what I want to buy, and knew it was going to guide me to the answer in a deterministic fashion. If I do this 100 times, and believe me, I have, I’m going to get the same answer every single time. And so right now, it’s giving me some options.

Maximum 75 licenses, Okay. I have to… Upsell. Okay, so let’s say my… I talk to my customer, they say, yeah, let’s go for Broom Enterprise. Quarterly billing, net 90. And, give me 150 licenses. But I want to see… 1, 2… And 3-year options. with a, 5% and 10% discount. for each… additional year. Now, in a normal system.

How long would it take you to create 3 quotes? I know, like, they can’t respond to me right now. The answer is a why. And so, with this, I was able to ask this. Okay, I would like 150 licenses, quarterly billing, because new is an end-to-end revenue architecture platform.

configuring my product, configuring my billing, configuring my quotes, configuring my billing. By the time I actually get to my billing, I have no questions. There are no problems for me. End to end. I’ve now got a close date, July 16th. No, no discount, 5% discount, 10% discount. And now it’s dug into my bundles.

It can actually read everything about the products I’m selling. Sure, I said Room Enterprise, but it knows there’s professional services with an add-on. There’s Platinum support. There’s optional add-ons.

And I can choose between two different API usage things, because new can support all of your consumption modeling, whether that’s credit burndown, cash burndown, commits, or just standard overage usage. And so, as I go. I can say, it’ll continue to ask me questions. Which AI credit options do I want? Let’s go PAYGO. And… Name it what you want.

And now, our Gentex solution actually takes the next step. Not only is it gonna say, here’s… your opportunity. It’s gonna do all the work for your rep. But it’s actually gonna build my 3 quote options for me. Immediately. It’s gonna say, here’s your first year, total amount, Second year. With my 5% discount, Third year, with my 10% discount.

All in one, you can see my year over year with ramping. It’s designed to be simple, it’s designed to be super easy. And then I can also do a quantity comparison and a quote comparison. on everything. End-to-end, it’s giving me the discount guardrails. Now, I run into the next thing that we run into with reps all the time.

Which is, hey, what happens if I discount too much? Well… with this, I’ll say, I want the… 3-year… Add in Platinum Support. And, give them a 34% discount. Now, a normal solution? Would say, great, let me get a preview for that, but… You’re not gonna have… you’re not gonna run into any issues.

And it’s not until you go and hit the submit for approval button. With our solution, it’s gonna lay everything out, give me a quote summary, and then be like, you know what? you’re gonna have manager approval on the other side.

We’re putting those guardrails in place for your reps, so you don’t actually have to worry about it, so that every single step of the way, they’re doing exactly what you want them to do, as you anticipate them Doing it. And so with this, I can very quickly go, yes please. And it’ll go save the quote for me.

Now… in our world, everyone here operates on skills. We’ve seen a number of people talk about skills. how they’re using Claude’s skills, all the different ways we apply skills in our daily lives is something that… Every single person here I’m assuming it’s doing that. But with this, I have baked into this… the Generate a Quote PDF.

It’s a part of new, we have a document generation, but we’ve built this out so that I can quickly say. Generate the PDF, And go. And this is gonna go ahead, load all my quote templates. and generate that lot… that PDF right away. Now, you’ll see here on the right.

I’ve got my lifecycle Manager MCP, Price Builder MCP, which means I, as an admin, I can go and I can change price, and it will authenticate based on my Salesforce permissions, so I don’t have to worry about overriding anything. And I can also pull in fellow AI when I’ve actually got the OAuth set up. And here’s our quote. Once it’s done loading?

There we go. And as I was talking and pausing, we were able to do this with a complex pricing engine, with price tags, everything, all within a few minutes. And all of it designed… Using natural language in a format we’re all very used to. Markdown. This is the actual behavior. So if I needed to modify it, all I’d have to do is hit one button.

Alright, that’s… That’s new, that’s my demo. Julia, what do you think?

Julia Nimchinski:

This is super impressive. James, we have, an interesting question from the community. How does this compare to Salesforce CPQ?

James McArthur:

Whoa. We can sell it? No, that’s a bad answer. No, the, the actual… the actual… from a Salesforce CPQ standpoint, we have… New is built for the modern age. New is built to run headless. So, when we’re talking, all the things you saw me do, all of that can be done via API. All of that can be put onto your website.

Everything you do on a day-to-day basis can be done using our MCP servers, it can be done using our Agentix solutions, or it can be just… if you want to build a website with pricing on it, like Chili Piper, you can go and do that. And it’s designed to create a lot of simplicity in your day-to-day workflow.

Our end goal, one year from now, and I’m comfortable saying this here, I’m pretty sure Mark said it yesterday. is… We don’t want the sales reps to ever know a CPQ existed.

Julia Nimchinski:

Amazing answer. Next question is, what’s the biggest challenge around implementation that customers run into?

James McArthur:

It’s the same… it’s the age-old challenge we’ve all run into, which is the data. Getting the data right, getting your products and… products configured, because what we run into a lot of times is customers will come in. They’ve got years speaking on the Salesforce CPQ question.

They’ve got years of bad practice built On a platform that didn’t work for their needs. And so what you end up with is… you end up with people who have built bad practice on bad practice. And then they get into a new solution, and they’re afraid of the change management that’s gonna drive for their reps.

So they try to lift and shift bad practice over to the new system. Or… Worse, their data’s terrible. And then we take… we start importing bad data into a new system, and it doesn’t matter how good a system is. If the data underlying it is bad. the entire outcome is going to be bad.

And so what we’ve done as a company to de-risk this situation is we’ve built, an agentic deploy tool that… where we can actually scan customers’ orgs as they come in to determine risks, to figure out, okay, what are the configurations they have? that are potential risk for this platform.

It also will scan their products and pricing, and see what is the best practice for us to migrate this over. And so, what we’re doing is to solve the age-old problems everyone’s had. Since forever? We’re creating the Agenthic tools to actually do that without needing the people. And eventually, the end goal is that this becomes a hit a button.

migration… the sandbox done. Not… we’re not migrating to production right away. Please don’t make us do that. It’ll make me cry.

Julia Nimchinski:

Tricky question. The question is around, basically, how do you prevent AI-generated quotes or pricing recommendations from creating compliance or margin issues?

James McArthur:

Well, that’s our approvals Engine. So, New… Approvals Pro from New is an approvals Engine built in… built as… as a part of this entire flow. So, sitting behind everything I showed you, yes, I showed you a quote, yes, I showed you a PDF, all of that was for this demo, because I had 15 minutes to do a thing.

But… In the real world, I have, parallel and sequential approval sitting behind everything. All of those special things I said, net 90, Quarterly billing? a 34% discount, there’d be a total of 4 approvals generated off of that, in that system.

And so that rep would not be able to generate a PDF, and they would end up running into those approval flows almost immediately.

Julia Nimchinski:

And last question, James. The question is around, working with the platform itself. Is it… do you always need, technical users, admins, folks are asking, or it’s just completely adjustable and you can use it as a business user?

James McArthur:

The answer is… For a technical business user. you could get by. Like, I was an accidental admin at one point. As an accidental admin, I could do this for an SMB. Fine, great. But… At scale, no. At scale, this requires, admins who understand Salesforce, understand permission sets, understand how APIs work.

And, while our Agentix solutions are incredibly easy once configured, at the end of the day, you do need to understand how your product and pricing work, you do need to understand how your data architecture is structured, so then you can make best use of what we’ve built.

Julia Nimchinski:

Fantastic demo. What’s the best next step?

James McArthur:

Best next step is you can always go to new.io and register for a demo, or if you have personal questions for me and want to bug me about stuff, my email is james at new.io. It’s, incredibly difficult. But, but yeah, I’m always happy to talk. I’m also on LinkedIn, where you can find me, and I’m… Open door. Anyone here?

Julia Nimchinski:

Thank you so much, James.

James McArthur:

Thank you.

Julia Nimchinski:

Thanks, and that’s a wrap for our summit, Agent to Agent GTM. Thank you to every speaker, sponsor, moderator, all of you watching. As you know, we explored one central idea, the last three days, agent-to-agent GTM, and we are returning on July 30 and inviting you to join us again.

It’s going to be a practice session for some of the elements and some of the Friday’s conversation here today, and all over the three days, we’re gonna be basically showcasing that. Executives, for executives, thank you again, and see you soon.