SPS Commerce, Inc.

Good day and welcome to the SPS Commerce second quarter 2026 earnings conference call. All participants will be in a listen-only mode. Should you need assistance, please signal a conference specialist by pressing the star key followed by zero. After today's presentation, there will be an opportunity to ask questions. To ask a question, you may press star then one on your telephone keypad. To withdraw your question, please press star then two. Please note that this event is being recorded. I would now like to turn the conference over to Armina Blaszczyk, Investor Relations for SPS Commerce. Please go ahead.

Good afternoon, everyone, and thank you for joining us on SPS Commerce second quarter 2026 conference call. We will make certain statements today, including with respect to our expected financial results, go-to-market strategy, and efforts designed to increase attraction and penetration with retailers and other customers. These statements are forward-looking and involve a number of risks and uncertainties that could cause actual results to differ materially. Please note that these forward-looking statements reflect our opinions only as of the date of this call and we undertake no obligation to publicly update and revise any forward-looking statements, whether as a result of new information, future events, or otherwise. Please refer to our SEC filings, specifically our Form 10-K, as well as our financial results press release for a more detailed description of the risk factors that may affect our results. These documents are available at our website, spscommerce.com, and at the SEC's website, sec.gov. In addition, we are providing a historical data sheet for easy reference on the investor relations section of our website, spscommerce.com. During our call today, we will discuss adjusted EBITDA financial measures and non-GAAP income per share. In our press release and our filings with the SEC, each of which is posted on our website, you will find additional disclosures regarding these non-GAAP financial measures, including reconciliations of these measures with comparable GAAP measures. And with that, I will turn the call over to Chad.

Thanks, Hermina, and good afternoon, everyone. Thank you for joining us today. At SBS Commerce, our foundation has always been our cloud-based supply chain network. Today, our network stands as a massive, interconnected retail ecosystem of tens of thousands of suppliers and 3,500 buying organizations, including all the major retailers and distributors in North America. We work with more than 2,000 logistics providers and over 400 technology partners. which enables us to integrate our network with all of our customers' supply chains and business systems. We are protocol agnostic and enable fulfillment models and channels with enterprise-grade security certifications. That foundation makes everything that follows possible and represents every AI use case on our network. The SPS Commerce network took over 25 years to get to where it is today. Through its effect and scale, we are building partnerships, supporting evolving supply chains, and helping our customers grow. Having recently divested the 3P revenue recovery business, we have sharpened focus on strategic relationships with 1P suppliers who operate multi-retailer trading relationships and benefit from our intelligent supply chain and portfolio solutions. As the network expands, we continue to capture proprietary intelligence from trading partner activity, transaction patterns, digital specifications, and compliance rules, strengthening the supply chain rules engine that powers Max, SPS's AI agent. By leveraging SPS's network intelligence within everyday workflows, Max enables customers to interact with their supply chains in a more intuitive, Proactiv, and Connected Way. Users can instantly compare business requirements and business performance between major retailers like Target and Costco. With Proactiv monitoring, Max serves up 24 by 7 extension of a customer's team, detecting anomalies and flagging critical business errors. Max puts the expertise of the SPS network at the customer's fingertips. to instantly diagnose business issues and determine actionable solutions, shortening the time it takes to address risks in trading partner relationships. For example, Branch Furniture is a fast-growing wholesale brand selling to major retailers like Williams-Sonoma, Lumen and Office Depot. They rely on SPS Commerce to manage order flows across multiple channels and have already experienced significant efficiencies using Max to resolve order issues in minutes as opposed to days. For one of their key retail partners, Max helped achieve 90% weekly time savings in managing overdue orders. Other customers have recognized tangible results since Max's beta phase launch. Max successfully caught a $290,000 invoice failure due to an incorrect UPC code. It identified 100 stalled dropship orders for an outdoor brand. It flagged $70,000 in unacknowledged purchase orders for a food manufacturer. Delivering this immediate ROI, Max is quickly becoming the default starting point for customers inside the SPS user interface. They trust Max's proprietary supply chain expertise, and they are increasingly allowing it to take automated actions on their behalf. continually improving operational efficiencies with their training partners. By pairing SPS's network intelligence with our agentic capabilities, we completed our first AI-powered customer onboarding, including pre-sale contacts and account provisioning. We're working toward a future where agentic technology can engage in new customer immediately after a deal closes with more of the onboarding processes shifting to AI as we continue to reduce the time it takes for customers to transact with their trading partners. Agent-assisted customer functions and onboarding, as well as the agentification of our internal operations, are the two pillars in our agent strategy already in motion at SPS. We are also exploring new AI-powered use cases and products which we believe will drive ARPU expansion and increase the size of our addressable market. The initial launch of MAX to all SPS fulfillment customers is expected by the end of the summer, and we plan to launch additional products at scale later this year. One of the key learnings from our beta program is that users of Max through the chat interface are more likely to explore advanced Max features, and we expect this usage trend will define the path to monetization of our AI solutions. We are excited about these AI capabilities and the immense value they will bring to our network, and so are our customers. In a recent study of SPS customers, We quantified and validated the value and impact SPS delivers to their business. 83% of customers said that the data in the SPS network improved their AI readiness. 87% said it improved scalability. And 100% of the surveyed customers said that without the SPS network, they would need more headcount, more tools, and more time, or in some cases, Thank you for joining us today. Over their decade-long relationship with SPS Commerce, they have scaled from one trading partner to dozens of customers, multiple 3PLs, and a growing supplier network. To prepare for their next phase of growth, Chosen Foods migrated to a new ERP and trusted SPS to manage the transition. Through a fully integrated Acumatica deployment, SPS Commerce delivered a unified approach across their order-to-cash Procure2Pay, and revenue recovery workflows ahead of schedule and with zero operational downtime. Crucially, with growing deduction complexities across major retailers like Walmart, Amazon, and Target, SPS's automated dispute management successfully recovered approximately 30% of outstanding deductions. which represents hundreds of thousands of dollars while helping chosen foods identify why these deductions occurred and how to prevent them. Other customers realizing real ROI from SPS revenue recovery include Owlette, a leader in infant health technology, recovered $1.4 million within six months of using the solution, including 100% recovery on our recent settlement, totaling $423,000. Serta Simmons Bedding, one of North America's largest bedding manufacturers, saved $200,000 by successfully challenging a post-audit with a large retailer. Turning to our analytics business. SGS's new analytics solution is now running on a new enhanced platform that delivers significant gains in both power and scale. This new platform brings an improved user experience while enabling faster time to insight so customers can move seamlessly from data to decisions. It expands what's possible for customers, supporting growing data volumes, broader use cases, and future AI predictive capabilities. With these platform enhancements, our analytics solution helps customers protect revenue, margin and shelf space by catching risks early while uncovering new growth opportunities across products, customers, markets and distribution. It also gives teams the agility and efficiency to act sooner, align inventory, forecasting and planning while strengthening retailer relationships with a single view of performance. Ruffle Butts, a children's clothing company based in Texas, is leveraging the platform to gain significantly better sell-through visibility into one of the nation's largest retailers, capturing critical insights from data across more than 400 retail locations and the retailer's e-commerce channel. To sustain this momentum, automated data feeds and scheduled reporting will drive ongoing day-to-day analysis. Early feedback from RuffleBus on the platform's granular product and location insights has been highly positive, prompting this supplier to consider adding another major retailer to their reporting. In summary, SPS's customer success stories demonstrate that navigating today's increasingly complex supply chain requires an intelligent network. As businesses continue to expand across technology platforms and connect with new trading partners, they view SPS Commerce as a vital partner for scaling their operations and improving AI readiness. No other company can match the unique combination of AI capabilities 25 years of proprietary data, deep domain expertise, and an expansive network access to drive this kind of tangible value and collaboration that SPS offers today. With that, I'll turn it over to Joe to discuss our financial results.

Thank you, Chad, and welcome everyone. We report a strong second quarter of 2026. SPS Commerce's core business, which excludes the invested 3P revenue recovery business, grew into high single digits, driven by the acceleration of 1P customer ARPU growth, resulting from continued upsell and cross-sell momentum. On June 30th, we announced the sale of the 3P revenue recovery business. We believe this divestiture sharpens our focus on the strategic opportunity with 1P suppliers, who operate multi-retailer trading relationships, and are positioned to benefit from our intelligent supply chain network and purchase additional solutions like fulfillment, revenue recovery, and analytics. SPS Commerce received a cash payment of $9.5 million at closing, and we incurred a loss on sale of $23.5 million in Q2 2026 in connection with the transaction. Now let's review our Q2 results. Revenue was $197.8 million, a 6% increase over Q2 of last year. Recurring revenue grew 6% year-over-year. As a result of the sale of the 3P revenue recovery business and its approximately 7,300 customers, the total number of recurring revenue customers in Q2 was approximately 46,650, and an average revenue per customer was 15,100. In Q2, ARPU skewed higher due to the divestiture's impact on our ARPU calculation, which used an average of beginning and End of Quarter Customer Counts. Because the quarter end investiture significantly reduced our final customer count, Q2R proof reflects full period revenue divided by a lower customer base. Adjusted EBITDA increased to 66.6 million, highlighting the health of our business as we scale. Strong operation execution, the realization of past investments and benefits of improving process efficiencies. Turning to liquidity and cash flow, We ended the quarter with total cash and cash equivalents of $173 million. Pre-cash for the quarter was $57.4 million, bringing our trailing 12-month pre-cash flow to $198.7 million, up 40% year-over-year. In Q2 2026, we deployed nearly 90% of pre-cash flow to repurchase 51.2 million of SPS shares. Now turning to guidance. As a reminder, As a result of the divestiture of the 3P revenue recovery business on June 30, 2026, guidance factors in a reduction of approximately $10.5 million to revenue to the second half of 2026. The divestiture is expected to be neutral to adjusted EBITDA in the second half of 2026. For the third quarter of 2026, we expect revenue to be in the range of $196.3 million to $198.3 million. We expect adjusted EBITDA to be in the range of $67.4 million to $69.4 million. We expect fully diluted earnings per share in the range of $0.72 to $0.76 with fully weighted average shares outstanding of approximately 36.8 million shares. We expect non-GAAP diluted income per share to be in the range of $1.20 to $1.23 with stock-based compensation expense of approximately 16.4 million, depreciation expense of approximately 5.4 million, and amortization expense of approximately 8.5 million. For the full year 2026, we expect revenue to be in the range of 788.4 million to 793.4 million, representing approximately 5% growth over 2025 at the midpoint of the guided range. Excluding the impact of the divested business, We expect our core business revenue to grow high single digits. We expect adjusted EBITDA to be in the range of $264.6 million to $269.1 million, reflecting an adjusted EBITDA margin of 34% at the midpoint, and an increase of approximately 300 basis points compared to full year 2025. We expect fully diluted earnings per share to be in the range of $2.24 to $2.33, with fully diluted weighted average shares outstanding of approximately 36.9 million shares. We expect non-GAAP diluted income per share to be in the range of $4.84 to $4.93 with stock-based compensation expense of approximately 69.8 million, depreciation expense of approximately 23.4 million, and amortization expense for the year of approximately 35.6 million. For the remainder of the year, on a quarterly basis, investors should model approximately a 30% effective tax rate calculated on GAAP pre-tax net earnings. In summary, SPS's strong second quarter performance reflects the strength of our core business, driven by upside on cross-sell momentum. We continue to demonstrate operational rigor, exceeding our margin expansion goals, while simultaneously rolling out our AI strategy across our network. With that, I'd like to open the call to questions.

Thank you. And ladies and gentlemen, we will now begin the question and answer session. To ask a question, you may press star then 1 on your telephone keypad. If you're using a speakerphone, please pick up your handset before pressing the keys. If at any time your question has been addressed and you would like to withdraw your question, please press star then 2. Our first question today will come from Scott Berg with Needham. Please go ahead.

Hi, Chad. Hi, Joe. Nice quarter here. I got a couple questions. Chad, first of all, I want to talk about the divestiture of the third-party revenue recovery business. You've been pretty positive on the long-term outlook of revenue recovery in general, and I know that part's been a little bit of a thorn in your side, but why divest it? Why divest it now or obviously a month ago? Just help us understand the thought process to move on from that segment.

Yeah, Scott, so, you know, overall, we remain very confident in revenue recovery. We're seeing the cross-selling to our fulfillment customers be good and also seeing new business come in as this is kind of a new emerging category of SaaS solutions. Where more of that positivity was, though, is on the one-piece-eyed seller. So Those that are selling primarily wholesale to multiple retailers, Amazon being one of those, but the 1P sellers really can use our whole portfolio revenue solutions across multiple retailers. Whereas the 3P business, those were more Amazon marketplace sellers. There didn't turn out to be a lot of overlap with the other parts of our portfolio for those customers. I think that combined with the take rate revenue model and some of the policy changes in 3P we saw from Amazon all clearly pointed out that the 1P side of this business is much more attractive for us and has much more overlap with our ideal customer profile than the 3P side does.

Got it. Helpful. And then, Joe, I think we kind of probably understand the and a number of customers that are leaving the platform with the divestiture. ARPU seems to be moving around, but I guess I've got a couple questions on the ARPU side. One, are you calculating any differently than how the company has before? I only ask because you took a bunch of revenue in the quarter, but obviously lower customer accounts exiting the quarter. And then, I guess secondly, in conjunction with that, how do we think about the impact going into Q3 because of the revenue shutdown?

Yeah, so we didn't calculate it any differently. And I think, you know, because we kept the calculation consistent, it kind of skewed in the quarter. And the reason for that was, if you think about the way the calculation works, Scott, it's the average customer count in the beginning and the ending of the period. And so we had those 7,300 3P customers in the beginning customer count, but they weren't in the ending customer count. and then that's compared against the revenue in the quarter and the full 3P revenue was in the quarter but not the ending customer count and so because of that, the ARPU overall skewed higher than it normally would have and so that's just, it's more of the impact in the quarter. Going forward, if you think about it, we'll just have 1P customers in the beginning and end of the period and so it'll be a little bit more consistent going forward than it was in Q2.

Awesome. And if I may, a quick third question here, sorry, out of etiquette. Joe, can you quantify what the third-party revenue recovery revenues were in the second half of 25? I know you said the business is going to grow high six digits here the rest of the year, kind of excluding that, but any further kind of quantification of that number I think would be helpful. Thank you.

Yeah, Scott, so the only other color we're providing on the 3P business outside of the fact that, to your point, that outside of the divested business, then we'd be growing high single digits. I think the other thing to pull out on the full year is, you know, we pulled out the 10.5 in the second half of the year, and you can assume the first half of this year was slightly lower than that, so you can kind of get a full run rate of the business for 2026. Thank you.

and our next question will come from Dylan Becker with William Blair. Please go ahead.

Hey guys, this is Jackson Bodley on for Dylan Becker. Maybe sticking on the revenue recovery side, now that the focus is solely on the one-piece side of that business, how are you thinking about the level of resources and investment dedicated to that business going forward? Is there more resources being redeployed toward fulfillment and analytics or does The retained 1P opportunities still warrant the incremental investment from here.

Yeah, so Jackson, the 1P business, I would say, is nearing consistent with our overall margin profiles in the business overall. It wasn't that way right out of the gate with the acquisition of Supply Pike and Carbon 6, but as that has gotten more integrated into our overall business, It's more approaching our overall margin profile. So I wouldn't say it's an area of our business that is sort of receiving, you know, oversized investment at this point in time. And I think the divestiture of the 3P side of that business really helps us, I mean, because there's quite a bit of good customer overlap. product portfolio overlap on the network with the 1P side and definitely think that revenue recovery business is definitely in line with the margin profile of our overall business.

Got it. Super helpful. And then maybe as a follow-up, with ERP, migration is still creating a little bit of timing noise. I mean, I would just be curious to get your thoughts if you guys are seeing any change in Onboarding duration, I know you guys talked about the AI-enabled customer onboarding. So is that changing anything with the customer readiness or attach rates once those projects are complete? Or maybe are there areas where migration delays are building pent-up expansion demand that could release once these go live? Thanks.

Yeah, so we're super excited about the progress around agentic onboarding. We did have in the prepared remarks that we did the first fully agentic onboarding. Now, keep in mind, that's with the more kind of simple onboarding that we have. You know, that's really taking things that would have been previously done in days and getting them down to kind of minutes. With the more complex onboarding, which is really where we have all the ERP integrations We do expect that we will continue to make great progress there. We've been making progress there over the last couple years, speeding that up. That's led to a better customer experience. It's also helped us on the gross margin. And that was really done all before this agentic capability was applied. So we do expect to speed up those more complex ERP onboardings as well. There's just still a little bit more work to do there. Once we have that in place, That speed to time on the network can be a barrier for adopting the SPS network. So we think any efficiencies we gain there will help with customers and speed up that access to the network. I wouldn't necessarily say that there's substantial pent-up demand just kind of waiting for this. Admittedly, the ERP market has been a little bit slower in 2025 and so far this year, especially at the kind of medium to large end of that market. But I do think our speed of onboarding with agentic ERP onboarding is going to be a massive differentiator for us and really speed up customer time to value.

And our next question will come from Chris Quintero with Morgan Stanley. Please go ahead.

Hey, Chad. Hey, Joe. Thank you for taking the questions and the nice execution here. I want to hear your thoughts on maybe the macro environment and kind of what you're hearing from your customers. You know, we're hearing about, you know, higher fuel costs, higher freight costs, you know, the K-shaped economy. So just curious kind of what you're hearing and seeing high level from your customers from a macro perspective.

Yeah, Chris, I mean, I would say, you know, no substantial headwinds we're hearing from our customers relative to the macro. You know, we were coming off a tougher 2025, especially on the supplier side of our network where they did cite some headwinds related to tariffs and that, you know, did cause some contract rightsizing last year. We anticipated that that would dissipate this year as we kind of did get those contracts right size and they were one time and that's playing out as we had expected. And so I'd say, you know, no overwhelming headwind in the macro. Of course, things like the fuel prices and still a little bit of looming tariff uncertainty are things that we continue to monitor, but those things are not coming up in our engagement with customers right now.

Got it. And then maybe go for you, just on the 1P customer counts, if I have my math right, it seems like that went down around 200 quarter-over-quarter. Is that right? And if so, curious what you're seeing on the community enablement side of things and new customer ads.

Yeah, no, your calculation there is right. We were down a little over 200 sequentially on customer counts. The driver of that was really just the timing effect of some of the retail enablement programs. Keep in mind, those customers that are typically churning or adding that are primarily affecting that customer count tend to be the real low ARPU customers. That's why we're able to still deliver the financial results, even having that customer count there. The overall pipeline for an enablement activity right now is strong. There's programs that we're running now that will contribute in the second half, plus the remaining pipeline that's to be closed in the second half looks positive. That said, I would expect for the year we're kind of flat to slightly positive on customer count, but I do expect some of that momentum from the second half enablement programs will carry into early 2027. Excellent.

Thank you so much. and our next question will come from George Kurisawa with Citi. Please go ahead.

Okay, great. Thanks for taking the questions here. Maybe if I could just ask about the max beta. You had some interesting anecdotes of customers saving in some cases it sounds like hundreds of thousands of dollars. I think you've done some work on market sizing. Maybe you could just share updated thoughts there on how you're thinking about a potential uplift maybe in a best case scenario or for a median customer. and then how that maybe has evolved your overall thinking on packaging and pricing as the product portfolio continues to expand.

Yeah, absolutely. So, yeah, as you noted and was in the script, we're seeing customers really identify different supply chain anomalies and disruptions using MAX, which today is, you know, through the chat feature. That's what they have access to in the beta. and using that chat feature, they're able to get to some of those problems in the supply chain, get them resolved and that's resulting in hard ROI savings for them. What we've seen through the good adoption of chat here is that a lot of the things that customers are doing via chat would be possible to automate with an agent. So today, it may take them 20 prompts in the chat to get to the right answer. We're seeing that that's something that actually could be automatically detected and potentially, in some cases, automatically resolved, which is great because We are developing those types of agents on top of this MAX technology now. And we believe that those agents that can do things more autonomously in terms of identifying these anomalies and, in many cases, resolving them, not only finds the kind of hard ROI in the supply chain savings, but also is going to be a very favorable kind of headcount and efficiency impact for our customers. So what we're in the process of now is converting the chat piece All newly deployed customers as of the last month have been onboarded with MAX included. And over the course of the next several weeks here, kind of through the summer, we'll be making it available to all our other fulfillment customers. And we'll be doing that as part of their standard subscription. But what we believe the major monetization activity will be is when we deliver those agents on top of that that are more autonomous and self-acting, that customers will be willing to pay for that. And that's really where the monetization would come in. And the way that that would work is there would be certain tiering or bundling of the packaging of those autonomous agents running on top. And then we would monetize the customers through subscriptions to those bundles. but what I'd say is gives us high confidence in this approach is we're already seeing customers using MaxChat to get to these benefits in their supply chain and the things that they're finding and doing, we have high confidence we'll be able to automate with the agentic architecture over the top.

Okay, that's great, Keller. And then one for Joe, if I may, just looking at the change in guidance for the second half, it looks like, from what we can tell, On the revenue side, it looks like basically the Q2 beat flowed through, excluding the divestiture impact. On the EBITDA side, it looks like the full beat was not flowed through. So I wonder if you could just maybe comment if there's anything, incremental spend, expense timing, conservatism, anything we should keep in mind on the EBITDA line.

Yeah, for sure. I think on the EBITDA side, I think there's a couple of things that we contemplated. One, you know, there was some movement of some of the expenses that moved out of Q2 into Q3 and Q4. And so that was some of it. I think the other piece is we want to make sure we're being very prudent with the way we're approaching our internal AI costs. As we're building out this stuff for Macs, as we're building out our internal agents, on the things we're doing internally. We want to make sure we give ourselves enough room to make those investments and make sure that we've got enough flexibility in the cost structure. And so that's the other part of that and why we didn't flow all that through the year, George.

Okay, makes sense. Thanks for taking the questions.

And our next question will come from Parker Lane with Stiefel. Please go ahead.

Hi, guys. Good afternoon. Thanks for taking the questions here. Chad, you talked about some of the advancements you're making on the analytics side of the house. Sounds like there's a new enhanced platform there, so good to see that. I think the revenue side, it was up maybe a percent in the first half of the year. Can you just talk about what you're seeing from a demand perspective around that? I know you had mentioned that historically it's been seen as maybe more discretionary, and that was an impact to that business last year. But looking to the second half of the year, what are your expectations around analytics?

Yeah, so we're really excited about this new technology revamp. I mean, I do think it will help us on the sales side. Some of the previous technology had gotten a little stale, a little dated. Our feedback from customers who are up and running on this new capability is one. Just the look and feel and ability to use the system and the prebuilt capabilities are much stronger than they were before. Plus, there's more tooling for customers to kind of do more on their own. And then probably the most important thing in all this is it really changes the underlying data architecture of that product, which now sets it up for many more AI features that we'll be able to add to that over time. So we are optimistic about that outlook for the analytics business. I think the fact that it is a little bit more discretionary is true still, but I think with this replatforming, not only will we be in a maybe a little bit stronger competitive position, but we should also be in a position then to add more AI features, which I believe we'll be able to monetize over time.

Got it. And we're coming up on two years of the entry into the first part of your revenue recovery. I think at the time there was about 300 customers that overlapped with SPS. How have attach rates or adoption rates trended at the two-year mark relative to back then, and what are some of the learnings you guys have had on the go-to-market front on how to effectively cross-sell both into the historical supply pipe base and back into SPS's base?

Yeah, absolutely. So we've had success in both directions, selling fulfillment to supply pipe customers. Obviously, that's not as big a population, so it's been a little bit less impactful. But the big win has been selling the supply pipe and really now the Amazon 1P that came out of Carbon 6 to the fulfillment customers. And, you know, we've... kind of hardened that muscle, I'd say, around cross-selling in the organization. We've done some things organizationally to have that work a little better. We've done some things with the sales team's incentives. And what I think is really powerful in all this is just the signals we get from the network, right? So the network actually tells us, based on trading volumes and trading partner relationships, who's the most likely candidates. in fulfillment for revenue recovery. And using that data, we're able to specifically go and target those customers. In some cases, come to them with an estimate even just based on our network data on what the potential is for them to recover. And I think this is critical for us going forward. I mean, we've been clear that we expect to drive a higher proportion of our growth on the ARPU. And of course, there's a big opportunity for more connections for fulfillment customers, but cross-selling our analytics and revenue recovery solutions to those fulfillment customers is key to that ARPU growth as well.

Great. Thanks, Chad.

And our next question will come from Matt VanVleet with Cantor. Please go ahead.

Yeah, good afternoon. Thanks for taking the questions. I guess following up on some of your comments, Chad, about the max monetization, I guess curious on what you're kind of baking in in terms of the adoption cycle for existing customers. And then when do you plan to have some of these bundles in place? And, you know, I guess early stage, but what are you expecting as sort of the uplift if existing customers plan to adopt, you know, whether it's a middle or high tier, like how much uplift can they get on an annual basis?

Yeah, great question. So in terms of the adoption, I mean, if we're to judge it based on the MaxChat adoption, I believe we'll have real strong agent adoption because we're already seeing customers sort of, if they're onboarded with MaxChat, it's quickly becoming like the main interface point that they use when using any of our applications. They're just sort of starting in MaxChat. and, you know, through that then I can, I believe that as some of the things that they're doing in MaxChat, we're able to automate with agents. There will be strong interest in having that all be automated so they don't even need to interact that much with the chat interface. They still can, but some of the things that are happening on a daily basis or weekly basis will just get automated with the agents. In terms of the timing of all of that, We expect that we will be in a position to be selling agents kind of by late Q4 of this year. Now, obviously, that will take some time to flow through to revenue, but we do think we'll be in a position where we're actually monetizing this agent architecture still here this year. Now, the kind of degree to which we're able to kind of do uplift on ARPU, that's some of the details that we're working through right now. I do think the first set of agents that we put out are going to be probably more addressable for the more highly complex customers with more trading relationships. And over time, we'll be able to bring that back down to more of our medium and small customers over time.

All right, helpful. And then, Joe, you mentioned on some of the cost structure of internal AI usage. Maybe just help us with the timeline of When internally you were really pushing that aggressively for a good portion of the employee base, just to get a sense for sort of when we might lap that and when growth could provide some operating leverage in the model, whether it's later this year, in the next year, beyond that?

What I would say there, Matt, is a lot of the leverage we're seeing out of the business right now is not based on some of the AI internal use cases that we're starting to talk about. I think a lot of the efficiencies you've seen in this business have really been driven by economies of scale, just being more operationally efficient over the last 12 months, people looking internally and making sure we're optimizing each of our processes. So I feel really good about how we've somewhat structurally changed this business going forward without Using AI, and if I think of the go-forward and some of the things we've talked about, about the onboarding process on the go-to-market side, we believe that those will all be added to some of the things we've already been able to accomplish without the internal use of AI. So we feel good about the trajectory of the margin going forward, not only this year, but going into next year. And as we exit this year, Matt, we'll have a little bit more color on how we think that probably impacts more of the longer-term focus of the business.

All right, great. Thank you.

and our next question will come from Mark Chapelle with Loop Capital Markets. Please go ahead.

Thank you for taking my question. Chad, you've had a new chief commercial officer on board now for a couple of quarters. I was wondering if you could just talk a little about maybe some of the changes that have been made or adjustments that have been made to the sales structure, maybe like customer segmentation or just even the coverage model for that matter.

Yeah, I would say we did evolve certain things in the go-to-market. They were kind of happened to be in conjunction with Eduardo's arrival, but I think he's all in line with that. You know, some of the things I mentioned earlier around driving a little bit more focus on cross-sell and aligning that as part of our incentives. We've also done some things to segment the sales force a little bit more between new and existing. That has worked effectively, especially on the retail side. And the other thing I would say is Eduardo and his team on our customer success are also responsible for all the customer onboarding activity. And that's an area where we've seen quite a bit of success. are continuing to drive more success as we agenify that onboarding process. So very pleased with the way that Eduardo's come in. He has brought some new ideas to the organization, having worked at some previous very scaled software businesses and just helping us overall mature our capabilities around go-to-market. And I will add, too, part of that is marketing. We brought in a new chief marketing officer. She's really helped us on some of the demand generation things. I mean, the company's been kind of in a luxury position to, you know, pretty much solely rely on these retail enablement programs as the source for new customers. We believe that there, over time, will be opportunity to drive more new customers through more traditional digital marketing capabilities. and that's something that Maria has brought into our organization. So the combination is working quite well. Thank you.

And our next question will come from Jeff Van Rie with Craig Hallam. Please go ahead.

Hey, this is Daniel Allen for Jeff Van Rie. On the beat this quarter, you know, the last few quarters have been a little bit more in line. Congrats on this quarter. Real nice beat on the top and the bottom. Just what played out in the quarter that drove, you know, the more than expected strength here in Q2?

Yeah, I think a couple of things. One, we talked about this coming out of Q1. We're not seeing the same amount of pressure, especially on the downsell and gross retention that we saw throughout 2025. So GRR continues to be a real strength of ours. That continues to You know, grow year over year and feel really good about the progress we're making on that front. And then, you know, we start to see, you know, more momentum within our existing customer base and adding new trading partners. I think we've talked about the land and expand model continues to be a big driver of our growth overall. And so I think the combination of our ability to expand trading partners within our existing customer base and then, you know, the positive momentum on the GRR side with the two big drivers on the revenue, you know, over performance. Okay.

And then on the customer count, obviously that's skewed by the 3P customers exiting the count. But in terms of just the 1P counting down 250 sequentially, just thoughts on that? Any updated thinking on expectations for customer growth? Anything to change there? Thanks.

Yeah, that was just really due to some timing of the retail enablement programs and how they contributed to customer count in the quarter. I'd say overall, The retail programs that are up and running and those are in the pipeline that we have high confidence in for the second half. You know, that all looks pretty positive. So I would expect the second half to contribute sort of a positive customer count, but kind of coming in on the year, probably, you know, kind of flat to slightly positive on the customer count.

Okay. Thanks, Chad. Thank you.

and our next question will come from Lachlan Brown with Rothschild and Company. Please go ahead.

Hi, Chad, Joe. Thanks for the questions. With your Max Beta customers, could you just run us through your confidence in being able to convert them when you make Max generally available at the end of the summer? Could you talk us through the go-to-market playbook that's in place to transition these accounts at launch? And I guess any feedback from preliminary customer discussions would be helpful. Thanks.

Yeah, so let me start with the preliminary customer discussions. In this beta, we've been very engaged with customers. I think you can see from some of the detailed examples that we shared in the prepared remarks, we're really engaged with customers, understanding the ROI that they're getting out of MAX. And I'd say, you know, this is one of the nice things about having a tool like this. I mean, we see all of their interactions. They're able to score their interactions. We have a separate agent that on top of their scoring goes in and scores the interaction. So we really can narrow in and see where customers are getting value out of the Max Chat capability. In terms of kind of then upselling them from Max Chat, which we're using kind of as a gateway into our overall Max architecture, we're going to target those probably larger, more complex customers that have high usage of Max Chat. and work with them to convert some of the things they're doing with MaxChat into autonomous agents that will just take care of those things automatically for them. And we think between the ROI that they're driving out of their supply chain and the efficiencies they get then from converting over from chat into an agent, an autonomous agent, that they'll be pretty high conviction from customers to move over to the more agentic approach, which will be monetizable.

Thanks. And looking at the implied Q4 revenue from the outlook, it suggests a nice step up from Q3. Could you just help us unpack the main building blocks behind that acceleration? For example, are there any specific enablement campaigns scheduled for later than in the year that give you that visibility?

Yeah, I'll talk through a couple of things, and then I'll have Chad talk about it a little bit more on the enablement campaigns. I think a couple of things are going on in the business. One, I just I talked about it a little bit earlier. The momentum we're seeing on the GRR side, so we continue to see improvements across our customer base. And so we're in a much better position, I think, going into Q4 and the momentum we're seeing there than we were a year ago. So I think that's the other big driver. And then the second thing is, you know, is on the enablement side. We're seeing more of these campaigns come through. We're seeing momentum in the back half of the business. We have a really strong pipeline. And so we believe there's going to be a solid number of these customers that land in Q4 that's really kind of driving that revenue in the quarter.

Yeah, I mean, I would just add, you know, although we see some positivity there, you know, kind of the big drivers in our revenue performance to finish out the year here are going to be more probably driven by the strong GRR that Joe mentioned and the ARPU expansion. We do expect to be positive on the customer count. But, you know, the customer count that we drive through these retail programs, certainly while important, we want to get customers. We want to further penetrate that TAM. Those tend to be very low ARPU customers when they come in the door. So they're meaningful over the long term, but not as meaningful in the short term to drive revenue.

That's clear. And congrats on the quarter, guys. Thank you. Thanks, Markman.

And our next question will come from Nehal Chokshi with Northland Capital Markets. Please go ahead.

Pardon me, your line is open.

Sorry about that. Thank you. Congrats on a good quarter. And congrats on the, as well, that the implicit acceleration in the business in the back half, especially in the 4Q here. And sounds like it's going to be driven by the improving GRR that you're seeing. Is that, is the driver of improving GRR max or is it something else?

Yeah, I would say it's a combination of things. I do think a little bit is macro. We did see some headwinds last year in our customer base that drove them to kind of right-size some contracts. We're not seeing that this year. The other factor is I believe, you know, we've made some improvements in our customer treatment strategy. I mentioned that both on the onboarding and also the way that we've organized the sales force. to have a little bit more attention, I'd say, to existing customers. And I think the new innovation that our customers are seeing us with Max, with adding revenue recovery to the product portfolio, with investing in our analytics product, I think these are all things that show to our customers that they want to be a long-term partner with SPS Commerce. Great. Thank you very much.

and once again, if you would like to ask a question, please press star then one. Our next question will come from Clark Wright with DA Davidson. Please go ahead.

Thank you. If we look at the growth mix after the 3P revenue recovery divestiture, how much of the growth now do you expect to come from ARPU expansion versus customer additions?

Yeah, Clark, what we've said is, you know, kind of in our growth algorithm over the long term, you know, we expect kind of, you know, roughly one-third of the growth to come from the customer count side and two-thirds to come from ARPU. This year, it'll obviously probably be slightly more on the ARPU side. And then if you were to take that to our And then can you help me understand, in your prepared remarks you mentioned that SCS Commerce is uniquely positioned to provide agents to automate tasks.

Can you help me understand why you're uniquely positioned versus other vendors in the market? And what that means going forward as you continue to invest to grow your competitive advantages?

Yeah, so we made that comment in the context of what we're doing in automating collaboration and supply chain transactions between trading partners. And what we found that is really key to that is The data that we have on the network. So, you know, three main components there. One, of course, the customer's data on the network. Often we have more of their supply chain data in our network than they have available to them in the ERP. It's just a broader set of data. We also see all the kind of macro transaction patterns. are all going across our network. So of course we can't let one customer look at another customer's discrete data, but what we can do is look at trading patterns, especially across the major retailers. So we may see some differences in the way that Walmart or Target are handling some of their suppliers and see that at the macro level and translate that into some changes that the suppliers to those retailers need to make. And then maybe most importantly, is, you know, over this 25 years of doing this, we have built out very deep proprietary databases of supply chain expectations that the major retailers and distributors in the US have around compliance and supply chain expectations. And a lot of this information we have is stuff that's not going to be available in a downloadable vendor guide that they're going to provide. And then a lot of them on the network don't even provide these types of vendor guides. and so we're really able to train the agents on this proprietary database and those agents are really able to guide these suppliers to execute their supply chain in a way that's going to be compliant with their retail and distributor customers.

Got it. That's helpful. Thank you.

And I'm showing no further questions in the queue. Ladies and gentlemen, this concludes today's conference call. Thank you for your participation and have a wonderful day. You may now disconnect your lines at this time.