Propel 26: The 5 Ideas That Redefined Professional Services in 2026

Something shifted at Propel26. It wasn't a product announcement, and it wasn't a trend report. It was a room full of PS leaders agreeing out loud that the way they've been measuring success needs to change. 

If you weren't there, this is your fastest way in. 

500+ leaders, 35+ powerful sessions on the current and the future of PS delivery, all in one venue for two days with interesting conversations both on and off stage (those you've got to be there for; we can't cover them in a blog, unfortunately).

The sessions didn't describe what the industry is doing. They challenged us to think about what should be done and how. Every speaker brought a different angle, but the same questions kept surfacing: What does success actually look like once the software is live? 

What does it take to earn a client's trust rather than just their signature on a project plan? And if AI is taking on more of the execution work, what does that free up a PS professional to actually become?

We've got you covered from the stage through this recap, with individual session summaries linked throughout and recordings of the sessions hosted on Rocketlane TV, where you can watch them all.

This blog is a quick recap of everything Propel26 was about: every session organized across five themes, so you can navigate to the ones that matter most to your team and go deep from there.

What is Propel26?

Propel26 is Rocketlane's annual flagship conference for professional services leaders, held in May 2026 this year. 

Over the two days, 500+ implementation, delivery, and customer success leaders gathered across 35+ sessions covering AI adoption, delivery operations, forward-deployed engineering, outcome-based pricing, and the shift from implementation-as-delivery to implementation-as-growth.

The overarching theme for the conference was the Outcome Era, a structural shift in how PS success gets measured, from go-live milestones to customer transformation outcomes. SPI Research 2026 puts the urgency of that shift in numbers: average billable utilization at a historic low of 66.4%, with 26.2% of projects running late.

Against that backdrop, the sessions at Propel26 didn't describe the problem. They built the case for what to do about it.

Rocketlane, the agentic-AI-powered PSA platform connecting delivery execution, AI automation, and customer outcome tracking in one system of record, hosted the event and used it to demonstrate what purpose-built AI looks like inside professional services delivery, not as a roadmap item, but as live infrastructure.

The 5 ideas that defined Propel26

This recap covers all 35 sessions organized by theme. Each section links to the full recap for that individual. Jump to the theme that matters most to your team, or read straight through for the full picture.

Idea #1: The finish line moved. Most PS teams haven't.

The most important idea at Propel26 didn't come from a product demo or a data slide. It came from a question that Srikrishnan Ganesan posed in the opening keynote: when is an implementation actually done?

For most PS teams, the answer is go-live. For most customers, the answer is a lot later, and it has nothing to do with whether the software is running. Go-live is a checkpoint. What comes after it- adoption, retention, and the customer's ability to run independently without leaning on the implementation team is where the real outcome lives.

That gap between how PS teams measure success and how customers experience it was the thread that ran through every session in this theme. 

Every speaker arrived at the same place from a different direction: the moment PS teams stop celebrating go-live and start measuring what actually changed for the customer, everything downstream improves.

Redefining the finish line

The Outcome Era: how professional services teams are shifting from delivery to growth

Most PS teams have a go-live metric. Almost none of them have an outcome metric, and that's a revenue problem disguised as a measurement problem. 

Srikrishnan Ganesan's opening keynote made the case that AI is reshaping what humans do in PS delivery, and the teams that will lead through that shift aren't the ones that automate faster. They're the ones who design implementations around what the customer's business should look like on the other side of the project.

The delivery-to-transformation shift: can your implementation team prove it matters?

Every PS leader knows adoption matters. Far fewer have wired their delivery process to actually measure it. 

Kevin Stanley from MaintainX made the case at Propel26 that delivery metrics are foundational, but they're not sufficient on their own. The teams creating the strongest growth outcomes treat adoption as an implementation metric, not something CS picks up after handoff. By the time the customer has a CSM, the adoption trajectory is already set.

Why net revenue retention is the most important metric in implementation

NRR belongs to CS on the org chart, but the implementation team influences it more than anyone in the business. 

Daniel Levine from Clutch made this case directly at Propel26: renewal decisions are shaped long before the renewal conversation begins. The habits customers build during onboarding determine whether they stay, expand, or churn. If the implementation team treats handoff as the finish line, they're handing CS a problem dressed up as a healthy account.

How to treat implementation as a product, not a service

Emily Garza from Unit21 cut implementation time from 90 days to under 60, and then discovered something that stopped the whole team: customers were reaching go-live without the configurations they needed to actually use the product. 

Go-live had become a checkbox, not a milestone worth celebrating. Treating implementation like a product means defining the experience, measuring adoption milestones, and not calling it complete until value is actually in play.

Why go-live is the wrong finish line — and what to measure instead

The real outcome in Chris Pinaire's customer data from Global Shop Solutions showed up at month nine, not month six, which means every team measuring at go-live was closing the book too early. 

The Milestone Value Map he introduced at Propel26 provides teams with five checkpoints tied to adoption signals, all of which are measurable before a project formally closes. If your measurement framework ends at go-live, you're evaluating the project before the story has actually played out.

Expansion is a delivery outcome, not a CS outcome

The expansion conversation usually takes place during a CSM quarterly review. The sessions in this group made a harder point: the expansion decision is usually made during implementation, months before that conversation ever takes place.

Expansion is decided before go-live: Graphite Connect's approach to implementation-led growth

Kasey Smith from Graphite Connect pulled her customer dataset and found a pattern so consistent it was hard to argue with: customers who had a positive implementation experience almost always expanded early, and customers who didn't, almost never did. 

The moat in an AI world isn't software features. It's human expertise built through years of embedded delivery. When a customer's internal team starts asking why they can't build this themselves, the implementation team is the last line of defense, and the answer they give depends entirely on how the relationship was built during implementation.

Why customer outcomes get lost between departments — and how to fix it

Sales promises a $15 million efficiency gain. Nobody validates the baseline. Nobody carries that number into delivery. By the time CS has the account, the value thread is completely gone. Jeremy Taylor and Robert Haukenberry from The Apricity Group showed at Propel26 that this isn't a communication problem — it's a structural one. A connected PSA doesn't just manage tasks. It keeps the outcome thread alive across every handoff so the customer never has to wonder what they're paying for.

How to set customer goals during onboarding — and why waiting hurts retention

Customers form their opinion of an implementation before they ever meet the CS team, which means the window to set the right expectations is narrower than most teams realize. Kristen Hayer from The Success League drew a sharp line at Propel26 between outcomes and milestones: "Reduce manual reporting effort by 50% within 90 days" is an outcome. "Complete system integration by week four" is a milestone. Success planning belongs at kickoff, not at CSM handoff, because by the time it gets there, the customer has already made up their mind.

The alignment problem: when every department hits its KPI but the customer doesn't

Idea 1 ends with a structural observation that several speakers circled from different directions: it's entirely possible for every internal team to succeed by its own metrics while the customer still doesn't feel value. That's not a performance problem. It's an alignment problem, and AI is making it more expensive to ignore. 

The PS outcome alignment audit: a framework for the AI era

Mariah Bayne from Authentically You Coaching brought five questions to Propel26 that every PS team should be able to answer about their work: What is the work stream? What is the team actually doing? Who does it serve? Where is AI touching this workflow? And what should be better because of this? 

The last question is the one most teams stumble on, not because it's harder than the others, but because the answer exposes whether the first four are pointed in the right direction at all. When a team can't articulate what should be better, AI doesn't solve the problem. It accelerates it, helping everyone move faster in directions that were never fully aligned to begin with.

How to drive organizational change in a professional services team

Every team that has tried to drive alignment through better reporting alone has learned the same lesson the hard way. Kristen Ware from Submittable and Whitney Eskenazi from LogicGate shared the cross-functional committee model at Propel26: change built with the people it affects tends to stick, and change handed down to them usually doesn't. 

Accountability, not documentation, is the linchpin. The teams that had internalized this were the ones who walked out of the session nodding.

Idea #2: AI is moving from copilot to participant

If Idea 1 was about redefining what PS teams are trying to achieve, Idea 2 was about redefining who is doing the work.

The first wave of AI in professional services was genuinely useful: meeting summaries, status updates, project risk flags, document drafting. It saved real time on tasks that were tedious but necessary. 

The speakers in this theme came from PS teams who have been living with that first wave long enough to see what comes next, and what comes next is more significant. AI isn't just assisting with the work anymore. It's beginning to execute it. 

Not summarizing a meeting after the fact, but attending the stakeholder call beforehand, gathering context, and arriving at kickoff with a head start. Not flagging a risk after it surfaces, but detecting the pattern days earlier and routing it before it becomes a conversation nobody wants to have.

From assistant to active participant

Agentic PSA is here: what Rocketlane announced at Propel26

The central question Rocketlane put to the room at Propel26 wasn't how AI can help teams work faster. It was what happens when AI has first-party access to your entire delivery system. 

When AI is embedded in the system of record, with real-time visibility into project plans, resource allocations, financials, timelines, and customer context, it can do more than surface insights. It can take action. Nitro AI's evolution from assistant to active delivery participant was the practical demonstration of what an agentic PSA actually looks like in production, not on a roadmap slide.

From MCP to multi-agent: the AI concepts every PS leader needs right now

Twenty-nine project templates configured in two hours, a task that previously took two to three weeks. That was the result Srikrishnan Ganesan showed in his closing keynote at Propel26, and it came from the same infrastructure every PS team now has access to. 

The session covered ten AI concepts, from MCP as connective tissue across systems, to Skills as reusable AI workflows, to voice agents conducting stakeholder interviews before kickoff. The gap between "AI as interesting" and "AI as infrastructure" is closing faster than most PS teams realize, which is why the challenge at the end of the keynote landed the way it did: go build something this weekend.

What AI actually looks like in delivery right now

Theory is useful. What the sessions in this group offered was something more grounding: actual results from PS teams who have moved past the pilot stage and are running AI across live engagements.

Practical ways to turn AI into your implementation copilot

Before Shehryar Malik from Tipalti changed anything about how his team worked, he asked a question that most AI adoption frameworks skip entirely: where are implementation teams spending time that doesn't directly help customers? 

Starting from that question, rather than from the technology, led to an 80% reduction in escalations and a 25% improvement in time-to-go-live. AI creates value by removing operational friction, and the friction that matters most is the friction that sits between the team and the customer, not the friction between team members.

Same team, 3x projects: how SaasGenie built a delivery system that scales

"Rocketlane alone will not bring us anything. With AI, we eliminate the busy work. Together, we are able to scale." That was Raj Rajasekar's summary at Propel26, and the numbers behind it were hard to argue with: from 60% to 90% on-time delivery with the same 35-person team, with $60K invested to unlock $1M to $1.4M in attributed value. 

The sequencing was as important as the outcome. Common templates first, standardized processes second, Rocketlane as system of record third, and then AI introduced one workflow at a time.

How AI is fixing professional services' oldest problem: bad scoping

Twenty-seven years in PS, and Joey Poarch from PSQuote's diagnosis was unchanged: scoping runs on tribal knowledge and spreadsheets, and the consequences show up months later as scope surprises on delivery. 

AI changes the equation with probabilistic estimates that carry confidence intervals, discovery transcripts converted to structured estimates in minutes, and continuous learning from the gap between as-sold and as-delivered actuals. The session results he shared at Propel26: 3x faster quote turnaround and 50% fewer scope surprises.

From feature factory to solving business outcomes

Ali Meyer from GlossaPro.ai traced every bad project her team had ever run back to the same place: a broken handoff between sales and delivery. The fix she shared at Propel26 was structural. Outcomes need a baseline, a target value, a financial impact, and an owner, all of which must exist before anyone writes a SOW. 

Once requirements are structured and mapped to those outcomes, they plug directly into Rocketlane as tasks, and the project has a spine from day one instead of discovering it's missing one at month three.

Why AI fails without an operational foundation

Every AI session at Propel26 eventually converged on the same warning: AI amplifies the system you give it. If the process is unclear, AI runs it faster in the wrong direction. If the data is fragmented, AI makes confident recommendations on unreliable inputs. The sessions in this group made that principle concrete enough to act on.

Design before you automate your PS implementation

Merlin Komenda from Zappi cut time to first value from 30 days to 15, and the result didn't come from automation. It came from redesigning the implementation before automating anything. Three principles shaped the work at Propel26: start with customer outcomes, lead with customer expertise, and let automation follow clarity. 

If you can't explain what a step is for, redesign it before you automate it. Every hour spent fixing a messy manual process before AI touches it saves multiples of that in troubleshooting downstream.

Why AI implementations plateau — and how to break it

Anuj Arora from Ada, in his session, spoke about how the plateau is created during implementation, not after it. If teams reach handoff without establishing momentum across people, process, and technology, CS rarely has enough leverage to create it afterward. 

His AI maturity scorecard mapped customers across four levels, and the outcome gap between them was stark: customers at Level 4 generated 150%+ NDR, while customers stuck at Level 1 fell below 90% and churned. The conditions for success need to be built before go-live, because the window to create them closes at handoff.

What aviation teaches PS leaders about AI readiness

In aviation, no single failure brings a plane down. It's always a chain of smaller failures that align at exactly the wrong moment. Brian Hodges from nCloud Integrators is a licensed pilot, and he brought that principle directly to PS delivery at Propel26. A mispriced project is recoverable. 

A sponsor leaving mid-engagement is recoverable. A bad handoff is recoverable. When all three happen on the same project, it usually isn't. His practical takeaway for AI: set personal minimums, a clear definition of what requires human review before anything reaches a client, and only raise those standards as a tool earns trust through results you can actually verify.

Three decisions every PS leader needs to make before 2027

Jeff Rosenbaugh from Lucid, Leanne Snoeck from Quickbase, and CJ Tully from Smartsheet came to Propel26 with a shared conviction: there are three decisions PS leaders can no longer put off. First, choose your model. AI efficiency that reduces billable hours in a T&M setup is optimizing for the wrong thing, and that tension only grows. 

Second, consolidate your knowledge before layering AI on top of it, because organizations with structured process documentation reach AI adoption milestones two to three times faster than those without. Third, build a formal AI innovation program so that when someone on the team figures something out, it becomes a capability the whole organization keeps, not a trick that walks out the door with them.

The CFO conversation: proving AI ROI to finance

Only one in eight CFOs has a clear formula for AI ROI. The PS teams that secure budgets aren't the ones with the most enthusiasm for the technology; they're the ones who arrive at the finance conversation with evidence, connected to the metrics finance already cares about.

How CFOs are rethinking AI budgets — and what PS teams need to know

Aly Khan Musani from Detechtion, Intekhab Nazeer from Lineaje, and Chithra Rajagopalan from Obsidian Security described a structural shift at Propel26 that changed the conversation for every PS leader in the room: AI spend moved from centralized IT budgets to department-level accountability in 2026. 

Token budgets, consumption alerts, and six-month planning cycles are the new governance model. The strongest PS business cases connect investment directly to gross margin, NRR, or expansion revenue, not to hours saved, because hours saved is a metric that finance has learned to discount.

How to prove AI ROI in professional services — when productivity gains aren't enough

Productivity is a leading indicator, not a final outcome, and CFOs in 2026 know the difference. Ana Gole from Verint, Justin Collins from Proofpoint, and Josh Rutberg from Glean recommended at Propel26 that PS teams track adoption, retention, customer value, and delivery performance alongside efficiency gains. 

The other warning from the panel: AI fragmentation, meaning deploying disconnected tools across the delivery workflow, creates exactly the kind of complexity that makes ROI harder to prove and organizational trust in AI harder to build.

Idea #3: PS teams are becoming growth functions

Professional services teams were built to deliver. The best ones in 2026 are being measured on something more.

The shift that several speakers articulated across this theme is worth sitting with: implementation teams don't just influence whether a customer succeeds. They influence whether that customer expands, renews, and becomes an advocate. 

When those outcomes get connected back to delivery decisions made at kickoff, the PS team stops being a cost center and starts being a revenue function. That's not a positioning exercise. It's a structural change in what the team is accountable for.

When implementation teams own the growth outcome

Proactive professional services: why the best teams have stopped taking orders

The most important skill in professional services isn't configuration. It's diagnosis. Gwen Thorn and Kristen Rosenberry from Moveworks made the case at Propel26 that the best consultants don't do what customers ask; they help customers achieve what they actually need, which is often a different thing entirely. 

As AI takes on more execution work, the ability to identify the real problem underneath the stated request becomes the primary differentiator between implementation teams that clients want to keep working with and those that get commoditized.

Beyond the bucket: how dbt Labs shifted from hours to value

Zero change orders after project kickoff. That was one of the results Erin Vaughan and Jess Williams from dbt Labs shared at Propel26 after introducing the Resident Architect Access model: instead of purchasing hours, customers purchase access to an experienced dbt architect. 

The other results were equally striking: a 9% increase in billable utilization and 52% year-over-year growth in attach rates. When the engagement model aligns with customer outcomes rather than team activity, the economics of delivery change.

Scale without restructuring first

The instinct when a PS team is under capacity pressure is to restructure. The sessions in this group made a different case: before redesigning the team, get visibility into how the work is actually happening. Structure should follow signal, not the other way around.

How to scale professional services without restructuring: the visibility-first model

Kate Brady and Eric Olson from OneSource Virtual grew more than 20% year-over-year while managing concentrated delivery spikes, and they did it by bringing six teams onto one platform before making a single restructuring decision. 

The insight they shared at Propel26 was straightforward but easy to skip: shared visibility creates evidence, and evidence creates better organizational decisions. Teams that restructure before they can see what's actually happening tend to optimize for the wrong bottlenecks.

How to automate project hygiene — replacing gut feel with real-time risk scoring

Grace VanderMolen, Russell Pope, and TJ Clark from OneSource Virtual inherited a portfolio of 150 projects that all showed green health statuses, with broken dependencies sitting underneath every one of them. Operation Stabilize took five weeks of manually applying a hygiene framework before a single check was automated. 

The principle they brought to Propel26 holds across every AI adoption context: prove the framework works by hand before automating it, because automation locks in whatever you give it, including the chaos.

Partner delivery at scale: why trust alone isn't enough

Partner programs don't create weaknesses in the business. They expose weaknesses that are already there. Justin Manduke from Optro, Josh Simpson from Gainsight, and Sabina Pons from Growth Molecules made this case at Propel26: mature partner ecosystems are built on operational consistency, meaning the ability to deliver the same customer experience regardless of which partner does the work, not just on contracts and good intentions. The teams that had made it work were the ones who had standardized delivery before they had scaled it.

Designing delivery systems that make good outcomes automatic

The strongest PS teams at Propel26 weren't relying on effort and individual excellence to produce consistent outcomes. They had built systems where the right outcome was the path of least resistance, which is a fundamentally different design philosophy.

Make good delivery outcomes the default, not the exception

Forty hours of work compressed to 20 minutes. That was one output of what Hannah Kinder from Greener by Default called operational ecology: building delivery systems where the right outcomes are natural, not dependent on heroics, reminders, or perfect individual compliance. 

The 72% reduction in client-facing staff effort she reported at Propel26 wasn't a technology win. It was a design win. Burnout in PS teams is often a systems failure, and margin pressure is often a workflow design problem dressed up as a resourcing problem.

Why your clients don't need to understand complexity — they need to see through it

April Slinger from HealthEZ manages close to a million possible plan combinations per client implementation, moved access-to-care rates from 43% to 90% in two years, and brought broker escalations to the CEO down to zero during the busy season. 

The time zone analogy she used at Propel26 clarified the design principle behind all of it: complexity doesn't disappear; it gets contained in a shared framework so clients only need to know where they are and what they need to do next, not how the system behind them works.

Idea #4: Trust and transparency beat polished reporting

Every PS team has a status update process. The teams at Propel26 who had the strongest client relationships had something different: a culture of shared accountability that made the status update almost irrelevant.

The sessions in this theme weren't about communication tactics or stakeholder management frameworks. They were about a more fundamental shift in how PS teams position themselves relative to their clients, as partners who share the problem rather than vendors who report on progress toward a deliverable. The difference in client relationships is visible, and so is the difference in outcomes.

What radical client transparency actually looks like

Why your best clients are still keeping secrets from you

At 9:47 pm, during a steering committee call, a major deliverable slipped, disclosed at the last possible moment because the client didn't feel safe enough to surface it earlier. Jacqui Morgan from Xceptor brought this story to Propel26 not as a cautionary tale, but as the starting point for a practice that eliminated it entirely. 

Open the Rocketlane portal live on every client call. Update everything together before anyone hangs up. Make shared visibility a ritual rather than a report. The renewal decision from State Street, she noted, was based not on the software delivered but on how it was delivered. Informed clients are manageable. Invested clients become partners.

Why clients are asking for outcome-based pricing — and what they're actually afraid of

Clients aren't asking for outcome-based pricing because they're trying to squeeze margin. Sarah Hurd from Faye made this case directly at Propel26: they're asking because they've been burned before. Every redlined SOW carries the scar tissue from a previous engagement that went sideways. 

The fix isn't a pricing innovation. It's trust repair: capture the baseline at kickoff, prove value throughout the engagement, and give clients a genuine opt-out if the trajectory isn't where it needs to be.

Authority over execution

Transparency builds the foundation of client trust. But the sessions in this group addressed the second element, one that's harder to train for: authority. The ability to lead a client conversation with genuine confidence rather than just report on it.

Why authority is the real differentiator in PS delivery

Technically sharp. Goes quiet when a CFO asks a pointed question. Stacey Milgram Potzka and Amanda Stewart from Actabl described this pattern at Propel26 because they had lived it, and then built the Customer Excellence framework to address it. 

The results: customer satisfaction consistently above 85% against an industry standard of 70%, more than 30% of completed projects generating unsolicited C-suite appreciation, and escalations dropping from out-of-control to near zero. Authority is a skill. It can be taught, and it compounds.

Idea #5: Operational maturity is the new competitive moat

There's a version of the AI adoption conversation that focuses entirely on tools. The sessions in this theme came from speakers who had moved past that version and were thinking about the more durable question: what organizational foundation makes AI trustworthy over time, and what does it take to build that foundation before the next wave arrives?

The organizations that will lead PS in the next three years aren't necessarily the ones deploying the most AI. They're the ones who built the foundation that makes AI reliable in the first place, and who have the discipline to keep building it as the technology moves faster.

The infrastructure AI needs to work

What Postman's co-founder learned about building for the AI-first era

APIs are not abstractions for engineers anymore. Ankit Sobti, co-founder and CTO of Postman, reframed this for the room at Propel26: APIs are the hands and legs of AI, the execution layer that turns agent reasoning into action. 

Three-layer readiness is required for AI to work at scale: the agent layer, the data and knowledge layer, and the tool and API layer. The warning that most people carried out of the session: good patterns compound, but so do bad ones. Build observability into your AI system before you scale anything, because the mistakes that are invisible at small scale become very expensive at large scale.

Forward-deployed engineering: a new model for technical delivery

Forward-deployed engineering sits at the edge of what most PS teams have tried before, which is exactly why the Propel26 sessions on it were so well attended. The question isn't whether FDE works in theory. It's whether the organization has the delivery discipline to make it work in practice.

Forward deployed engineering: finding the model that fits

One north star appeared across every FDE model represented on the Propel26 panel: the person who hears the problem should be close enough to build the solution. Beyond that, the models diverged significantly. Alex Laverty from Harvey positions FDEs as lawyers rather than engineers, with domain expertise as the bridge between AI capability and client trust. Miku Jha from ServiceNow requires every FDE engagement to produce reusable assets, not just customer-specific solutions. 

Rajkumar Irudayaraj from Alteryx runs time-boxed commercial acceleration and then hands off. The panel, moderated by Srikrishnan Ganesan, made clear that the right FDE model depends heavily on what your product and your customer need, but that the absence of delivery discipline behind any of them turns FDE into expensive custom work with no measurable return.

The FDE hiring decision: when you need one and when you don't

Three hats, one person: consultant, PM, software engineer. Kevin Bai from Rippling described the FDE role at Propel26 with a clarity that cut through most of the category confusion. The two-axis test for whether it makes sense: a technically complex product combined with a buyer who genuinely can't unlock value without a technical partner. 

Without tracked milestones, visible outcomes, and a connected system of record behind the role, FDE becomes the most expensive form of customer service a PS team can offer.

Beyond the sessions: What Propel26 felt like in person

Beyond the sessions, Propel26 had its own kind of energy. 

The breaks were just as lively as the rooms, with attendee Lego kits, customer booths, a very popular puppy booth, and plenty of conversations happening in every corner. 

The Nitro Cafe quickly became the crowd favorite, serving coffee and ice cream in exchange for Nitro tokens that attendees could earn through LinkedIn posts, quizzes, and demos. It was fun, interactive, and very on-theme. 

What Rocketlane demonstrated at Propel26

Rocketlane used the conference to show what an agentic PSA looks like in practice, not as a roadmap slide, but as live delivery infrastructure running across real customer engagements.

Configuration MCP tools: 150 tools covering setup and administration tasks that previously required weeks of manual work. Live results from the conference demo: 29 project templates configured in two hours; 50 template field updates in five minutes; 7 regional holiday calendars in ten minutes.

Customer builds on Rocketlane: OSV built Flight Path — an automated project hygiene scoring app — directly on Rocketlane's SDK and API. Real-time risk scoring replaced gut-feel escalation across a 150-project portfolio. Go-live criteria were tied to actual operational events: first payroll processed, collections cleared, disbursements run.

Nitro AI as delivery participant: Nitro has first-party access to project plans, resources, financials, timelines, and customer context. That access is what allows it to move beyond helping teams work faster and toward helping teams execute more of the work itself.

Skills: Every repeatable PS process captured as a reusable AI workflow. One person's best practice automatically becomes the team's default behavior.

More reads on this:

Agentic PSA is here: what Rocketlane announced at Propel26 ·

 From MCP to multi-agent: the AI concepts every PS leader needs right now

The conference ends. The shift doesn't.

AI is here, and it's not here to steal anyone's job. It's here to make everyone better at what they do. Srikrishnan Ganesan closed Propel26 with a direct challenge: use it to write code, build an agent, run a complex analysis, or build a presentation the way he built his. Not because you're out of time, but because now you have more time to do what AI can't do.

The infrastructure that felt experimental a year ago is now accessible to every PS team. MCP, voice agents, computer-use systems, multi-agent orchestration, Skills: none of these are research projects anymore. They are available right now. The leaders who develop fluency with them today will be the ones defining what PS delivery looks like in 2027.

The question Propel26 left on the table isn't whether AI will change how professional services teams operate. It will. The question is whether you'll be leading that change or catching up to it.

Explore every Propel26 session: Rocketlane Blog

Watch the sessions in full: Rocketlane TV

See how Rocketlane's agentic PSA works: Book a demo

Based on live session data from Propel26 (May 2026) and aggregate outcomes from 750+ Rocketlane