We’ve all heard the rumors. The implementation era is ending for technology services firms. While that assertion may be a bit exaggerated, there is an element of truth to it.
The services opportunity has been moving beyond basic platform configuration and ‘lift and shift’ deployments for a while now. Low-code configuration, and increasingly intelligent SaaS platforms are making parts of deployment easier and more repeatable. AI coding tools and AI-automated deployment platforms are exacerbating this.
Yet enterprises aren’t suddenly going to stop deploying and integrating new platforms. What IS changing is how much differentiation a firm can build just by focusing on implementations alone. Even if you’re a platform expert, knowing how to get a platform live no longer commands a premium.
The more durable services opportunity has always been understanding how a platform fits into the business and how to make it work with everything around it. The proliferation of applications, data, and now agents, just adds more dependencies and decisions that no individual deployment can resolve. The great irony is that enterprise architecture is becoming more challenging, even as traditional software implementation gets easier.
The great irony is that enterprise architecture is becoming more challenging, even as traditional software implementation gets easier.
It’s why ‘orchestration’ is this year’s biggest buzzword. There are lots of different definitions of orchestration, but at its core, it’s the ability to coordinate how data, systems, agents and people work together to improve a business process. Implementation remains part of this work, but the bigger need is making the whole process and environment operate seamlessly, especially if you want to put AI to work across processes.
All of this means more complex (and lucrative) work for services firms, especially those that can think beyond a single platform.
A massive orchestration opportunity
The average enterprise now manages 957 applications, yet only 27% are connected, according to MuleSoft’s 2026 Connectivity Benchmark Report. Systems have always had problems with silos, and IT teams already spend 36% of their time designing, building and testing custom integrations between systems and data. AI just adds another layer of fragmentation to that environment.
Enterprises now have to coordinate across more applications, data sources, models, agents and APIs than ever before. They must also protect against a growing number of security boundaries, evolving standards and shifting regulatory requirements. Not to mention trying to keep up with the AI vendor chaos.
This is not simply a bigger integration problem where we’re connecting System A with System B, C, and D. Today it’s about making an entire environment work together to accomplish an outcome, and then continuously optimizing how that happens.
That need is creating a significant new market for services. McKinsey estimates that agentic AI could generate nearly $200 billion in new enterprise spending on digital services such as orchestration, agent engineering, security and governance.
That work could include:
- Redesigning processes for human-agent collaboration
- Establishing the data architecture and enterprise context agents need
- Connecting systems, models, APIs, and specialized technologies
- Defining agent roles, permissions, and escalation paths
- Building governance, security, and auditability into workflows
- Testing and monitoring interactions across multiple systems
- Managing adoption and measuring outcomes across functions
The outcome (not the platform) is the new unit of work
Taken together, these capabilities expand the fundamental unit of services work. Historically, a transformation might have been framed as “implement Salesforce,” “move to Snowflake,” or “deploy ServiceNow.” But success is no longer defined by whether one platform goes live.
Today, an outcome such as AI-enabled quote-to-cash might span Salesforce, SAP, Snowflake, niche pricing and contract systems. Not to mention identity, payments, customer data, and multiple AI models. Consistently reliable outputs require more than system integration and data exchange. They require an understanding of how work flows across these components.
- Which system should an agent call?
- What data can it access?
- When is human approval required?
- What happens when something fails?
- How can the organization prove what occurred?
It’s one of the reasons why outcome-based pricing is so difficult. A partner might make quoting faster in Salesforce, but if pricing requires manual checks against outdated ERP data, or contracts sit in legal review, or invoices fail because customer records don’t match, the business may not collect cash any sooner. Delivering the outcome requires both visibility into and influence over the broader environment that produces an outcome.
The same applies to software delivery and customer support. AI coding tools may help developers write code faster, while testing, security reviews and release approvals continue to delay delivery. An AI support agent may handle routine questions, while recurring product defects or fulfillment problems keep generating new tickets. Local productivity gains will only translate into business value if they address what actually limits performance.
A partner doesn’t necessarily need to own every system or perform every task to move into outcome-based pricing. But it does need to understand the dependencies, work with the people who control them, and be explicit about what must change for the promised result to materialize. This is why outcome-based services are pushing firms toward orchestration. Their value increasingly depends on making the whole process work better.
Outcome-based services are pushing firms toward orchestration. Their value increasingly depends on making the whole process work better.
Valiantys is a great example of how a traditional implementation partner can move up the value chain. Valiantys built its reputation and business by being Atlassian’s largest, dedicated global partner. Over the last few years, they have built on their Atlassian expertise and understanding of core engineering systems to lean into AI-enabled service management and software delivery workflows. Their ability to go deep on the Atlassian system-of-record while orchestrating work and data across different systems has made them a go-to transformation partner for companies like McLaren, Siemens Energy, and AA.
Orchestration is the new black
In short, orchestration creates even more ongoing services opportunities. It gets firms into the higher-value domains of architecture, governance, exception handling, and operating-model design. They can also stay stickier after deployment. As models, business rules and systems change, firms can help to evaluate performance, monitor handoffs, and adjust workflows.
To learn more about other market shifts reshaping technology services, and to discover the software ecosystems offering the greatest opportunities for services firms, explore the latest edition of the Tercera 30.