Tonkean recently announced that it’s making its powerful agentic orchestration platform and autonomous enterprise AI agents generally available to Fortune 1,000 clients.
In a market newly awash in agentic AI offerings and hype, Tonkean stands apart for how it instruments AI agents for enterprise operations.
Tonkean Agentic Orchestration layers autonomous, collaborative AI with deterministic rules-based automation, which allows humans to set guardrails for what the agents can and cannot do, while at the same time empowering each agent to self-determine how to achieve the goals set for it.
Tonkean Agentic Orchestration facilitates true orchestration end-to-end, with interoperability across 200+ platform integrations. Tonkean Agentic Orchestration is 100% no-code, internal enterprise teams like procurement can build, deploy and orchestrate agents themselves.
Sagi Eliyahu, Co-founder and CEO, shared insight with Procurement Magazine on the impact this will have and why the hype this time is real.
Last year we saw generative AI become a popular means of creating content. It's equally good at reviewing content and understanding intention. When you combine these capabilities, you have powerful technology. However, like any technology, if it's not accessible to employees in day-to-day workflows, it remains just technology.
The real jump we'll see this year is treating LLMs not just as answer machines where you ask questions and get responses, but as entities you can assign work to. This is part of what makes agents exciting.
Tonkean's true differentiator is combining agents with an orchestration platform. The value of orchestration — sitting atop all your processes, tools and teams — in this case is that it allows you to make AI accessible to employees at strategic points in your processes. Sometimes the AI appears as a chat interface, but at other times as another entity working on tasks. You work and collaborate with agents as they become part of your process.
We're the leader in orchestration because we've been doing it for 10 years. Our capabilities extend far beyond procurement, with the ability to orchestrate downstream, upstream and throughout complex environments in the Fortune 500 landscape. For us, this amount of integration is just natural, but in the context of AI agents, it’s transformational.
P2P systems have become core systems for procurement. The problem is, they constitute a walled garden.
When you zoom out, very few employees outside of procurement want to work inside that environment. Employees who are buying things operate within a wide number of tools — Slack, Teams, Asana, etc. the same is true on the other side, you have reviewers and approvers from legal or finance or IT who need to engage with these processes and who work in their own tools.
The reality for large enterprises is they don't have just one P2P system. They have ERPs that might compete with the P2P they bought, plus through M&A they might have acquired five other ERPs and 20 CLMs. The real enterprise environment is far more complex than what the big P2P players might suggest.
Part of what makes orchestration so valuable for organisations is it allows procurement, in this case, to create and manage procurement processes that meet all these employees where they are — rather than requiring them to log into the big P2P system, which is central to the work procurement does, but which tends to be difficult for other teams and employees to use.
True orchestration means connecting all elements of your organisation so you can automate complex processes across them. True orchestration means being able to harmoniously and strategically manage all your organisation’s moving parts — its people, tools and data — much like musical components of an orchestra. You have to be able to leverage all the orchestra’s different instruments to make a song work as well as it can.
Orchestrating inside one system isn't true orchestration. Orchestration happens when you work outside the boundaries of individual systems.
That's why I'm confident in saying we're the best at orchestration — because no one else in the procurement space is doing true orchestration, in my opinion. They're doing automation, integration and some may even handle intake well. But that's not orchestration.
Orchestration is like working with 130 musicians playing different instruments and making them all work together to create harmony. Otherwise, it's just noise.
The impact will be immense, but it might not look the way people think. The concept of AI replacing jobs, for example, is nonsense. Every time there's a significant technological leap, it changes jobs, absolutely—but it also grows the pie and creates new needs and roles.
What AI does first is lower the barrier of entry for tackling problems. Think about different challenges in procurement: predicting changes in supply chain, handling various geographies and political landscapes—things that might become relevant or irrelevant within months.
Previously, you wouldn't apply technology to these challenges because the barrier to entry was too high — implementation was too complex or costly. But with the latest AI developments, accessibility of agents and orchestration allowing you to integrate these agents into processes, you're unlocking areas that were previously too expensive to handle.
This opens up new opportunities and strategic work that I can't even predict right now. What I do know is that priorities will change significantly. Procurement leaders entering 2025 with certain investment plans will likely have a surprisingly different conversation by year-end. I don't think people fully understand how much this technology will alter day-to-day operations—in a good way.
When companies think about unlocking the AI revolution, I believe it will be a combination of having the right tool stack and the right mindset. This isn't about looking for broken things and putting lipstick on them. It's about asking: what do my best people spend their time on and how can I replace the wasteful parts with AI that works with them?
This is a mindset shift that most people don't fully understand yet. They still think in terms of connecting systems or moving data. That's so last decade! We need to think about changing the paradigm.
Procurement and finance, but procurement specifically, have the opportunity to lead their organisations in this transformation. They occupy a unique position with domain knowledge that isn't easy to replicate, while simultaneously performing menial work that's easy to delegate. This combination means procurement can elevate itself relatively quickly.
It's similar to software development, where engineering examples are flooding the web with agents writing code. To write a programme that works well at scale requires expertise, but these experts waste time on mundane tasks. The ability to extend their impact by overseeing multiple agents that handle the routine work—while they focus on dictating outcomes, validating results and moving to the next challenge—is why people are so excited about the potential.
The first stage procurement had to go through was digitalising everything. That's happened for the most part and has become the baseline. Then you started seeing attention being paid to compliance and adoption: the data is there, the process is there, people are following the process. That's why we saw a surge in discussions about intake and adoption.
The next natural progression is efficiency: now that you have the data, the machinery and the people using it, how do you get more with less? After compliance and adoption, you think about how to gain leverage.
This might not fully materialise this year because market penetration of intake solutions isn't big enough. But in the next two to three years, I think we'll see a completely different approach to supplier diversification and category management.
I expect higher specialisation across different categories. Today, having a different SaaS product just for procuring steel seems ridiculous—it would be costly and hard to implement. So you use whatever you can from the big players, which lacks depth and throw people at the problem.
With agents and AI, within the next two to three years, you'll be able to get very specialised in specific industries and types of work, gaining more leverage than in the past. Domain-specific agents will come slowly at first, but then all at once.
We're still close with Coupa and also partner with SAP and other P2P systems. We're friends because we're not trying to be a P2P — we're an orchestration layer that sits on top.
In the last year, we've also partnered with the Big Four and many other management consulting firms. This aligns well with what I said earlier about domain expertise being where the rubber meets the road.
There's not much difference from what I've already mentioned. While people understand these concepts and they're part of the cultural discourse, market penetration is still low. Most companies are looking at these technologies but haven't fully implemented them or seen the ROI yet.
We still have intake solutions that have generated hype over the last two to three years, but most companies are still using carrot-and-stick approaches to drive compliance. Some companies still have digital transformation projects because their processes are paper-based. The question is what percentage of the market is in this state?
Intake will remain a significant part of our offering and focus. Orchestration is our core, but it's a never-ending journey because our mission is to bring the best technology and tools to every part of the business so people can do the work they actually want to do, unlocking human potential.
With AI and agents, there's real potential to leapfrog some of these problems. I'm interested to see how we can help lagging companies skip some early steps and move directly to agents.
The biggest value of LLMs and AI, in my opinion, is that they communicate like us. They understand things from a linguistic perspective. When we started the company, I felt technology and software weren't built for people — they were built for data. But business processes aren't about data; they're about people. That gap was why we started the company and it remains true nearly a decade later.
AI and large language models have the potential to leapfrog some of these challenges by allowing us to communicate naturally while orchestration and automation handle data structuring in the background. This is very exciting to me and will shape people's expectations of work in general—what they come to work to do. That's where we're headed.
Read the interview here.