Konstantin Bredyuk and the decision before the bill

An invoice is a late arrival. By the time it reaches the accounts team, someone may already have chosen the supplier, agreed the price and promised that payment will follow. Konstantin Bredyuk has made a business out of paying attention to that earlier moment. The interesting event, in his telling, is the permission that set everything moving.

In August 2026, he put the idea plainly: “It leaves at the commitment”. He was talking about money, and about the purchase orders and agreements that commit it before finance catches up. For the co-founder and CEO of ApprovalMax, the distance between a decision and its financial consequences has become a professional preoccupation. An inbox is a remarkably casual place to keep that much responsibility.

His route to this subject runs through consulting, product management, a shared software patent and a university classroom. Along the way, he has kept returning to the machinery of working together: how people pass tasks, share information and decide who has permission to do what. The drama is modest. The consequences can be expensive.

A familiar inbox, a different problem

Bredyuk studied software engineering at the Higher School of Economics, completing a master’s degree between 2008 and 2010. His early career included consulting at Esri CIS, followed by product work at NetCracker and Comindware. Those roles placed him in the business of making software useful inside organisations, with all the complexity that organisations supply free of charge.

One concrete trace survives in a US patent granted in May 2014. Bredyuk is one of four named inventors on a system for synchronising custom objects between Microsoft Outlook and an external server. The assignee was Comindware. The subject sounds forbiddingly technical until you consider the everyday problem underneath it: business processes living beyond the email application where people already spend their time.

The patent describes bringing workflow-backed objects into that familiar environment and keeping their information synchronised. It is an early example of the territory he would continue to occupy. There are people doing their jobs, there is a system meant to coordinate them, and there is a gap between the two that somebody has to design around.

A former Comindware colleague, Nikita Morozov, described Bredyuk’s involvement across strategy, product design and conversations with prospective and existing customers. The recollection gives some texture to the job titles: his work crossed the boundary between deciding what the company should pursue and understanding what its software needed to do.

The syllabus behind the software

By 2015, Bredyuk had put his name to a course specification for Software Business Management at HSE. Its premise was simple enough to be overlooked: knowing how to engineer software does not automatically equip someone to manage the business around it. The curriculum covered product management, marketing, sales, operations, finance, people and strategy.

The details are revealing. Students were expected to encounter product pricing, licensing, roadmaps and psychology in product design. A business model had to sit beside a working product. A feature needed a reason to exist. It is a less romantic account of software entrepreneurship than the familiar flash of inspiration, but considerably more useful on Monday morning.

Teaching also remains part of his public professional record. HSE’s 2025/2026 catalogue lists him as the instructor for the English-language course. Its learning outcomes include defining customer needs, prioritising features, analysing company strategies and understanding software-business finances. These are the same questions a product executive eventually finds waiting on the CEO’s desk.

The feature that customers kept asking for

There is a small, unusually concrete episode from October 2017. At XDHax, Bredyuk worked with Peter Volynsky and Mikhail Belov on an ApprovalMax budget-checking prototype. The project began with repeated customer requests. Approvers wanted budget information available while reviewing a bill or purchase order.

The team built a way to put that context inside the approval request. Their account includes an unglamorous complication: obtaining the budget detail they needed meant parsing CSV files. Bredyuk’s listed contribution was product management. The episode catches him in the middle of practical product work, turning a recurring question into something a customer could try.

The point survives the old prototype. A decision is easier to make when the necessary information travels with it. Otherwise, an approval tool can merely give the same old chase a tidier interface.

Enough structure to grow

When Bredyuk explained ApprovalMax’s origins, he described a team with a background in business process management and optimisation software. Those systems could automate many kinds of work, but enterprise implementations demanded configuration, time and effort. Small businesses had a different tolerance for the project required to get the product running.

ApprovalMax concentrated that experience on finance workflows. In an early interview, Bredyuk discussed conditional approvals, different authorisation levels, exceptions and the matching of bills to purchase orders. The ambition was specific: make the route from purchase to approval something a business could configure and follow.

His later product interviews returned to a balancing act. Customers might begin with bill approval, move into purchase orders, then want more extensive controls. Software had to accommodate those stages without loading the smallest organisation with expense and complexity it did not yet need. He also defended customers’ freedom to choose the other applications in their accounting setup.

That leaves the product manager with an awkward brief: be ready for the next stage without making the present one intolerable. Bredyuk’s answers repeatedly focus on change over time. A customer relationship can outlast the particular problem that first brought someone through the door.

When the email chain runs out of room

In January 2021, Bredyuk wrote about Data Friendly Space, a US nonprofit providing data applications and analytical services to humanitarian organisations. Its contractors submitted monthly invoices. As the organisation and its projects grew, an email-based approval process became cumbersome and a more systematic approach was needed.

It is an instructive example because the trigger was growth in ordinary work. More projects meant more decisions to coordinate. The original arrangement had acquired responsibilities it was poorly equipped to carry. Anyone who has searched an email chain for a definitive answer will recognise the administrative genre: a mystery in which all the suspects have replied “see below”.

Bredyuk’s professional writing makes room for these operational details. They explain why a business might buy approval software in the first place. A neatly recorded decision has value when the alternative involves reconstructing it after the people involved have moved on to something else.

A larger company around the product

ApprovalMax’s founding team includes Bredyuk, Helmut Heptner and Peter Volynsky. In 2024, the company publicly welcomed Bredyuk as CEO. He also announced senior additions in operations and people leadership that year. The organisation around the software was developing alongside its capabilities.

June brought ApprovalMax Capture and ApprovalMax Pay. In November, the company announced £10 million in primary growth investment from Yttrium, after operating on a self-funded basis. Yttrium also acquired an undisclosed secondary stake from existing shareholders. The announcement put the customer count above 16,000 businesses worldwide.

For Bredyuk, the funding supported a broader product ambition. He described investing further in research and development and expanding the accounts payable and receivable offering. Engineering, product and commercial hiring formed part of the next phase. The brief was growing from handling approvals toward more of the work surrounding them.

The ladder, and the bill for climbing it

In his writing about moving from Xero or QuickBooks Online toward NetSuite, Bredyuk uses a ladder as his organising image. A growing business can add connected applications to its existing accounting system before undertaking a larger migration. Capabilities arrive in steps, while the core ledger remains familiar.

He illustrates the argument with an imagined café expanding into several locations and a coffee-bean business. The example is a teaching device, and a revealing one. He follows the operational complications of success rather than treating growth as a number that simply goes up. More locations bring more information to consolidate and more people making financial decisions.

By June 2026, he was applying another question of fit to ApprovalMax itself: how customers should pay. He explained the move toward pricing based on document usage, with migration for existing customers beginning at subscription renewals from August. The change also brought Capture into the plans. His argument linked the charge to activity and the need to keep investing in the platform.

“The price you pay for software should reflect how you use it.”Konstantin Bredyuk, June 2026

The permission worth keeping

In August 2026, ApprovalMax announced its UK Small Business App of the Year win in the Xero Global App Awards. The accompanying figures described more than 18,000 Xero organisations across over 100 countries and more than 37 million approval decisions in the preceding year. Those are company-reported measures, and the Xero count describes a particular part of its customer base.

The scale gives weight to Bredyuk’s original subject without changing its size. Each approval still belongs to a particular purchase, person and set of permissions. His stated ambition is to give finance teams control from the moment spending is committed. After years of product work, that remains a recognisable task: get the right information in front of someone before they say yes.