The feature was finished by Sunday. Three developers had given up their weekend to build something a prospective customer wanted. Anthony Welgemoed was one of them. As he remembers it, their boss arrived in a Porsche with the roof down, called out his thanks, and drove away. The prospect never became a customer. The software worked; the bargain between boss and team looked rather less convincing.
Years later, Welgemoed still tells that story. The car supplies the comedy. The wasted effort supplies the lesson. He now runs Ziflow, a company concerned with how work moves from one person to another, and his account of that weekend ends with a practical rule: “Never waste my team’s time.” A brief acknowledgment, he suggests, would have mattered. Even paying for the developers’ KFC would have been a start.
It is a revealing place to begin a career that has involved two companies in the same field. Welgemoed has spent years helping creative and marketing teams get work reviewed, corrected, and approved. The subject sounds administrative until you consider the people waiting at each stage. A designer can finish a file and still have a considerable amount of work left before anyone agrees it is finished.
The problem survived the sale
Welgemoed came to this corner of software through technical work and product management. He completed an MBA at London Business School in 2007 and worked at Mtivity, which supplied marketing procurement software and services. Through that business he met Mat Atkinson. Atkinson was working on ProofHQ and needed a technical co-founder. Welgemoed became its CTO.
The product addressed a specific nuisance: gathering feedback on creative assets without losing the thread. A file could acquire several versions, several reviewers, and several opinions about which version deserved attention. Giving that exchange a shared place to happen was useful. Giving it a process made it useful to organizations with more complicated approval requirements.
On July 22, 2015, Workfront announced that it had acquired ProofHQ. The buyer and the acquired business already knew each other. ProofHQ had powered proofing inside Workfront for nearly three years, and the companies shared more than 350 customers. The deal brought a specialized review tool into a broader work-management offering. It was an extension of an existing relationship, with customers already using the two together.
A sale can settle ownership without exhausting a problem. Welgemoed and Atkinson began looking for their next project. They initially pursued automation for routine creative production tasks. Conversations with customers then pointed them back toward online proofing: there was demand for an independent product that could serve teams without requiring them to adopt a larger platform.
- 2015ProofHQ acquired
by Workfront - 2017Ziflow launches;
proofing pivot follows - 2023$20m Series A
announced
The dates help untangle the second venture. Welgemoed has described beginning development in 2016 and launching the initial product in January 2017. Later that year, the direction shifted toward standalone proofing. The useful discovery was quite precise: customers still wanted help with a job the founders understood. Their first idea supplied a starting point, and customer conversations changed what came next.
The advantage of a familiar partner
Atkinson appears throughout this story. He was the founder looking for a technical partner at ProofHQ, then the person with whom Welgemoed built Ziflow. In discussing their partnership, Welgemoed describes a division of work based on knowing what each enjoys and does well. They stayed close to agencies, marketing, and creative teams rather than treating the second company as an invitation to enter an unfamiliar industry.
“We play to our strengths and trust each other.”
Anthony Welgemoed, on building with Mat Atkinson
That is a modest description of an important advantage. A new business asks its founders to answer enough questions without also having to discover whether they can work together. Familiarity gave them a shared vocabulary for the customer and the product. It also gave Welgemoed someone whose judgment he had already seen under the ordinary pressures of running a company.
He is pleasingly unsentimental about another founding detail. Asked how Ziflow acquired its name, he explained that they searched for an available domain. There is no need to embroider the syllables with a grand theory. For a business devoted to getting things through a process, finding a name that could actually be registered seems appropriately practical.
He also recalls struggling earlier in life to find a business idea large enough to pursue. Building twice in the same space changed his advice. Start with a small, real problem and let feedback expose the surrounding difficulties. In that account, the opportunity becomes clearer through use. The customer who tries something can tell you where it fails, or where the next piece of work begins.
A spreadsheet, then a $20 million check
By July 2020, Companyon Ventures had published its reasons for investing in Ziflow. The case included prior experience with the ProofHQ team and a close interest in how the business operated. Its account singled out a shared performance spreadsheet tracking the journey from website visitor to lead, sale, upgrade, and departure. The spreadsheet made the company’s workings visible to its senior team.
There is a useful detail in that choice of evidence. A product demonstration shows what software does. An operating report shows whether a company understands how customers reach it, stay with it, and leave it. Companyon’s argument was that the founders brought an established habit of measurement to their next venture. Their earlier experience had become a way of running the business.
On February 2, 2023, Welgemoed announced a $20 million Series A led by Riverside Acceleration Capital and Companyon Ventures. His plans included team growth, integrations, a redesigned dashboard, and more automation. He put customers at the center of the announcement. The money would support another stretch of product development around the same review and approval work.
Dallas, too, was a return visit. ProofHQ had opened a US office there. Ziflow went back to the city for go-to-market operations, including sales, marketing, and customer success. In February 2023, the company had 140 employees globally and 28 in Dallas. Those are a dated snapshot, but they show the second business taking shape across places and functions, beyond the original product and founding partnership.
Only two layers away
The developer in Welgemoed has remained close to the executive. He has described being involved in product architecture while trying to keep unnecessary administrative work away from the people building it. That combination makes his early-career story more than a colorful complaint about an old boss. It provides a test for his own decisions: what is being asked of the team, and what useful work will the request produce?
In a 2026 LinkedIn post, he described only two management layers between himself and Ziflow’s most junior engineers. His preference is for people who understand the roadmap and take initiative. He acknowledges that growing teams need guidance; he also argues that the right people can work with fewer layers for longer than companies sometimes assume. Ownership, in his account, reduces the need for repeated handoffs.
The distinction matters in a company selling smoother collaboration. An organization can introduce delays through its own reporting lines just as a customer can introduce them through a review process. Welgemoed’s stated approach puts responsibility near the work. Its appeal is easy to understand from the developer’s chair: fewer people standing between a question and someone able to answer it.
His hiring irritation is equally concrete. Candidates often apply without trying Ziflow’s free trial. He wants them to upload something, explore, and arrive with a sense of the product. Twenty minutes of curiosity, in his telling, can reveal an instinct that a prepared interview answer conceals. Someone who wants to work on software has an obvious opportunity to investigate it.
The checklist enters the AI age
In November 2025, Welgemoed introduced ReviewAI in public preview. Its first surfaced capability used AI to evaluate review checklist items against textual criteria. A team could define requirements, have the software perform an initial check, and receive an explanation or suggested revision when something appeared to be missing. Reviewers could accept, reject, or edit the recommendations.
The allocation of responsibility is the interesting part. The organization supplies its standards. Software helps examine an asset against them. A human keeps control of the decision. For Welgemoed, this extends a long-standing interest in repetitive work surrounding creative production. It gives the technology a defined assignment within an existing process.
Company-reported activity for 2025. These figures describe the platform, not personal achievements or revenue.
Ziflow’s year-end account reported more than 6.5 million proofs reviewed in 2025 and more than 150,000 stakeholders involved. At that volume, even a routine check is work repeated many times. The figures give a sense of why consistency interests him. Each asset still has to reach people who can decide whether it meets the requirements and is ready to leave the building.
A longer clock for creative work
Welgemoed also hosts conversations about the wider business of creativity through In the Flow. In one discussion with marketing leader Temilola Agbede, the subject was the tension between immediate returns and long-term brand building. He used his own familiar choices, including Nike, Apple, and Marriott, to consider how years of exposure influence a purchase long before it appears in a sales report.
That discussion adds something to the portrait of a founder interested in operating metrics. Measurement can explain part of a business while leaving other valuable work difficult to see. A team can organize review carefully and still need room to make something distinctive. Speed through approval has a purpose when it helps that work reach its audience with its intent intact.
The unfinished business in Welgemoed’s career lives in those everyday decisions. Which version are we looking at? Who needs to respond? Is the request worth someone’s weekend? He has returned to the same field with a familiar partner and another product because the questions keep arriving. Somewhere, a piece of creative work is ready. Someone still needs to say yes.