Building a design-reference business, selling it, and moving to dynamic QR codes: Ramy Khuffash's Page Flows and Hovercode

The Design-Reference Business That Began with Shared GIFs

Ramy Khuffash built Page Flows to reduce the time designers spend studying other products. He gathered recordings of real user flows from live services and grew it into a paid library, then sold it in November 2023. Afterward he focused on Hovercode, a dynamic QR code service he had been running alongside it, moving his main business over.

His background was in accounting and information systems. After graduating he went through several jobs and worked with WordPress before switching to web development, and he kept building side projects in order to create a business of his own.

He began working with design material in 2014. At first he shared GIFs of interesting screen interactions from iPhone apps, and after changing the format several times he arrived at a newsletter called UI Movement in 2016. Running it by picking animations from other design sites and introducing them, he quickly passed 10,000 subscribers.

He talked with UI Movement subscribers to find a product that could earn money directly. Those conversations revealed demand for material that let people study how well-known products actually work. Based on that idea he built User Flow Pro, the predecessor of Page Flows.

From Custom Research to a Free Library

Early on he also tried a custom research service that recorded the flow of a competitor product the customer named and reported changes each week. The first customer paid about $1,000 a month, and he picked up more customers. But as the services he studied began blocking repeated signups, preparing the recordings became troublesome, and re-investigating the same flow every week did not suit him either. He shut the service down and returned to a library where many customers could view the same videos.

The library's first paid launch went poorly. He charged $14 a month, but for weeks he had only one paying customer, and the interest from Product Hunt did not turn into payments. He refunded that customer and published the videos for free.

The turning point came in 2018. Seeing that visitors kept coming to the free videos, he tested a one-time $29 payment. Before finishing a real checkout, he verified buying intent with a test screen, and once he saw a response he attached actual payment. He then expanded the videos and relaunched under the name Page Flows.

Making Page Flows and What It Was Worth

The product was built with Django and HTML, CSS, and jQuery, and the initial preparation took about a month. The labor-intensive part was screen recording and adding explanations. After recording websites and iPhone apps, he attached annotations and screenshots with a tool he built himself, and the videos, which started at about 100, passed 1,000 around the time of a 2020 interview. At one point he outsourced the recording and annotation work, but he later recounted that it did not work well because the process was not documented.

Page Flows' value lay in comparing, in one place, the steps a user goes through to finish a given task. A designer planning signup or paid conversion could look at which steps other services take and find what they had missed. Khuffash described this business of collecting and organizing information as "research as a service."

His own write-up analyzing Dropbox's plan downgrade flow shows how the material was used. He walked through the process of asking why someone wants to downgrade and for which specific reason, then offering annual billing to users who cited price. At the same time he pointed out that the button to continue with the change was not visually prominent, examining both the product's conversion design and the user's difficulty.

In a February 2020 Starter Story interview he disclosed Page Flows' monthly revenue as about $4,500. At the time he was the only founder, with no employees, and the library had become his main source of income. The business, which had gone through an early one-time payment experiment, was by then being run on a subscription fee.

UI Movement and search were important for customer acquisition. In particular, searches for product names like Slack, Dropbox, and Spotify together with a specific user flow accumulated. He began choosing which products to record after researching the potential for search traffic, and made sure that adding material raised both the value for existing customers and the inflow of new ones.

Khuffash also cited the relatively low operating burden as a reason a small content business suits developers. Compared with software that directly performs the customer's core work, the pressure of incident response is smaller, and he saw that he could use his development skills to improve the search, classification, and browsing experience. He also counted it as an advantage that the more material accumulates, the more valuable it becomes to use.

Hovercode, Found Through Small Experiments

Even after Page Flows settled, he tested other businesses. In June 2021 he announced a plan to put out five or six small products and then choose and grow the one or two that drew a good response. He set the development period at a maximum of ten days per product, five if possible, and decided to spend the time after launch on improvement and marketing.

Hovercode was one of the candidate products he built at this time. In May 2022 he recorded that people were coming every day to create QR codes and that cumulative scans had reached about 1,700, even though he had not launched it on Product Hunt. Seeing the signal that the codes people created were actually being used, he decided to invest a few more months and secure paying customers.

The question he wanted to answer was what people would pay for in a market full of free QR code tools. At the time he focused on features business users would pay for, such as the ability to connect their own domain. Saying that he had a tendency to give up on projects too quickly, he stated that he would push Hovercode longer than usual even if monetization was slow.

Hovercode's core is dynamic QR codes whose destination can be changed even after printing. Because the code contains an intermediate link address and that address forwards to the final page, the destination can be changed without replacing the printed material. It also provides analytics showing scan counts, the devices used, and approximate locations.

The two businesses ran in parallel for a while. In a 2023 Postmark interview he explained that he was running Page Flows and Hovercode together, and in November of that year he sold Page Flows. When he appeared on Indie Bites in May 2024, he was running Hovercode full time. It was a sequence of keeping the business that generated income, confirming real usage of the next product, and then switching his main focus.

Changing Priorities While Focusing on Hovercode

Hovercode's monthly pricing as of September 2026 separates free use from business expansion. The free plan includes three dynamic QR codes, unlimited scans, and basic statistics. The $12 per month Pro plan provides up to 100 dynamic codes and a custom domain, among other things, and the $39 per month Business plan provides 600 code creations per month plus API access, bulk creation, and team management.

The boundary of the subscription is also spelled out. Codes added beyond the free allowance become inactive after cancellation, and connections using a custom domain stop as well. Customers pay for the management, linking, and analytics of the codes they continue to use, and the service maintains those features on an ongoing basis.

Hovercode also provides integrations that let you create QR codes inside Canva and Figma. There is also a Chrome extension and an Airtable integration, so codes can be created right in the environment where they are needed. This reduces the hassle of moving to a separate website each time, and attaches the product to the design and data management workflow.

The cumulative number of trackable QR codes shown on the official site in September 2026 was about 2.72 million. The cumulative scans for those codes were shown as about 113.51 million, which shows that actual usage had grown a great deal compared with the roughly 1,700 scans disclosed in 2022.

What stands out in this case is that he kept running small experiments while changing his attitude to spend more time on products that showed a response. With Page Flows he went back to look at the visitors to the videos he had left free, and with Hovercode he checked whether scans actually occurred after codes were created. The common thread between the two businesses is the approach of observing user behavior to narrow down what is worth charging for and then concentrating on the validated candidate.