Building a Business on Merging Duplicate Tickets: How Sorin Alupoaie Grew Swifteq

An Eleven-App Business Run by One Person

In 2023 Sorin Alupoaie shipped five new apps at Swifteq. The number of apps listed in the Zendesk Marketplace grew to eleven, and by December of that year he had 230 customers and 150,000 euros in annual recurring revenue. Once he could pay himself a salary and still keep a profit, he could run the business without depending on another job or on investors. Revenue coming in from several apps had started to support a developer's independence.

Before that, Sorin had bet his livelihood on a single product. When he built CX Moments in 2016, he went full time on the strength of a working prototype and potential orders, and he had to raise money early to cover his salary. The business ended without finding a product that matched real customer demand. On the next attempt he kept his job while building, and he saved roughly a year of salary before leaving. Swifteq was the second business he prepared for that way.

Finding a Product in Zendesk's Repetitive Work

Swifteq's starting point was support teams using Zendesk. Reading Zendesk's official community and posts from Support Driven, where support practitioners gather, Sorin looked for the complaints that kept coming up. The work of merging, ticket by ticket, the inquiries the same customer sent several times stood out, and he built an app that merged duplicate tickets according to defined rules. Narrowing the customer first also narrowed where to look and which problem to solve.

He had also run a business on outside investment and felt his freedom to make decisions shrink as the tension grew. With Swifteq he wanted to take responsibility for the product and the customers himself and build a profitable business that could support his family. Starting alone and taking no outside money were choices aimed at that goal.

The first app shipped with only the features it needed and took under two weeks, and he charged for it from the start. Sorin limited development to what he could handle while working a job. According to the approach he described in 2023, most apps offered only enough of a screen for users to switch them on and off, and he handled the complicated configuration on his own server side. Apart from two exceptions, he never spent more than a week on the first version of an app, and he filled in features and settings after watching how real users worked. When he added features he favored problems several customers shared.

The ticket merging app connected into how Zendesk already handled work. When a new ticket appears, Zendesk's triggers and webhooks pass it to the app, and the app checks the configured criteria to decide whether it is a duplicate and merges the tickets. It could match on shared identifiers such as an order number as well as on the same customer, and it could leave out tickets created by internal staff or bots. Support teams kept the system they already used and added automation that followed their own working rules.

Extending One Customer's Work Across Several Products

The apps all grew in the same direction. Automation that cut an agent's repetitive work, tools for managing help center articles, and tools that helped with translation and writing replies all attached to the daily routine of a support team. Within one organization they took on different jobs: handling inquiries, maintaining documentation, and supporting several languages. Swifteq widened its product range inside the support market.

The help center products show how the expansion worked. Help Center Manager finds and replaces outdated wording across articles, fixes broken links, and edits many articles at once. Help Center Translate translates documentation into several languages while keeping the formatting, re-translates only the parts that changed, and applies the company's glossary. Because the people managing documentation and the people translating it overlap, a team that adopts one tool has a reason to hear about the next.

Help Center Analytics shows whether that documentation is doing its job. It tracks views and usefulness ratings per article, the share of readers who still file a ticket after reading help, and the paths users take. Managers can use that to decide which article to fix first. Tidying documentation, translating it, and checking the results run as one product line.

The portfolio also included free apps. Help Center Export exports help center articles and related information to CSV files and checks published documentation for broken links or images. A user can create a Swifteq account and audit the state of their documentation, and when they need more export formats or bulk edits they look at the paid Help Center Manager. The free tool finishes real work and points customers with bigger management needs toward a paid product.

Free and paid apps shared the same account system. Sorin saw people who used a free tool go on to look at other paid apps, and he kept the relationship going with a weekly newsletter to app users. Each app therefore both earned revenue and helped people discover the other products. Adding a product could also draw on the touchpoints he already had with customers.

Customer Stories That Led to the Next Product

The outdoor route service Komoot, a real customer, used three of these products together. Running documentation in six languages meant moving translations back into the document formatting and hunting down related articles whenever a link changed. In the customer case Swifteq published, Komoot said the time spent producing and managing translations dropped by half and that improving its help center based on the analysis cut the share of customer inquiries by 20 percent. The three apps covered consecutive steps of the same team's work.

The buying process at the home exchange service HomeExchange shows the connections between products even more clearly. While handling inquiries the same member sent through several channels, the person leading support found Swifteq's ticket merging app in the Zendesk Marketplace. After a trial they confirmed the effect, and the published case says more than 10,000 inquiries a month were being merged automatically. The first purchase began with one problem that was wasting time right away.

The same person later subscribed to the Swifteq blog, saw the launch of the conversation translation app, and tried it. Having already had a good experience with the merging app made reviewing a new product easy. When a customer's thank-you note on a closed inquiry reopened the ticket, they brought that problem to Swifteq as well, and Sorin's side helped handle it with an existing AI automation app. One sale turned into a conversation with the same customer that surfaced a second and then a third problem.

Being established inside Zendesk also helped secure a sales channel. The marketplace brought in customers looking for a specific capability, and it made relationships possible with Zendesk staff and with the consultants and partners who help companies adopt it. When those people set up a customer's workflow and run into a need the basic features cannot cover, they look at outside apps. Telling them about the problems Swifteq solves was another way to reach new customers.

Refining Pricing and the Way It Sold

The billing reflected the nature of the work. The ticket merging app splits its monthly price tiers by how many tickets it checks for duplicates each month, and even the entry tier includes custom rules and exclusion conditions. Cleaning up a backlog of past tickets all at once is sold as a separate service. Ongoing inquiries became a subscription and one-off cleanup a separate charge, turning two different kinds of demand into revenue.

The way it charged also matched how customers used it. Most bought a monthly or annual subscription, but some features were needed for only a few weeks, such as duplicating a help center. For those customers Sorin introduced a 30-day pass, and some came back later and bought an annual subscription. Offering a purchase method that fit temporary work let short-term users in.

The way he met customers was matched to the same group of practitioners. Sorin publicized Swifteq by offering free tools and useful articles, with help from Supported Content, a content studio in the support field, and specialist writers. The topics came from the questions and discussions that moved through Support Driven. The customer observation he used to build products also fed the content customers would search for and read.

Pricing involved trial and error. In a retrospective Sorin admitted that he had sold most of his early apps too cheaply and had taken too long to fix the prices. Low prices delayed the point at which he could secure profitability and reinvest in the business, by his account. Help Center Manager and the analytics tools in particular had more value for mid-sized and large companies, so the price needed to reflect the value they provided and the profit required to keep the business running.

The results Sorin published in January 2023 were 120 paying customers and 5,000 euros in monthly recurring revenue. The customer list included companies such as FREE NOW, Zapier, and Hotjar. In June of that year annual recurring revenue reached 100,000 euros. A set of apps solving a few specific jobs was starting to become a recurring line item for business customers.

Operating Burdens That Grew With the Business

More products did not mean every app grew at the same pace. In the second half of 2023 churn increased, growth in one of the main apps stalled, and adoption of some new products was slower than expected. At the same time support requests took up so much of Sorin's time that he had less to spend on product development. A business holding eleven apps still had work to do after eleven launches.

Sorin started by cutting the repetitive operations. He automated part of the trial and customer journey and, with the help of a writer, rebuilt the help center to match customer questions. He also worked with a content studio that knew the support field to publish articles useful in practice. He kept the shape of a solo founder and drew on outside expertise for the work that needed it.

He also saw what focusing on development had cost him. He admitted that he missed a search indexing problem on his website for several months and that by keeping too much of the development to himself he lost time he could have spent growing the business. He later advised founders to divide their time between building new apps and improving and promoting the ones already out. At the stage of running a portfolio, the decision is not only which products to launch but how much time to give the products already launched.

In June 2024 the monthly recurring revenue Sorin published passed 20,000 euros, with more than 300 customers. At that point he said every recently launched app had paying customers and that some of the new products were contributing significantly to growth. In an interview published in November 2025 the monthly recurring revenue he reported was about 44,000 euros. Selling several apps on repeat had turned into rising recurring revenue.

Choosing Customer Problems amid Platform Changes

Alongside the growth, the risk of depending on a platform became real. When Zendesk began offering a feature similar to his attachment cleanup app as a built-in, Sorin worried about falling revenue and churn in that app. Having other apps limited the blow to the business as a whole, but the fact that Zendesk's product decisions affected his revenue did not change. Adding more apps and reducing platform dependence were separate problems.

Help center translation went differently. According to a September 2025 retrospective, two months after Zendesk shipped its own translation feature customers had stayed and revenue was still rising. Sorin also reflected that while he had been avoiding the possibility of overlapping with the platform, he had stayed with peripheral features where demand was small. Afterwards he decided to take on problems that matter to customers even when Zendesk might address them.

At the same time he moved Help Center Translate to Intercom. The two platforms' data integration works so differently that he had to redesign part of the product structure, so it was not a matter of copying the existing app across. Even so, the problem of translating support documentation and the experience he had built up came with it. A portfolio that had grown by adding apps within one customer group added the choice of widening to another platform.

The important moment in Swifteq's expansion is when the HomeExchange contact decided to try a second app. Trust earned from the first product and its support made reviewing a new product possible, and they passed along other problems Sorin had not solved yet. The business value of a portfolio in this case lies in getting to know customers you have already helped more deeply and meeting them again as customers of the next product. Doing one small job well became the starting point for the next sale.