How we opened a digital agency. The ups, the downs, and all the in-betweens…
This article takes you behind the scenes on how I, alongside two of my college classmates, started and grew our digital agency in Toronto, Canada. What started with wanting to just “gain experience” and hopefully have substantial outcomes, turned into accomplishing both of those goals as well as achieving a scalable system for web software development with a long list of dos and don’ts that came with it. Over the course of two and a half years, we obtained immense experience that will forever be priceless. By sharing it in this article, I hope it brings value to individuals who are considering opening their own agency, are currently doing so, or are simply newcomers to this industry.
Introduction:
The thought of starting a digital agency struck me rather abruptly, so let me take you back to 2019 when it all came about. It was mid-semester, pre covid, and most of my friends and I’s time was spent working on minor freelancing tasks. We each had 1–2 years of individual work experience at that time but had never worked as a team. I was very aware of the talent each of us possessed, so I mentioned possibly joining forces to not only grow as specialists but achieve something on our own. However, we had no concept of what a collaborative project even was, as well as zero experience in designing our own product.
We were all still interested in somehow starting something together but we didn’t know where to start. So before we drowned in our own project, we started by choosing to learn how to operate efficiently in a team, build good development habits, and make some early funds. Among us were a UX/UI designer, a full stack web developer, and someone who understood the team’s overall responsibilities (former front-end developer). It made sense that providing our services together as an agency would be a smart first step, as well as a way to build even more experience learning from other companies we would eventually work with.
The decision was then made and we came up with a name, designed our first logo, applied for company registration in Ontario, and most significantly, started a work chat all in one day!
Forming our business proposal:
We were soon ready for our first test projects, but first, we needed to figure out who we were selling our services to and what our ideal offer was. It was important that both our team and our customers benefited as much as possible.
After taking the time to analyze our previous freelancing experience, we decided it was best to specialize in the development of small websites and simple web applications.
Clients?
To create a more thorough offer, we first had to identify a market segment where our services would be most in demand.
Following a brief examination of the client market and prospective rivals, it became evident that our services were only valuable to startups and young companies like ourselves. We found numerous resources for market research, the most important of which was awwwards.com (which publishes the best websites from around the world every month). We used this site to look through the work of other agencies and gathered what ones we felt the closest to.
When we compared the services and costs that the finest agencies in Canada were displaying, we realized that it didn’t make sense for us to attempt to offer our services to big, solid businesses at the time. Due to our lack of experience and resources, even if big customers would be interested in us, we would cause them and ourselves more issues than advantages in the end. As a result, we made the smart decision to make our ideal customers SaaS startups in pre-seed investment rounds with a maximum investment of approximately 1 million dollars.
Ideal offer:
When startups obtain their initial funding, they often start by upgrading their product, as well as their website, in order to develop further as a company. They most likely had a previous website done badly on services like Wix or Squarespace, which works perfectly for the early stages but is very inconvenient for future development.
Our plan was to get experience with as many projects and firms as possible, which meant executing little projects so we can not only move on to the next one quicker but avoid drowning in bigger projects as well. We didn’t want to just create template landing pages that everyone had and that looked similar to sites built using Wix, Squarespace, or Webflow. As a result, we decided that the development of small, non-template sites for businesses in the early phases of investment was our perfect solution.
We preferred SaaS businesses since we were interested in developing our own SaaS application in the future, but we didn’t limit ourselves to working with companies in other fields. We also set the average cost of projects at 1–5k Canadian dollars after evaluating our rivals and prospective customers in Canada. This pricing was 3–5 times cheaper than the market average and precisely demonstrated the kind of clients and projects we were interested in.
First clients:
Our first clients were a tiny start-up helping people to find and hire professionals in a variety of industries (plumbers, cleaners, movers, etc.) throughout Canada. We had our first meeting in the UofT library, where we met the creator of this little app. We spoke about our agency’s concept and that we were in the process of looking for our first customers. They were interested and told us more about their application as well as the fact that they didn’t have a UI/UX designer in their team.
We then advised them to forgo hiring a designer and instead collaborate with us for a few hundred dollars in exchange for helping them with their early stages of design and research. They eagerly agreed since the cost of working with a freelance designer was almost the same price.
Our network started growing rapidly, as, before quarantine and isolation, we were unwillingly surrounded by a large number of professionals in many disciplines, as well as entrepreneurs throughout or around campus. As a result, seeing such individuals at a library or a coffee shop near the university was unavoidable!
Client acquisition:
We explored a variety of tactics to find new clients, including cold phoning, touring around corporate buildings, and developing a lead-generating funnel. However, for us, word of mouth had always proven to be the most efficient method. Even with other strategies like social media marketing to promote our agency, it still never compared to the leads we received through word of mouth. It’s as if our customers became our own salesmen, armed with the most persuasive reasons to work with us simply by showing others the websites we created. These simple conversations turned into the completion of numerous successful projects for us.
After speaking with the owners of other, more successful agencies, it was revealed that word of mouth continues to be the primary source of leads, even for businesses with more than ten years of experience!
LinkedIn, on the other hand, had also proved to be a very useful tool for us. We were able to track the proprietors of small businesses, communicate with them, and eventually arrange phone conversations. It was a powerful tool at the start of 2020, with an average conversion rate of 1 in 20 prospects who agreed for a call.
Other agencies’ projects!
Only a few relatively tiny tasks were completed over the first six months of starting our agency. We were moving in the right direction, however, we lacked the necessary expertise to take on more substantial projects and expand our range.
So we decided to try and persuade our acquaintances from other agencies in Europe and Canada to assign us some of their minor projects. This would help them to minimize their operational expenses as well as provide us with additional expertise at a lower cost.
Over time, collaborative projects accounted for half of our business, all while we developed expertise and were able to save the other agencies substantial time and effort. As a result, we were able to have a steady pipeline of new projects coming in. Such initiatives often outpaced our operational capabilities and fueled our company’s expansion.
Finance:
Our client-interaction system was designed in such a manner that we would only go on to the next stage of the project once the previous stage had been approved and paid for. As a result, the customer paid after each stage was completed, and that way, we avoided the possibility of being left without payment after completing all of the work.
We didn’t expect to make a lot of money as our major purpose was to get experience. However, the financial plan was designed to balance expenses and provide a modest cushion. This gave us security in case there were issues with the project and we needed to pay our personnel for certain phases while waiting for the customer to pay the final bill.
Unfortunately, even with a payment mechanism after each step and a relatively cheap cost of labor, similar scenarios occurred. We were able to pay from our “financial cushion” on many occasions.
Portfolio:
The value of a well-designed portfolio cannot be overstated. Our portfolio site served as our primary word-of-mouth marketing technique. This is the sole way for a prospective client to determine whether or not a new agency is competent. Following that, our clients’ websites arose, which we utilized for the same reason.
Work from the inside out:
It took an average of 2–3 months to construct one website, including research, mockup approval, design, development, and testing. As a result, we worked on three to four projects at the same time in a quarter. With such a workload, the ability to organize the work of all professionals in the chain is critical to success.
Team:
Our team was composed of three regular workers and perhaps ten freelance professionals in different disciplines who worked with us on a project-by-project basis.
Collaboration with freelancers is far more costly and complex to organize than recruiting permanent personnel. However, on the other hand, for a start-up agency, working with freelancers has two major advantages. It provides a chance to swiftly increase the team based on the amount and complexity of projects, as well as the rate at which work is completed.
When working with freelance employees, we immediately understood that if payment is given for the job accomplished rather than the hours worked at the end of the contract, they would be more motivated to work as rapidly as possible. A strategy like this is also much more clear and easy to control when working with beginner professionals.
Hiring freelancers:
Looking for developers was one of the most challenging aspects of the job. We didn’t hunt for senior or highly qualified developers because of the typical complexity of our projects; instead, we sought out even absolutely inexperienced junior developers and taught them everything on our own. In this situation, training took place alongside a more experienced developer or CTO on real projects.
As a result, some developers and designers started their professional careers by working with us, and when their abilities surpassed the complexity of our projects, they found major projects for themselves on Upwork and in other firms within six months or a year.
For us, the lower cost of labor was also an essential rationale for working with beginner professionals. A paid test task was a key stage in the process of finding young and inexperienced freelancers. We quickly learned what specialists we wanted to collaborate with after seeing the results of their well-designed and paid test task. We knew right away whether we’d have troubles and what type of problems we’d have if any.
We had a limited amount of time when doing research to find designers and/or developers for a new project. As a result, when we needed a new freelancer, we accepted applications from 3–5 individuals at once and assigned them a test task. On average, 1–2 freelancers performed well on their test tasks, and their expertise, as well as skills, fulfilled our standards. After ongoing quick interviews, this then resulted in having new freelancers ready to work within a week after posting our application on Upwork.
Work process organizing:
At the start, supervisors were assigned to each project to monitor and verify the work of the specialists throughout each step. The planning and search for development/ design freelancers took us on average 3–5 days. This occurred concurrently with the project research stage, therefore this procedure was usually unnoticed by the client.
Scalability:
Each project was distinct and challenging in its own way since the order and number of steps in each project varied based on the customer’s requirements. The key challenge, however, was to run 2–3, and occasionally four projects at the same time without disrupting the process’s sequence or integrity. Doing this right is incredibly tough, so it was critical to plan each project from the start, as well as at the stage of producing the SOW.
In my role as a PM at the agency, in addition to project planning and development, I had to arrange all of the work for each of the three essential employees, including myself. Our project management system allowed us to handle up to four projects every quarter with good time management.
To overpass 3–4 projects for each quarter, the supervisory and team lead workforce would need to be increased by an additional two people (supervisor for working with designers and team lead for working with developers). It would then be possible to accomplish up to eight projects every quarter.
We tried this method for one quarter but then abandoned it because when working on 8 projects at once, we started to lose emotional connection and weren’t as engaged in the development of each project.
The record of all our project choices and activities was a crucial success element. As a result, if the project failed, we could always go to our timeline to see where the error occurred and how to prevent it in the future. These were initially simply noted with dates, but then evolved into a full-fledged document on Confluence. Even today, we have access to a comprehensive history of our choices and activities for all initiatives that go back to two years ago.
Our mistakes:
- The projects that were too big:
We were often contacted about developing major SaaS applications. However, we generally understood that huge projects mixed with our speed and style of work would never work out. There were occasions when it seemed that the planned project was scalable for us and not that enormous but once starting so-called projects, we realized we were absolutely doomed and in situations where success was clearly not conceivable and our estimations were incorrect.
To prevent similar scenarios in the future, we 1. Constantly examined the API of any third-party services with which we had to deal with 2. Performed research on comparable projects to identify where other people had previously run into problems. 3. As if it were a different project, we went through all of the essential microservices. 4. Always interacted with the customer on their grasp of those areas of the project and inquired if he had any special requests. If the client had specific technical needs, we thoroughly examined them and came to an agreement on our SOW with them.
Even though this is just a tiny portion of what we’ve learned to pay attention to, we could have prevented half of our unsuccessful projects in the first year of operation and saved time, money, and stress.
An example of such a project from personal experience: We were once approached with developing a service for the European taxi drivers. The web application was designed to be a supplement to the company’s current CRM for organizing and coordinating taxi driver operations. The goal was to leverage CRM’s current API to build a simpler and user-friendly version of CRM for operators that was free of unnecessary buttons and complexity.
We took on the project after confirming that CRM does indeed provide an API and that it works with all of the functionalities we want.
We soon learned from our back-end developer, just two days after the start of development, that the API was broken and that we couldn’t do anything. We were taken aback by such a statement and decided to investigate, only to discover that the API provided completely random data that had nothing to do with the documentation.
We wrote to tech support 1,2,3 times right away… After speaking with the most inept tech support in our life, it became evident that the API could only be sorted by the “poke method.”
We informed the client, but they were dissatisfied with the result. We were then informed that because we had accepted to take on the job, we would have to do our best to comprehend the API.
This CRM and its API turned out to be quite outdated and incomplete, and we weren’t the only ones to run into a similar problem with it. As a consequence of such events, there was an incomplete project, four months of client time squandered, and a significant amount of resources spent in vain.
2. Employee overload and burnout:
When dealing with young and inexperienced specialists, whether developers or designers, we couldn’t rely on them to notify us if they were burnt out. Since most of our workers were experiencing burnout for the first time in their life, they weren’t aware of how to communicate it or stop when the early signs of it arose. Instead, a developer would work on 2–3 projects in a succession, then continue to work on 2 more projects until he went missing for a week, with half a project still incomplete, and returned with the words “I can’t do this anymore…”. The implications were terrible: overloading the remainder of the project’s staff, rearranging the project on the go, missing deadlines, and so on…
To prevent such situations, regardless of their expertise, we learned to engage with staff after each project, offer them relaxation, and interact with them on a regular basis about burnout and work-life balance. Secondly, we learned that in the event of such an incident, we should always have one developer and/or designer on standby in case we need to replace an employee fast, even if it comes at a larger cost for us.
Conclusion:
After more than two years of hard effort, we have honestly outgrown our business model and decided to shut the agency six months ago. We chose to go back to our initial intention and work on personal projects after gaining a lot of expertise. We had gotten the most out of our time at the agency, in my perspective, and were ready to move on.
We’ve all gathered invaluable experience dealing with a large variety of companies and, most crucially, all sorts of people over this period. We’ve learned how to plan projects, organize our work and the people around us, get things done, analyze our strengths and the projects we’re working on, and uncover and address difficulties.
Here is a link to our portfolio, which includes some of our most successful projects:
We experienced an astounding amount of wins but also a lot of failures including uncompleted projects, and disappointed customers over the course of two+ years. However, it was, nonetheless, a crucial event for our development. We progressed, honed our talents, as well as identified and resolved obstacles. The first six months of work were the most challenging, and we would not have made it through this period without the help of our friendly team, and our customers that developed with us. We truly would not have learned the skills that are essential today in the development of our new projects. Alio Agency will forever live on as something that changed our lives.
Follow me on social media to keep up with my projects!
Twitter: https://twitter.com/LeoFedoseev
LinkedIn: https://www.linkedin.com/in/leofedoseev/
Email: lev.fedoseev@gmail.com
