The story of Flatland

Just over one year ago, Conor and I decided to start a consulting company together. We had both thought about doing consulting separately, so when we each found out about the other’s plans it just made sense to team up. Teaming up also had the added benefit of holding us accountable. If I had started this company by myself it’s doubtful that I would have accomplished anything. The same probably can’t be said for Conor; he’s a workaholic. In any case, this is the story of year 1 of Flatland Data Solutions.

The beginning

At the time, both of us were working at the same tech company in Saskatoon (and we still are at the time of writing). Saskatoon is a pretty big tech hub (we’ve got a big tech IPO on the horizon), despite being a flat, cold, unpopulated, agrarian province. Despite this there were few companies with any sort of data program, and even fewer with machine learning programs. This is understandable since good, full-time data scientists can be hard to find and costly to retain. However, we believe most companies are ready to start capitalizing on data, and the only thing most are missing is expertise. We realized that we could provide that expertise and we started thinking about how to make that available to local companies. We decided to form Flatland with the goal of getting local companies started with data.

It was March 3rd when we got our business license together and set up a business account. We exited the bank, sunlight gleaming, a sparkle in our eyes, and started looking for networking events we could attend, drop some business cards, and shake some hands. We managed to get to one event before covid shut everything down.

Who we are

Conor Lazarou
Conor Lazarou
I suspect that Conor’s attempt to replicate Einstein’s haircut is intentional

Before we go any further, I should introduce ourselves. Conor is a machine learning genius. He has an amazing medium blog, he’s done some fantastic work with GANs, and developed a unique neural network for a complex application at the tech company we mentioned earlier.

Myself, I’m more of a generalist. My background is in software engineering, my master’s thesis was focused on deep learning, and I’m the data lead at the same company, so I offer a broad range of skills and a unique perspective.

Together, we’re able to do everything from setting up data collection, building analytics and business intelligence tools, training an organization to make data driven decisions, and, most impressively, building machine learning tools to offer as products or improve business performance. I’m currently putting together a multi-part series about how data can be used to accelerate your business, so feel free to stick around until those come out, or if you’re really curious, send us a message and we can tell you exactly how your business would benefit from a data strategy.

Logan Kopas
Logan Kopas
Not pictured: everyone in the audience falling asleep while I talk about data warehouses

Building a business in a pandemic

Let’s get back to March. Conor and I had just attended our first networking event as Flatland Data Solutions when, suddenly, everything was shut down. Personally, having a new company to work on was a bit of a blessing. While most people were baking sourdough and trying to find any hobby to stave off boredom, we were able to spend our time setting up our website, writing blog posts, and figuring out how to run a business. For the first 6 months we spoke to a few companies, but nothing ever really came of those conversations. Which was fine, because we were soon going to figure out that it’s a lot of work to run a business while having day jobs.

Networking in person wasn’t really a strength either of us had, but we soon found out that networking virtually is even harder. Neither of us are the type of people to cold call someone in order to build connections, so we mostly just put content out and hoped people would reach out to us. We’ve had moderate success with this; we made a few connections and had a few conversations with various businesses that were tentatively interested in what we could offer. I’d also like to make a side note here: we’re happy to talk with people, even if there isn’t a business fit. We started this company because we want to see local companies succeed, and that’s true even if we don’t have a business relationship. I’m always happy to network and meet new people, and offer some advice or suggestions!

Conor even combed his hair for the networking event

Remember how we said we started this business with the goal of helping Saskatoon businesses? Naturally, our first real contract was with an Australian company. They had read one of Conor’s blog posts, borrowed some of his code, then hired us to help out when the business outgrew that borrowed code. Another side note here: if you want to run a company in the evenings, but you work 9–5, pick up a job halfway across the world; your day job hours will never overlap!

Working with this client was an amazing opportunity. We were able to build an amazing machine learning system similar to the kind of thing UberEats would use. We’re incredibly proud of it and know it will be a core part of the client’s operations. Now that we’re nearing the end of this contract, we’re hoping to get back on track with our original goal of helping Saskatoon businesses, so if you work at a tech company in Saskatoon, we may drop by your LinkedIn inbox. Hope to see you there!

What we do

You may have heard the saying “Data is the new oil.” Conor and I believe this is true. However, if you found oil in your backyard you wouldn’t know what to do with it, you can’t just toss it in your car and start using it. You need to know how to use it, it needs to be refined, you need tools and distribution systems. That’s where we come in (except for data, not oil). The beautiful thing is that if you have a business, no matter how small, you probably have at least a little bit of oil (data). I’m working on a series of articles that will go into this in more depth, but this section will give you a bit of an overview.

Now, at this point you might be thinking, “my business isn’t online, so how would I have data?” The idea that you need to have a strong web presence and a lot of technology in order to capitalize on data is a myth; your business generates a lot of data, but you might just need help collecting it.

Small/new businesses (0–2 years)

This is possibly the most crucial time to start collecting data. Data can act as your compass as you’re figuring out your business and determining how to achieve sustainable growth. There are so many insights that can be gained at this point that can drastically change your business trajectory.

From a business growth perspective, data can inform you about your customers, which will allow you to be more targeted in your customer acquisition. Whether you’re just getting started with an online store, building out an app for some core customers, or prototyping ag equipment, it’s extremely valuable to know where your business is coming from.

From a business operations perspective, it’s extremely valuable to know where you’re spending your time. How do you know when it’s time to start hiring people to take some load off your plate? You might be able to judge by intuition when it’s just you and a handful of teammates, but what about when you’ve got a sales department? Should you spend that extra budget on marketing? Proper data collection and analysis can help you allocate your time and resources, find inefficiencies, and maximize your ROI for any money you spend.

You might think that it’s a lot of overhead in order to set up this data collection, but that’s not the case at this stage. It’s completely reasonable to collect data in spreadsheets and notes at this stage, and being intentional about it will streamline the process and allow for efficient workflows to ensure you don’t get bogged down in the details.

Whether you have a full team, or it’s just you working from a coffeeshop, it pays to capitalize on your data. Photo by Andrew Neel on Unsplash

Medium business (2–5 years)

At this stage you’re possibly getting to the point where spreadsheets aren’t cutting it anymore and you’d like to set up something a little more automated. Now, you’ll want to evaluate the cost of setting up a more effective analytics infrastructure. You might also want to start building a data culture throughout the organization. This culture can ensure that everyone in your organization is making effective decisions and that each one incrementally pushes the organization forward.

Cultivating a data culture can be tricky. Not only is it important that people have access to data that can inform them about the decisions they need to make, it’s also extremely important to ensure that people can feel safe discussing data that conveys “bad news” as well as “good news”. It’s an easy trap to fall into where people only feel comfortable reporting charts that go “up and to the right”, but identifying problems and bad news through data is how an organization learns, grows, and gets stronger.

Mature business (5+ years)

Once you’re comfortable with analytics and business intelligence, the next step is to start getting revenue from your data. This could take the form of reports and dashboards that you offer to your customers, or it could be through machine learning and AI products.

Now, machine learning doesn’t have to be offered as a product, it can also be used to optimize internal processes, streamline workflows, and predict risks and losses. The real value here is recognizing that you likely have enough data to start implementing smart systems that solve problems for you or your customers.

Conclusion

Conor and I started Flatland Data Solutions in order to help local businesses maximize their potential through data. The pandemic may have taken the wind out of our province’s sails (and you know that’s saying something if you’re from here), but with a reopening boom on the horizon we’re ready to help your business soar. I like to compare data to a compass; when your ship is in a storm you can probably stay afloat without it, but if you know how to use it you can navigate when everyone else is just trying to get by.

Spreading the love for data, and maybe other things too. flatland.ai