Building in public - episode 1


Getting up to speed: an overview of where we are today and how we're building our data analytics SaaS application in public.

Starting out

We've always had the ambition of being more than an accounting firm. We started out with the vision of adding real insight to fast-growth and complex businesses. That's part of the reason we chose the word flinder, a name and a brand that can span verticals. As such, we've been working hard for over 12 months building our own SaaS application – a data analytics platform. A platform that can provide real-time relevant business-wide KPIs and metrics. The really complex ones that investors ask for. Something that translates the complex into the simple. Reporting that’s fully integrated removing manual interventions and mistakes. That’s real-time and gives much faster insight and more proactive decision-making. That’s accessible, anytime, anywhere. Metrics you and your Board care about. Ones that are aligned to your strategy and customised for you.

Why build in public?

This building in public series is the start of being open and sharing more about the journey to date, and the crazy ride we're probably going to go on. Let's start with where we're at right now. Why did I choose to build in public? Well, a few reasons...

  1. To keep us accountable.
  2. To get early feedback.
  3. To share our experiences/inspire others.
  4. To document, and reflect, on our journey.
  5. To live and breathe our philosophy of transparency.
  6. To show we truly understand our client base!

And I should probably give some kudos to one of our clients Heights, who have been building in public for some time now and inspired this move for us.

Getting up to speed

If you're not aware, almost six years ago Luke and I left PwC (big accounting firm) with a vision of building and running outsourced finance functions for fast-growth businesses - something we now call Smart Finance Functions®. We saw a gap in the market where accounting firms were too focused on yesterday, and too focused on just the financials. They didn't understand their client’s business, they didn't care enough, and they didn't look at it from business perspective.

While we (mostly) had a great time at PwC, we also set out to create a place we would be energised to get up on a Monday morning and get to work. We figured if we built a place that was fun for us, it would be fun for others too.

If you aren’t embarrassed by the first version of your product, you shipped too late. Reid Hoffman.

But, at the heart of this vision was delivering insight through game-changing technology. And this is where our passion continued to be. As we were building the finance function side of flinder, we settled on an application called Klipfolio (after exploring PowerBI, Tableau, Qlik, Grow etc.). While this kind of worked, it was always going to be an interim solution. Ultimately, we ran into too many limitations: speed, configuration, data sets, data volumes and lack of inspiring visualisations.

So, we decided to embark on building our own SaaS application to own our future. We used profits from our finance function business to (bootstrap) invest in building our own tech product, internally known as Project Liberty.

Where are we today?

This is episode 1, so it’s largely about bringing you up to speed for the past year. I think the first thing I’ve learnt is, it takes a lot longer to build a SaaS product when you’re bootstrapping!

Most founders only look forward, so writing this is a nice way to reflect on what we’ve achieved. Here’s a snapshot to get you up to speed.

Milestones achieved

  • Product launched in beta for invite-only.
  • Clients migrated across from legacy (Klipfolio) platform to Basic plan.


Our DevOps team is based in Belfast, Northern Ireland. We have three in our team there. Also, both me and Luke spend a fair amount of time involved in the strategy and execution of Project Liberty. The team comprises:

  • CEO - Alastair
  • COO - Luke
  • Product Manager - Chris
  • Front-end React developer - Matt
  • Back-end Python developer - Jamie


We’ve built and rebuilt. We built out as an MVP, and then rebuilt as a full product (and then rebuilt the front-end again by decoupling it and building in React to make it faster). As a general user, there’s so much you take for granted when you use a cloud application – I have a whole new appreciation of what it takes! Even designing the login sequence took an immense amount of time. A very high-level summary of what we’ve achieved with the product so far:

  • Built the core back-end application.
  • Designed and developed native Xero integration (we're a developer partner) and API catalogue to many other applications i.e. Hubspot, Salesforce, Shopify, Airtable among others.
  • Designed and built back-end tables for storing, mapping and manipulating data.
  • Add/remove two-factor authentication.
  • Add/remove anti-phishing code.
  • Designed and built portfolio functionality.
  • User profile and allocation of client, roles and profile.
  • Designed and built user journey.
  • Designed and built portfolio functionality.
  • Applied chart.js catalogue and customised specific charts e.g. dynamic range waterfall.
  • Built standard dashboards integrating into Xero.
  • Built modal pop-ups by tile.
  • Added custom filters by dashboard on a right-hand nav bar.
  • Built left-hand nav bar for dashboard selection.

There were too many challenges to talk about, but a few that come to mind are:

  1. Multiple navigation bars.
  2. Charts misbehaving.
  3. Data loading speed and streamlining data volumes to pass to front-end.
  4. Dealing with bi-directional data.
  5. Heavy security emphasis and eliminating risk of cross-contamination of data between tenants.

Legal and security

There are a fair few legal aspects to cover off. We engaged with legal support and other professionals but it's possible to do this a little bit cheaper. Although, you get what you pay for.

  • Engaged with legal advisors (The Legal Director) to develop user agreement, license agreement and terms of use.
  • Became Cyber Essentials certified (ramsac) – this is important to demonstrate good practice and high integrity.
  • Performed initial penetration testing (Dionach) and remediated initial high-level issues.
  • Implemented continuous penetration testing process (AppCheck).
  • Developed our security statement (you can take inspiration, crib or just copy from ours security statement).


  • Migrated 100% standard dashboard clients from Klipfolio to Project Liberty.
  • Migrated 40% of non-standard dashboards clients.


After some discussion and debate, we concluded (for now) to adopt the same brand for our tech product as for the finance function. The investment it takes to build a brand is significant, so doing this twice over wouldn't be realistic. As a young company, the wise thing for us is to gain synergies across both, which there are many, especially as our ICP (ideal customer profile) is the same. Both sides enhances credibility which each other too.



While this is the part most want to see on building in public, it's not as meaningful as the story. Not at this stage anyway as it’s been about building the application and foundations.

  • Basic (free): 58 logos
  • Custom (paid): 3 logos
  • Basic (free) 171 dashboards
  • Custom (paid): 12 dashboards
  • Individual users: 77
  • MRR: £2k
  • ARPA: £667

What’s coming next?

We have quite an ambitious roadmap ahead of us this year. We’re in the middle of some custom builds and have a few already in the pipeline for Jan/Feb but from a product perspective, we’re building out the following features:

  1. Standard dashboard custom P&L groupings: allowing greater configuration regardless of how the chart of accounts or P&L is set up in Xero.
  2. Knowledge hub: help videos on how to change various settings and basic navigation around the application.
  3. Demo environment: this is a big one for bringing to life how the application can work for to get the best out of financial and non-financial data for SaaS and tech businesses.
  4. Migration: continue to migrate non-standard dashboards from legacy app to Project Liberty.

I would love to hear any feedback, and what you would like to hear more about! Feel free to get in touch with me.

Until next time 🤟

Want to read more insights like this? Sign up to The Journal!

Sign up for our latest insights