Invoicing, compliance, tax, contracts, and general paperwork can be categorized under the umbrella of "meta-work", namely work that is about work. This type of work does not directly add value, but rather enables other work to be valuable.
So, the obvious thing is to try and be as efficient as possible and reduce the amount of meta-work required to completed real work, something that can be expressed with the following equation.
Meta-work/real-work, which gives an efficiency metric.
Today we will discuss our one year long journey towards significant automation and digital transformation across a large part of our back office operations. This was an involved process that required a significant amount of upfront work, with the idea of then having massive long-term benefits.
Some of our goals for this transformation.
- Reduce or Eliminate Manual Work. Where possible, we prefer an API-centric (Application Programming Interface) approach where systems can share data between themselves and so we have zero data (re)entry.
- Improve Accuracy. Highly linked to our first goal, we wanted to improve the accuracy of our financial
- Transparency. We wanted to provide real-time data to shareholders, employees, and start to be a lot more driven by dashboards instead of gut feelings.
- Tax Efficiency. While we respect the need to pay taxes, we also have fiduciary responsibly to shareholders to ensure that our tax burden in as low as allowed under existing regulations.
How Things Stood.
Before discussing some of the more interesting aspects of this digital transformation, it's worth taking some time to review how we ran back office operations prior.
We ran three legal entities:
- Mad Creative Global Pte. Ltd. This was our Singapore-based holding company where shareholders held their shares and profits flowed up. Some billing was also done as this level, with bank accounts at DBS Bank.
- M.A.D (Cambodia) Co., Ltd. This was our main entity that employed the majority of our team, made payments to vendors and was the legal entity that entered into contracts with most of our clients. Banking was done at local level with ABA Bank, Canadia Bank, and Vattanac Bank.
- MAD Creative (Thailand) Co., Ltd. This was a legal entity based in Thailand that we used to bill our Thai clients from, which was actually a JV (Joint Venture) between Mad Singapore and a local law firm for compliance reasons as foreign entities cannot own 100% of a business in Thailand.
We can already see where the complexities start coming. Three separate legal entities across three different countries with three different currencies and a whole set of different regulations with regards to reporting periods, structures, as well as different law firms and accounting partners representing each corporate structure.
With regards to operations, contracts and invoices had to be manually printed out, signed, stamped, and delivered to client offices for counter-signing as there was no availability for e-invoicing. We, quite literally, had a full time employee whose sole job was dropping off and collecting paperwork.
The vast majority of payments to us were made by cheques, which had to then be picked up and banked one of our banks. Reminders to clients for upcoming or late payments was tracked manually and done via email, phone calls, and WhatsApp between our finance team and our client's finance department.
We mostly made payments via cheque or wire transfer, but for larger amounts this often meant that a director with signing privileges had to then physically visit the bank to sign, which often delayed payments to vendors.
As a final point, getting real-time or close to real time data, especially on projected cash-flows, was close to impossible, which caused significant problems in forecasting hires, general expenditures, and generally planning more than a few months at a time. This directly led to mishaps where we had to let go of over half of the team during one weekend, after the management team received the financial reports that were significantly different from expected.
That's not an experience that any of us wants to repeat again.
It's interesting to look back and understand why things were like they were, and the main reason is purely that we grew extremely quickly in our first few years in business, and then the above became our default modus operandi, and then nobody ever challenged it again.
There is also an inherent conflict of interest with anyone involved in the bureaucratic side of business, in which simplification and automation actually threatens their livelihoods by making their roles irrelevant - which is exactly what happened in our case, as we now have zero employees in administrative or back-office roles. This is a slightly dark side of digital transformation that is not often discussed, but the facts are that with greater efficiencies there will often be some roles that are made redundant, and often it would be futile to try and find a new, badly-fitting role for this person.
Patty McCord gives an interesting example if her HBR article How Netflix Reinvented HR:
One Netflix manager requested a PIP for a quality assurance engineer named Maria, who had been hired to help develop our streaming service. The technology was new, and it was evolving very quickly. Maria’s job was to find bugs. She was fast, intuitive, and hardworking. But in time we figured out how to automate the QA tests. Maria didn’t like automation and wasn’t particularly good at it. Her new boss (brought in to create a world-class automation tools team) told me he wanted to start a PIP with her.
I replied, “Why bother? We know how this will play out. You’ll write up objectives and deliverables for her to achieve, which she can’t, because she lacks the skills. Every Wednesday you’ll take time away from your real work to discuss (and document) her shortcomings. You won’t sleep on Tuesday nights, because you’ll know it will be an awful meeting, and the same will be true for her. After a few weeks there will be tears. This will go on for three months. The entire team will know. And at the end you’ll fire her. None of this will make any sense to her, because for five years she’s been consistently rewarded for being great at her job—a job that basically doesn’t exist anymore. Tell me again how Netflix benefits?
“Instead, let’s just tell the truth: Technology has changed, the company has changed, and Maria’s skills no longer apply. This won’t be a surprise to her: She’s been in the trenches, watching the work around her shift. Give her a great severance package—which, when she signs the documents, will dramatically reduce (if not eliminate) the chance of a lawsuit.” In my experience, people can handle anything as long as they’re told the truth—and this proved to be the case with Maria.
We reviewed each aspect of our back office operations and identified the biggest pain points, as well as dependencies between different aspects such as country of operations, different tax codes, and the software available in different jurisdictions.
We also looked at some overall principles, and the key thing that we wanted to achieve was simplification.
This means intentionally doing less, but ensuring that whatever software and processes we implement are scalable and efficient.
Simplifying Tax & Legal Entities.
Our first port of call was to look at the low-hanging fruit, and after a significant amount of research we decided to register a new legal entity called Mad Creative, Inc. based out of Delaware, United States.
Delaware State is often considered a Tax Shelter, which can be loosely defined as any method of reducing taxable income that results in a reduction of tax payments, recovering more than $1 in tax for every $1 spent within a four-year period.
Delaware also does not have any sales tax , Value-Added Tax (VAT), or capital shares and stock transfer taxes. There is, however, an annual franchise tax but this is a flat-fee based on the number of authorized shares and a few other factors, and is essentially negligible.
The other big benefit of registering in the USA is that it then enabled us to use an interesting suite of business tools and software that could start to automate our back office.
Let's now discuss payments and billings.
As a PSF (Professional Service Firm), we historically took the vast majority of our payments via cheque locally, and wire transfer internationally.
The issue with taking cheques is obvious...you have to send someone to pick them up! Additionally, these can often be hit with significant delays if the person responsible for signing off on the cheque is not available because they are on a business trip.
Accepting traditional wire transfers is also problematic, because these then need to be manually reconciled. Additionally, this means that you have to expose sensitive bank data publicity to third parties, and also allow access to employees to bank statements and sometimes the backend of various bank accounts.
We solved these various issues by using Stripe Billing, where we can issue invoices to customers, who then have two methods to pay:
Debit/Credit Card. We enable this for payments that are only a few thousand dollars, as there is a % fee applied and so it's not attractive to invoice tens of thousands via this payment method. However, this is absolutely perfect for smaller amounts, and there is the advantage that we can pull funds from a customer's card instantly once they have approved the payment.
Wire Transfers. This is where things become interesting, Stripe creates a unique virtual bank account on a per-customer basis, and then this allows Stripe to monitor incoming Wire Transfers and automatically reconcile any incoming payments with open invoices and mark them as paid. The cost for this service is $7 per invoice which is negligible, and saves us time.
Stripe then pays us out into our Mercury Bank Account (more on that later) at the end of each day, and it also reminds customers to pay prior to the invoice due date and also automatically follows up if the invoice is overdue.
We then turned our attention to banking, and what we wanted from our ideal bank. Transparency and future innovation was an important consideration when deciding on a bank, and so this meant that our bank would have an easy-to-use and flexible API that we could leverage to create dashboards for shareholders, management, and employees.
Mercury has some quite interesting features:
- Full Online Account Opening. The entire account opening was done online, without any need for an in-person visit or even a phone call.
- Self-Service Card Issuance. We're able to easily issue both virtual and real cards for staff expenses and to ensure spending limit controls are enabled.
- Beautiful User Interface. We benefit from a delightful user interface. Given the fact that we design user interface, this is actually an important added bonus!
- Extremely Low Fees. Mercury has almost no fees for any of the typical banking actions, and where there are fees these are on a cost basis. We can send money to anywhere in the world for $20.
The API is also quite interesting, as we can instantly issue read-only tokens to our developers who can then build dashboards, and we can even make programmatic payments if we wish to do so.
Another key pain points, as mentioned in the introduction, was the way we did accounting across multiple currencies, and also how we reconciled all of our bank statements, both for income and expenses.
We did not just want a new accounting software, but also a third-party accounting firm instead of an in-house accountant so that there would be more accountability to shareholders from senior management.
We found our ideal partner in a company called WaveApps, who have an extremely interesting offering. They have a SaaS (Software as a Service) accounting system that also has an Accountant (AaaS - Accountant as a Service?!) component as well.
Essentially, their software connects directly to Mercury to pull in real-time data of both revenues and any payments, and then WaveApps uses both machine learning as well as human oversight to categorize expenses.
For cash expenses, these are logged separately and we have a dedicated mobile app that we can use to upload receipts, and then it is up to the WaveApps team to actually categorize the expenses appropriately, instead of our in-house team doing so. This provides a significant amount of accountability and transparency in the decisions behind how we categorize individual expenses.
While the steps above took a significant amount of time to research and implement, the benefits both in the short term and long term are obvious. We reduce the amount of time spent on meta-work, which enables us to keep as much focus as possible on our client's projects.
We also gain real-time insights into our finances that are always 100% correct, which allows us to make better, more informed business decisions as we expand Mäd.
However, we are not finished yet. We have purely built the foundation that will then enable us to roll out further innovations across our entire back-office.
Key next steps include:
- Client Dashboards. Rolling out a client billing dashboard to enable clients to view their historical spending with us, download invoices, and generally manage their payment methods.
- Becoming an Open Startup. Creating a /open page on our website, that showcases key financial statistics that are publicly available. We believe that this level of transparency helps everyone understand the bigger picture.
- Create new products and services. Now that we have a significantly automated back-office, we can start to further innovate on the products and services that we offer. We already have Bloo, our project management software startup, and we will look to launch other services and products that customers can self-serve to themselves.
This journey towards a complete digital transformation has also had the additional benefit of being highly educational, as it's forced us to have a clear process for evaluating competing vendors, software solutions, and choices to reach optimal solutions.
If you're an existing client, or just an engaged reader on our website, we invite you to reach out and see how we can help you kickstart your own digital transformation to enable innovation at scale.