Earlier this year, Front co-founder & CEO Mathilde Collin announced the company’s $66m Series B, and, two months later, shared its Series B deck publicly. This followed the company’s $10m Series A in early 2016, with Collin similarly publishing its Series A deck and some learnings on the process.
When reading through the Series A deck and commentary, one item in particular stood out:
“Slides that were really weak:
Projections: you should have a model ready and explain how you’re going to spend the money you raise — I did not have one.”
Flux decided to use this opportunity to build Front’s financial model, specifically through the lens of its Series A financing. Using publicly available information (namely, the two decks, a tweet, and the company’s team and jobs pages), Flux reverse-engineered Front’s operating metrics and created an income statement and a cash flow statement for the company from mid-2014 through year-end 2017. From this analysis, we believe that Front initiated its Series B fundraising conversations near the end of 2017 after reaching $10m in ARR (more on that later).
Note that this model does not contain official data from the company. In fact, this model is missing key information that would better inform the financial story. That said, the structure exists and could easily incorporate said inputs.
The first step to this analysis is understanding Front’s historical financials. In order to accomplish this, we take a close look at slide 8 in the Series A deck: Front’s MRR and customer count from June 2014 through March 2016. Because each data point is not individually labeled, we are forced to make some eyeball assumptions from the graph, which produces this output:
Let’s do a quick sanity check: the $1.4m ARR in March 2016 matches the same figure Collin references in slide 14, so we feel comfortable in our estimates.
Modeling revenue projections from this point would involve looking at the company’s three pricing tiers (Basic, Premium, Enterprise) and making some assumptions around customer growth, churn, seats, and revenue per customer using previous cohort data coupled with the company’s sales & marketing plan. In this particular analysis, however, we decided to not make those assumptions. Instead, because this analysis comes in hindsight, we have the benefit of some company data that helps us reasonably estimate Front’s MRR for the subsequent 20 months following it Series A.
That data surfaced last month, when Collin tweeted the following:
Using this data, we are equipped with enough information to estimate Front’s MRR. Let’s use the first data point as example: it took Front 608 days to reach $1m in ARR. Referencing our previous chart generated from Front’s deck, we believe this milestone was reached in September 2015 (i.e., $1m in ARR implies $83k in MRR – $1,000,000/12). If we backtrack 608 days from this point, it corresponds to January 2014, two months after Front was co-founded (October 2013, as per Collin’s LinkedIn). We’ve now done a second sanity check on the analysis.
So, if we believe Front reached $1m ARR in September 2015, then Collin’s tweet suggests it hit the $2m ARR milestone 231 days later, implying that in May 2016, the company was at a $167k MRR (again, $2,000,000/12). We use the same method for the subsequent ARR milestones, and assume “smooth” revenue growth in intermediate months to “fill in the blanks”, producing this result:
Let’s pause for a moment and take a closer look at what these numbers suggest. Collin notes that she kicked off the Series A fundraise after Front “had just reached $100k in MRR” (thus, having reached over $1.2m in ARR). Similarly, Collin began the Series B fundraise after “having reached the milestone [we] had set for the previous round.” This tells us that Front began the Series B fundraise after reaching $10m in ARR. We confirm this by reviewing slide 18 in the Series A deck which states the milestone explicitly and is also supported by the data we produced above.
Also worth noting is MRR per customer – a metric we calculated using the aforementioned MRR and customer count estimates – which we will use as a proxy for average selling price (ASP). At the time of Front’s Series A, MRR per customer ranged between $130 and $150. By the time the company raised its Series B, this grew to ~$362. Alluding to the Series B, Collin noted that “ASP is growing and larger teams are using Front”, suggesting that the growth in ASP was driven by an increase in number of seats per customer (i.e., selling into companies with more users). Collin later states that “[t]he average selling price (ASP) is still low”, which suggests that in order to improve this metric, the company will continue targeting larger organizations (i.e., increase quantity) and/or increase the average revenue per customer (i.e., increase price). The latter emphasizes the importance of the efficacy of Front’s “land and expand” strategy with customers, and is reinforced by the company’s transition from a “shared email” client to a “collaborative communication” platform.
Now that we have “modeled” revenue (the word “modeled” being used liberally here as we backed out Front’s revenue numbers using historical data vs. a bottoms-up build), let’s model Front’s expenses to illustrate what its net income, cash position, and financial story would roughly explain.
We gathered employee information from Front’s team page post-Series A and cross-referenced it with LinkedIn data to determine employee tenure, assuming $100k in annual salary per team member (a simple, modular input). Similarly, we noted the roles the company was hiring for following its Series A and assume those jobs were filled 1-2 months after its raise. This exercise helps us size the majority of Front’s expenses: its employees.
We also make some high-level assumptions around infrastructure costs. Let’s assume that Front was utilizing one or more cloud infrastructure services (AWS/Azure/Google Cloud/Digital Ocean), and also leveraging free credits for the first N months of operations. Thus, this model assumes Front only started incurring infrastructure bills in May 2015. Ideally we would be privy to some details on the company’s other operating expenses (e.g., sales, marketing, software, rent, legal), but in lieu of this, we add placeholders for these expenses. This generates a hypothetical income statement for the company, featuring historical data (June 2014 through March 2016) as well as forward-looking projections (April 2016 through December 2017).
The cash flow statement also includes embedded assumptions. In cash flow from operating activities, we dismiss any notion of working capital, assuming Front collects all its revenue for the period in the period in which it occurred (in this example, the cadence is monthly). Why is this important? Front offers a 17% discount to customers that subscribe to its annual plan and pay upfront for the year. This upfront annual payment essentially funds Front’s working capital. That is, Front is not allowed to recognize this one payment as revenue in the month it was collected, but rather, it must spread it out over the months it delivers the service to customers (decided by the lifetime of the contract). Accordingly, it recognizes the fractional value as revenue for the month in which the service was delivered and creates a liability on its balance sheet called “deferred revenue” to account for the remainder. Because an upfront payment is literally upfront cash for the company, it creates a favorable cash position. Though Front very likely employs this method to run a negative working capital balance, we ignore its potential impact in this model.
We also dismiss any cash flow from investing activities (notably, capital expenditures), and assume the only cash flow from financing activities is comprised of the company’s Seed, Series A, and Series B financings. Front raised $3.1m in Seed funding; we assume this was accomplished through a mix of rolling-close SAFEs, starting from founding and continuing through September 2014. For simplicity, let’s assume that prior to June 2014, the company had raised $1m of SAFEs and had spent $300k, thus leaving $700k in cash on the balance sheet. For the remaining $2.1m, let’s assume it was funded in September 2014. We include the $10m Series A in May 2016, and if this model included January 2018, we would have included the $66m Series B.
Under this particular model and its assumptions, Front would have reached profitability and positive cash flow in April 2017. Given the company had a net cash burn of $4.8m from its Series A to its Series B (see Front’s capital efficiency), it is unlikely this played out. First, the employee salary placeholder we assumed ($100k annually) is almost certainly understated, suggesting that profitability was further off than we projected here. Secondly, and more importantly, Front has likely increased focus on growth in lieu of profitability, a logical strategy given its business model (SaaS) and its understanding of its cohorts' behavior across lifetime value, cost of customer acquisition, and payback period. Updating these assumptions based on the data underpinning Front’s strategy would help inform when the company expects to achieve these milestones, though it is fair to believe that the company is still hyper-focused on growth as it broadens its product offering and tries to expand its wedge in the market.
While we certainly benefitted from hindsight in this analysis, the intuition and exercise are similar to what an early-stage company will need to work through – whether explicitly or implicitly – as it thinks about its path to fundraising, growth, and scale. The ability to deconstruct a business into its core, granular inputs and thoughtfully assess the levers of growth & profitability is a powerful tool. We believe this enables founders and CEOs specifically to be more data-driven, purposeful, and prudent across capitalization (i.e., raising money) and asset allocation (i.e., spending money).
To be clear, we believe Collin has done a marvelous job at this (again, see Front’s capital efficiency) and we suspect she performed some version of this analysis during or after the Series A. More broadly, our goal in this analysis was to illustrate what the financial planning & analysis process would entail, and hopefully this paints a more concrete picture. If there are any questions, feel free to reach out.