The Metrics that Matter for Growth Stage Startups in 2020
Early lessons on SaaS metrics
Early in my career in growth equity, I was (willingly) pulled into the rabbit hole of SaaS metrics.
As strange as it might sound, the change of pace was actually refreshing to me, as SaaS metrics felt relatively straightforward at the time. No complex DCF or LBO models and only a handful of inputs to worry about? As a former investment banker, I considered this a win.
I was also excited when it dawned on me that SaaS metrics don’t change much… as in almost never. David Skok published his widely-read SaaS Metrics 2.0 framework over a decade ago in 2008. That article remains highly relevant to this day.
As I developed a deeper understanding of SaaS metrics, I read many blogs and frameworks on the web. If you’ve done any Googling in this realm, you’ve probably reached the same conclusion I did: everything that could possibly be said already has been. So, once I had found my favorite articles and formed my perspective, I stopped pushing myself to learn more on the subject. In other words, I thought I had it covered.
In hindsight: boy, was I wrong.
Fast-forward a few years, I can tell you that SaaS metrics are anything but straightforward. There are important variances in the way that each company calculates and perceives them. Board members often disagree on which metrics are most important and provide mixed guidance to entrepreneurs. And even if you get the calculations right, knowing what to do with the data is a problem in its own right. Finally, business models are perpetually changing, and digitally-enabled platforms like Uber and Peloton have recently thrown into question how we should measure technology company financials altogether.
Today, I know that the devil is in the details when it comes to understanding SaaS metrics. There is no one-size-fits-all, and because of that, the learning never stops. In short, I’m still deep in the rabbit hole of learning, and that’s probably the way it should be.
Why are we writing this?
The purpose of this article is to shed some light on my learnings, as well as Georgian’s overall perspective on SaaS metrics. As an organization, Georgian is dedicated to supporting and adding value to entrepreneurs throughout their growth journey. With this article, we hope to provide insight into the way we think about measuring efficiency and predictability, how we assess company performance and how we make investment decisions.
We will also cover some less frequently discussed nuances, including where SaaS metrics fail in practice. Finally, we’ll see how SaaS metrics change (or not) in the context of software companies leveraging AI.
Note: We’ve shared links to some of the best SaaS metrics sources throughout this article and in the accompanying infographic because as mentioned, there’s a lot of great content out there. The metrics we deep dive into are the ones we deem the most important.
Measuring what counts
When companies hit the growth stage — usually around their Series A or B financing — investors expect a level of consistency in the company’s growth plans. The product, team and company vision are no less important, but the evaluation of metrics alongside those factors becomes critical. The addition of a standardized set of metrics allows investors to connect the numbers with the founders’ vision, as well as measure the quality of the company as it scales.
At Georgian, we begin this measurement by asking two important questions: the first is “is this business efficient?” and the second is “is this business predictable”?
Is this business efficient?
Some companies spend more money to acquire customers than others. The science behind measuring this is called sales efficiency. In layman’s terms, it measures a business’s ability to repay an additional dollar invested in sales and marketing.
Sales efficiency is critical because companies with longer, more costly sales cycles require more capital to grow. Raising additional capital dilutes existing investors and employees, limiting the economic upside. As a result, more efficient businesses are typically viewed as better vehicles for investment.
To assess sales efficiency, our metric of choice is Gross Margin CAC Payback. This metric calculates the number of months required to repay the initial cost of acquiring a customer accounting for the business’s gross margin. We pay close attention to Gross Margin CAC Payback because it accounts for the velocity of customers (revenue) being acquired, in addition to their profitability. Benchmarks for the metric range between 12 and 20 months, with longer paybacks being more common for enterprise deployments.
Other payback metrics such as the SaaS Magic Number or Payback Ratio measure sales efficiency based on the recurring revenue dollars being acquired. While we do certainly track these metrics, we place less weight on them because they suffer from a major drawback: they assume all revenue is equal, regardless of profit margin. In practice, we know this isn’t true, and companies with lower gross profit require more capital to grow.
Sales efficiency is a science in and of itself and goes far beyond the metrics mentioned above. Many companies find it useful to build a comprehensive sales dashboard in order to measure levers such as lead velocity, conversion ratios and quota attainment (each of which impacts sales efficiency). Ultimately, from an investor’s standpoint, what we want to see is a repeatable go-to-market playbook that remains efficient as the business scales.
Is this business predictable?
Predictability is a two-part question, broken down into retaining and monetizing customers.
We would argue that retaining customers is, in the vast majority of cases, the most important indicator of health in a software business. This is because churn management becomes exponentially more challenging to offset as companies scale. Backfilling for $20K in lost revenue is one thing, backfilling for $20M is another—and can be the difference between massive growth and failure. As a software business grows, churn becomes the determining factor for the maximum size a business can reach.
The three most commonly cited metrics to measure retention are Gross Logo Retention, Gross Dollar Retention and Net Dollar Retention. Of these, Net Dollar Retention is the one we focus on most and— in our view—the most important SaaS metric to track.
Net Dollar Retention answers the question, “at what rate would your business continue to grow (or shrink) if you were to rely only on sales from your existing customer base?”. It measures what percentage of revenue from current customers a business retains from the prior period, after accounting for upsell, downsell and churn.
Software companies targeting the enterprise should aim for Net Dollar Retention of greater than 100%, or at least in-line with the average for public company Net Dollar Retention at the time of IPO. In our experience, the best enterprise software companies have Net Dollar Retention in excess of 120%, with 110% being our benchmark.
Up For Renewal Analysis is another retention metric we rely on frequently—and one we would argue does not get enough time in the spotlight. It measures the number of customers that renew as a percentage of the number of customers that are up for renewal during the period (either in dollars or number of customers). The benefit of the calculation is that it strips out the noise of multi-year contracts and customers who are not “up for renewal” during that period. These factors can artificially inflate the true retention characteristics of a business, particularly in the early stages of growth. This calculation is covered extensively by the SaaS CFO in this article, and we strongly recommend beginning to track this metric as part of your KPI dashboard. For reference, good renewal retention varies from 85% to 92.5%, with anything above that considered best-in-class.
The second piece of predictability comes in the form of monetizing customers, commonly measured by the Lifetime Value (LTV) to Customer Acquisition Cost (CAC) calculations. These metrics test how much a customer is worth (in contribution margin) in comparison to what it costs to acquire them. An LTV/CAC of greater than 3x is considered good; however, earlier stage investors often target figures in excess of 4x or 5x. The crux of it is that 3x LTV/CAC allows a business, operating under normal software margins, to be operating profit break-even in year four. An interesting article here gives more details.
We sanity check LTV/CAC calculations in a variety of different ways. For example, we often normalize for shorter customer lifetimes (between 3 to 7 years), rather than relying on the lifetime as calculated. This allows us to test whether unit economics would still make sense if retention assumptions are overstated, as they can be early on in a company’s lifecycle. You can also switch between using Gross Logo Retention or Gross Dollar Retention to calculate your customer’s average lifetime. Our guidance is to explore which of the many LTV/CAC calculations makes the most sense for your business – then stick to your chosen methodology and be prepared to back it up.
It can’t be that easy.
If all that seems too basic, it’s because at the highest level, it is. The calculations themselves often aren’t the biggest challenge. More often, issues arise in one of two areas.
- Lack of data availability: For example, calculating customer-level revenue is often an issue, as is the inability to allocate sales and marketing expense properly.
- Leveraging the metrics: Even if you calculate the metrics properly, it’s difficult to know what exactly to do with the numbers (i.e., what tangible insights can I take from this analysis?).
Because of this, we wanted to share some common pitfalls we’ve seen in using SaaS metrics to their full potential. This guidance is based on what we’ve learned from our portfolio.
- Use a KPI dashboard: You should measure SaaS metrics in relation to one another. No one metric is useful in isolation. The ideal way to track SaaS metrics is to build a KPI dashboard and measure performance over multiple periods. Once you create a dashboard, it’s important to update it regularly and make the information available to key decision-makers. Two popular starting points are these ones by Redpoint Ventures (Tomasz Tunguz) and forentrepreneurs (David Skok).
- Benchmark against the best: Comparing your company performance against median and top quartile benchmarks can help you understand where you’re ahead or behind versus peers and adjust accordingly. Openview and Keybanc annually publish well-known SaaS benchmarks for private software companies.
- Build metrics into your financial forecasts: One of the biggest mistakes we see companies make is failing to understand what their forecast suggests in terms of SaaS metrics performance. Putting a plan in front of investors that suggests your Gross Margin CAC Payback will drop from 53 months to 5 months next quarter—without any significant changes to operations— can damage the credibility of your projections. Calculate your future SaaS metrics to tell the story of how your planned operational improvements will impact business performance.
- Align with the board: Every investor is different and will have an opinion on which metrics are most important. The key word here is opinion. Investors, including us, come with preconceived ideas of the importance of certain metrics based on their own experiences. Those lessons may not be as relevant to the company you are building. Have a view of which metrics matter most for your business and focus on executing on those metrics as you scale. Get your board aligned with your approach and you’ll be set up for success.
- Permeate metrics across the organization: Too often, the discussion on SaaS metrics begins and ends with senior management (or worse yet, the board). The best companies speak openly about company metrics and use them as fuel to team drive performance.
- Align compensation with desired outcomes: Peter Drucker famously said: “If you can’t measure it, you can’t improve it”.; The same goes for compensation. Once you’ve identified key SaaS metrics, set annual goals and base some portion of executive compensation on those metrics. (To be clear, actual goals are often based on higher-level figures such as ARR added, recognized revenue or EBITDA burn, but have trickle-down effects on SaaS metrics.)
So what about AI?
Georgian invests in companies that leverage disruptive technologies to gain a competitive advantage. Today, we focus on the intersection of applied AI, conversational interfaces and trust. Because of this, founders often ask how SaaS metrics change for AI companies.
The answer is more boring than we’d like to admit: not that much. At the end of the day, the basic laws of software economics continue to apply to AI companies.
AI can help shape business outcomes, similar to cloud, mobile and other transformational technologies before it. Like all technologies, AI needs to be aligned with and positively impact organizational goals and metrics. In other words, AI is another lever that can be used to accelerate growth and reduce costs across the organization, ultimately improving SaaS metrics over the long-term.
A good example of this in practice is aggregating cross-customer data sets to augment machine learning models. This can improve the speed at which you can draw inferences from customer data for new customers, solving what’s commonly known as the “cold-start problem”. This means you can onboard users to your products faster and reduce friction at the start of the customer journey. Shorter implementation cycles mean companies recognize revenue faster, which improves sales efficiency, LTV/CAC, and a host of other metrics.
In practice, there’s more complexity to it than that. The board and leadership team must be aligned in shared understanding that AI-based investments require capital and time to play out. R&D spend is often higher for companies pursuing AI, as they require time and investment to scale data science efforts. Sales cycles for AI products can sometimes be longer as well, due to the inherent complexities of accessing customer data and the trust issues involved.
Being an AI company also fosters questions about product development, such as effective model QA, ethical use of data, data moats and overall model accuracy. All of these factors have an impact on SaaS metrics over time.
Zetta Partners has some interesting thoughts on AI-specific metrics, much of which we agree with and would look for during our technical diligence.
Our recommendation to B2B companies leveraging AI is the same as any other company: understand your metrics in the context of traditional software metrics. Take SaaS metrics for what they are worth, but also understand your differences and use them to your advantage in scaling your platform. In the end, it’s AI’s long-term impact on operating performance that counts, and in our portfolio we have seen significant revenue multiples placed on companies with established data moats. This may not be captured in SaaS metrics over the short term, but certainly counts in the end.
There’s a lot of information out there on SaaS metrics and we don’t pretend to have all the answers. However, we do recommend keeping things simple. After all, SaaS metrics haven’t changed much over the last decade. And that’s because efficient and predictable businesses are not going out of fashion any time soon.
When it comes to measuring efficiency of your business we recommend tracking your Gross Margin CAC Payback. When it comes to predictability you can’t ignore key indicators such as Net Dollar Retention, Up for Renewal Analysis and LTV/CAC. These measures work together to provide remarkably good insight into the health of a software business.
And as for AI? While it changes many things about building and scaling software companies— in our experience—it doesn’t change the basic metrics for measuring great companies. We would love to hear your feedback and questions, you can reach me at email@example.com.
This piece was produced for the Georgian Growth Program, a new program built for leaders of high growth B2B SaaS companies. To learn more about the program, contact Evan Lewis (firstname.lastname@example.org).