How to Find the Right Marketing Metrics for Your Business

You want to know if your marketing metrics are any good. So you Google “good ROAS for eCommerce” or “average CAC for SaaS.”

You find reasonable looking numbers and you aim for it. Then six months later, you’re hitting benchmarks. But the business still isn’t working. Or it’s working, but you’re missing the benchmark by a huge margin, even though your bank account looks fine.

What gives?

Metrics without business logic are dangerous. The logic comes from understanding your business model, not from copying someone else's targets.

A lot of folks pull industry benchmarks just to have a starting place. It’s a sensible enough thing to do, to be clear, but without context, they can be misleading. A 4:1 return on ad spending (ROAS) that's profitable for one company could be bankrupting another. A $200 customer acquisition cost (CAC) that's excellent for enterprise software could be a disaster for a $50/month subscription box.

In this post, I’ll walk you through how you can figure out which metrics you need to track. It all comes down to your research process, so I want to help you develop your own benchmarks through experimentation.

But first, I want to talk about a really fundamental question here…

There are three factors that tend to make reasonable metrics vary wildly, even within the same industry, from company to company. They are: gross margins, customer retention patterns, and sales cycle length.

As an example, I have a client that’s a third-party logistics company (3PL). They ship other companies’ products. But I can’t just pull trade journals to find out how much I should spend to win this 3PL one more client. That’s because some 3PLs work with massive enterprise businesses, and others work with 500-order Kickstarter campaigns. So what works for their competitors won’t necessarily work for them, and vice versa.

Every industry I’ve ever worked in has some nuances like this. I’ll break down what I mean by exploring these three variable factors—gross margins, customer retention patterns, and sales cycle length—below.

It’s not surprising that a SaaS company with 80% gross margins and an eCommerce company with 20% gross margins should aim for different ROAS numbers. If that SaaS company spends $1 to make $3, they’re fine. If that eCommerce company spends $1 to make $3, then they’re losing money on every sale.

But even looking for data on “ROAS for eCommerce” or even something as granular as “ROAS for online clothing retailers” can lead you astray if you’re not careful.

Break-even ROAS is calculated as 1 ÷ Gross Margin Percentage. If your gross margin is 25%, you need 4:1 ROAS just to break even. If it's 50%, you need 2:1. If it's 80%, you need only 1.25:1.

What I’m getting at is: you can use industry benchmarks as a starting point to figure out things like ROAS, CAC, and lifetime value (LTV). But if you don’t cross-check this data with your unit costs, then you simply do not have the data you need to make informed decisions.

Acquiring customers is great, but retaining them is a lot better. This is true whether you’re working in B2B or B2C.

There are a whole lot of companies out there that spend a lot of money to acquire clients, but it’s OK, because the clients tend to stick with them for a long time. Likewise, you’ll sometimes see eCommerce brands selling certain products at a loss, knowing that it’s a great way to get people into their ecosystem, and then upsell them later.

If you retain customers for a long time, then LTV will go way up. That means the amount that you can feel good about spending on CAC goes up accordingly too, as does the amount you can spend on advertising.

Cash is king. And that means that even if your customers are likely to keep paying you for a long time, if your bank account reaches $0 and you have to start taking out loans, that’s a real problem.

Let’s say you’re selling B2B enterprise software, and it takes 12 months to get payback on your CAC. That’s all well in good, but you can’t forget to add in the 3 months of sales cycle time too. It takes a while to get people onboard. That means you’ll be waiting 15 months to recoup cash, not 12.

Long sales cycles are entirely fine for companies with high contract values and low churn. But if you need customers to come in fast and pay fast in order to stay afloat, then you have to consider this alongside all your other metrics.

Before you ask "what should my metric be?", ask "what are my gross margins, retention rates, and sales cycles?"

Those answers determine everything else, but they don’t exclusively determine everything else. Because you also need to remember: you can’t treat a century-old Fortune 500, a VC-backed SaaS firm, and a mom-and-pop marketing agency the same.

Companies have different strategic priorities depending on their growth stage. And knowing which stage you’re in is going to have a huge impact on your strategic priorities. That, in turn, will have a huge impact on which metrics you need to use as your lodestars.

In my view, companies fall into three stages:

Let’s take them one by one.

Your priority here is to answer one simple question: “can you get customers to give you money for this thing?”

You need to be focused on revenue per sale, time from contact to cash, basic customer feedback, and gross profit per transaction.

Thinking in terms of LTV:CAC ratios is premature at this stage. You don’t have a lot of data and your early numbers are going to be all kinds of skewed.

To set reasonable benchmarks, track every transaction in a spreadsheet. Document what worked and what didn't. Compare this month to last month.

You are the benchmark. Talk to 5-10 customers. Test small—if an offer gets 10% response in 100 emails, scale it. If it gets 1%, don't.

For example, if you’re running a local service business, you can test an offer via direct outreach. After 50 contacts, 5 respond (10%), and 2 buy at $1,000 each. That's $2,000 for maybe 4 hours of work. That means you’re making $500/hour, which I’d say is pretty good! You’ll likely want to do more.

The point is proving the offer works and documenting what "working" looks like. The scientific method applies here—form a hypothesis, test it, document results, refine.

At this stage, your priority is to acquire customers consistently and profitably.

It’s here that you’ll need to focus on CAC by channel, payback period on CAC, and LTV (since you have enough customers now). It’s probably also a good idea to start tracking channel-specific ROAS.

Then once you do that, figure out what your breakeven ROAS is based on your gross margins. Set a target that’s going to consistently put you 20-30% above breakeven on the full cost to do whatever it is you do (whether making a product or delivering a service).

Or in plain English: make sure the channels you’re marketing on are making enough money to be worth it.

Once you figure out which of your channels are doing well, you can scale up spending or time spent on them, and you can even test new channels in small batches against your best performing ones.

If you were, for example, an eCommerce business owner and your margins were 25%, you’d need a ROAS of 4 to breakeven. So if Facebook is giving you 5.2 ROAS and Google is giving you 3.8 ROAS, you’ll want to spend more on Facebook and probably cut Google entirely if you can’t get it to improve.

Once you reach this stage, your priority is to improve margins, reduce waste, and optimize what’s working. And if your business is near-spotless as it is, then your job is to keep it that way.

It’s here that you’ll likely find yourself focusing on LTV:CAC ratio, customer retention and churn, operating margins, and efficiency of spend.

To set benchmarks for a business like this, look back at 12-24+ months of your data. Calculate your historical best performance by channel, campaign, and product. Set targets based on the top quartile of your own performance. Look at other companies or industry benchmarks to sanity-check your figures, but definitely give more weight to your own first-party data if you’re already profitable and otherwise happy with business performance.

As an example, if you had a SaaS company and you had marketing channels with LTV:CAC ratios ranging from 2:1 to 6:1, and your top quartile was 5:1, then use that as your benchmark. You want as many of your channels as possible reaching that benchmark.

It’s at this stage that you build benchmarks through experimentation, documentation, and tracking. What you find in trade journals or online is useful, and it’s good for broader context, but your own data is going to be a far better guide.

Individual metrics are interesting in their own right, but the real magic comes from how they connect with one another.

At 2x LTV:CAC, you have about 16% operating margin. That makes it kind of tough to reinvest in other areas of the business. But at 3x, operating margin jumps to 33% and gives you a whole lot more capital at your disposal to invest in growing the business.

This is the kind of thing that investors pay a lot of attention to. In fact, I read on a16z that bringing LTV:CAC from 2x to 3x can nearly triple valuation. Investors like having businesses who have capital available to, well, invest.

On the flipside, it’s easy to pat yourself on the back for a high ROAS. There’s many a startup out there that has thrown parties over an 8x ROAS even as they run out of cash. That’s because a ROAS of 8 can’t save you from tiny gross margins.

When you look at your CAC and sales cycle length together, that can give you a sense of CAC payback time. That tells you how long it will take to recover cash that you spent winning customers or clients. And that, in turn, tells you how long it will take before you can reinvest that cash.

If you have fast payback (such as, perhaps, 6-12 months) you might be able to self-fund growth. But if your payback is slow (like 18+ months), you might well find yourself needing an injection of capital to stay afloat in terms of cash.

As you look at your individual metrics, really stop and pause for a second and ask yourself: “what story is this telling about how we make money, and how we spend it?”

The numbers don’t exist for their own sake. They exist to answer that question.

First-party data grounded in your gross margins, customer retention patterns, and sales cycle length is the best. But I also recognize that telling you to look inward for data can sometimes feel like creating something out of thin air.

So here are six external sources of data that you can use to sanity check your metrics:

Again, the best use of this kind of data is to check if your own metrics are directionally reasonable. If people in your industry typically see a ROAS of 5.3, you probably shouldn’t be targeting 20—or 2.

Any benchmarks you pull from outside sources should be used to inform your internal discussions. But they shouldn’t be used to make decisions for you.

Googling "good ROAS for my industry" won’t get you what you need.

Far better is to start tracking your own data rigorously. Pay attention to what works for your situation and start building benchmarks from that reality. Pull in external context for validation, not direction.

Companies that win with metrics track fewer things but track them well. They build benchmarks from their own data. They test systematically. They update targets as they evolve. They focus on relationships between metrics, not isolated numbers.

More than good metrics, you need good process. And for that, there’s no beating experimentation, documentation, and honest analysis of what’s really going on in your business.