CAC vs. ROAS: which matters more, and where nCAC and LTV fit

If you run ads, you've been handed a ROAS number. It's the metric every ad platform leads with, and it's easy to understand: spend a dollar, get four back. The trouble is that a healthy ROAS can sit on top of a business that isn't growing.

This guide explains the four numbers that give you the full picture, CAC, nCAC, ROAS, and LTV, using one worked example throughout, and shows which to use for which decision.

The short answer: ROAS tells you how much revenue your ads brought in for every dollar spent. CAC tells you what it cost to win a customer. ROAS is useful for day-to-day decisions inside an ad account, but CAC, and especially new customer CAC, tells you whether your marketing is actually growing the business.

The four metrics at a glance

One example, all four numbers

Take an online store in a typical month:

• $20,000 spent on ads, plus $5,000 on agency fees and tools
• $80,000 in revenue attributed to those ads, from 1,000 orders, so an average order of $80
• 320 of those orders came from first-time customers. The other 680 came from people who had bought before.
• A 40% gross margin, meaning $32 of profit on an $80 order before marketing costs

We'll use these figures for each metric below.

What is ROAS?

Return on ad spend is the revenue your ads generate for every dollar you spend on them.

ROAS = revenue from ads ÷ ad spend

In the example, $80,000 ÷ $20,000 = a ROAS of 4, often written 4x or 4:1. That looks great, and it's the number most platforms will put front and center.

ROAS is the right tool for comparing campaigns, audiences, and creative inside an ad account, because it responds quickly and every platform reports it. It gets misleading when it's used to judge the business as a whole, for reasons we'll get to below.

What is CAC?

Customer acquisition cost is what you spend, in total, to win one new customer.

CAC = total sales and marketing costs ÷ new customers acquired

In the example, that's $25,000 in ads, fees, and tools, divided by 320 new customers, for a CAC of about $78. Include everything it takes to bring customers in: ad spend, agency or freelancer fees, software, and the share of your team's time spent on it. Leaving costs out is the most common way CAC ends up looking better than it is.

What is nCAC?

New customer acquisition cost, or nCAC, narrows the question to one thing: how much are your ads paying for each first-time customer?

nCAC = ad spend ÷ first-time customers

It exists because of a blind spot in platform reporting. Ad platforms usually report cost per purchase, which counts every order, including repeat customers who may well have bought anyway. In the example:

• Cost per purchase is $20,000 ÷ 1,000 orders = $20.
• nCAC is $20,000 ÷ 320 first-time customers = $62.50.

Same spend, same month, and the real cost of growth is more than three times what the ad account suggests. When most of your conversions come from existing customers, heavy retargeting, or searches for your own brand name, the gap gets wider.

Some teams calculate nCAC on ad spend alone and CAC on fully loaded costs, as we have here. Either approach works, as long as you calculate it the same way every month so the trend means something.

Why ROAS alone can mislead

• It measures revenue, not profit. A 4x ROAS on a product with a 40% margin means $1.60 of gross profit per dollar spent, which is healthy. On a 20% margin product, the same 4x means $0.80 back per dollar, which loses money. Your break-even ROAS is 1 ÷ your gross margin, so at 40% you need at least 2.5x just to cover the ads.
• It gives credit for sales you'd have made anyway. Retargeting, branded search, and ads shown to existing customers tend to report high ROAS because they reach people who were already likely to buy.
• Every platform claims the same sale. If someone clicks a Meta ad and later a Google ad, both may count the purchase. Add up each platform's reported revenue and you'll often get more than you actually sold.
• It says nothing about new customers. You can hold a steady ROAS for months while your customer base quietly stops growing.

ROAS tells you what your ads brought in. nCAC tells you whether you're growing.

Where LTV comes in

CAC and nCAC tell you what a customer costs. Lifetime value tells you what that customer is worth, which is what makes the cost meaningful. A $62.50 nCAC is expensive if customers buy once and never return, and cheap if they keep coming back for years.

LTV = average order value × orders per year × years as a customer × gross margin

In the example, if customers place 2.5 orders a year and stay for two years, LTV is $80 × 2.5 × 2 × 40% = $160 in gross profit.

Two ways to put LTV to work:

• LTV to CAC ratio. Divide LTV by nCAC: $160 ÷ $62.50 is about 2.6 to 1. A common rule of thumb is to aim for 3 to 1 or better, which leaves room for the costs that don't show up in either number. Much lower and growth costs more than it returns. Much higher and you may be underinvesting, leaving customers to competitors.
• Payback period. How long it takes a new customer's profit to cover what you paid to acquire them. Here, the first order brings in $32 of profit and the second brings the total to $64, just past the $62.50 nCAC. At 2.5 orders a year, that's roughly five months. The shorter the payback, the faster you can reinvest in growth.

LTV also changes what you can afford to pay. Turn the ratio around and a 3 to 1 target means a maximum nCAC of $160 ÷ 3, or about $53. The example store is paying $62.50, so either acquisition costs need to come down or LTV needs to go up.

Raising LTV is often the cheaper fix. Better email and SMS follow-up, subscriptions, a loyalty program, or simply a stronger second-purchase offer all make every new customer worth more, which makes the same nCAC more affordable.

Keep LTV honest. Base it on gross profit, not revenue, and on a time frame you can actually see in your data, usually 12 to 24 months. A five-year LTV built on assumptions will justify almost any acquisition cost.

Where ROI fits

Return on investment is the broadest view: did the marketing, all of it, make more money than it cost?

ROI = (gross profit from marketing − total marketing cost) ÷ total marketing cost

It's the number leadership usually wants, and it works well for looking back on a quarter or a campaign. Two things make it more useful:

• Count all the costs, the same way you do for CAC: ad spend, fees, software, and people.
• If you sell through leads rather than a checkout, give each lead a value. Multiply your average deal size by your close rate. With a $5,000 average deal and a 10% close rate, each qualified lead is worth about $500, which lets you calculate ROI and a target cost per lead long before the deals close.

Which metric should you manage by?

All of them, for different decisions:

• Day to day, inside the ad accounts: ROAS or cost per result, to compare campaigns, audiences, and creative.
• Month to month, for the business: nCAC and the number of new customers, measured against a target set from your LTV.
• Quarter to quarter: LTV to CAC, payback period, and ROI, to decide how hard you can afford to push growth.

If nCAC is too high, a few changes usually help:

• Report new and returning customers separately. Most ecommerce platforms can split first-time and returning orders. Put nCAC in the same report as ROAS so nobody looks at one without the other.
• Stop prospecting to people who already bought. Upload your customer list and exclude it from campaigns meant to find new customers. Google Ads and Meta both have settings that favor new customers or limit spend on existing ones.
• Keep retargeting in proportion. It's worth running, but if it's taking a growing share of the budget, ROAS will rise while new customer growth stalls.
• Improve what happens after the click. A better landing page or checkout lowers nCAC without touching the ads.

The bottom line: use ROAS to run the ad accounts and nCAC to run the business, and let LTV set how much a new customer is worth paying for. If you'd like help working out those numbers for your own business, we're happy to talk it through.