A typical talk with the owner of a coffee shop, salon or car workshop goes like this: “A person from the ad costs 600 UAH, and the first receipt brought 400 UAH of profit. That is a loss, so we switch it off.” The decision is understandable but often wrong. A business people come back to earns not on the first visit but on all the following ones. If you look only at the first purchase, you can switch off something that works.
First, the limits of this text. The numbers in the examples below are invented, just to show the arithmetic. How many times a year your customer buys, only you know, so the calculation has to be done on your own data. About the Google Ads account itself we write only what we read in Google’s help, and we say plainly what is not there.
Why the first purchase misleads
Take an example for illustration. A new person came from the ad, and it cost you 600 UAH. The first receipt left 400 UAH of profit, so after it you are 200 UAH down. If you stop the calculation there, the conclusion is single: paying for such customers does not pay.
But someone who liked it comes back. Suppose over a year they came five more times and left 400 UAH of profit each time. Six purchases in total, which is 2,400 UAH from one person. The 600 UAH expense looks completely different: for every hryvnia spent, four came back.
This picture is typical for coffee shops, salons, workshops, pet shops and everyone whose customers buy repeatedly. Where a person buys once and disappears, the logic is different, more on that below.
What LTV is and how to count it
The abbreviation LTV stands for “lifetime value of a customer”, and in plain words it is the total profit one person brings for as long as they buy from you. The word “lifetime” misleads: you do not need to count “forever” and cannot, because nobody knows how long they will keep coming. It is more practical to take a limited period, six months or a year.
The formula is simple:
- profit from one purchase × how many times a year the average buyer purchases.
Profit, not revenue. If a receipt in a salon is 1,000 UAH, and after materials, the stylist’s pay and rent 400 UAH remains, 400 goes into the formula. Otherwise the result will be pretty but untrue.
One more nuance: “how many times they buy” is the average across everyone, including those who did not return. If you count only regulars, the number comes out inflated.
How to count on your own data
You do not need an analyst, a list of buyers with dates is enough. It exists in your CRM, your till software, your booking system or even a spreadsheet.
- Pick a month six months ago and list everyone who came to you for the first time then.
- Check how many of them bought a second time, a third time and so on during those six months.
- Add up the profit from this group and divide by the number of people. That is your six-month figure.
- Compare it with the price of one new person from advertising: divide that month’s ad spend by the number of newcomers.
If the group’s money has already covered the spending, the ad paid off, even though on day one it is not visible. If not, see whether the group will pay back within a year.
That leaves the question of who counts as new. In your spreadsheet it is a person who was not there before. It helps when the record has a phone number or an email: they show it is the same person.
Why you should count by newcomers
Showing an ad to people who come to you anyway adds nothing: they would have bought without it. Such an ad merely takes credit for purchases that would have happened anyway. So counting across all purchases in a row gives a nicer number than the real one. It is more honest to separate the new from those who have already bought.
Google Ads has customer lifecycle goals for this. Here is what Google’s help says, which we opened and read.
- The customer acquisition group has three settings: “New Customer Value”, “High Value New Customer” and “New Customer Only”. The first assigns higher bids to new customers than to existing ones. The last assigns bids only for new customers.
- They do not work in all campaigns. “New Customer Value” and “New Customer Only” are available for Search, Performance Max, Shopping and Demand Gen campaigns. “High Value New Customer” does not include Demand Gen.
- “New Customer Value” and “High Value New Customer” require at least one Purchase conversion goal and a value-based bidding strategy. For “New Customer Only” the Purchase goal is optional, but Google recommends it.
- Google suggests “New Customer Only” when the acquisition budget is limited or when conversions are not tied to purchases, for example when they are leads. Existing customers then need a separate campaign.
- Who counts as “existing” in your case is for you to define: the help advises relying on your own data, such as a customer list and the website tag.
The page on activating these goals adds technical conditions: customer lists must contain at least 100 active members, and automatic detection takes purchases over the last 540 days into account, so for accuracy it is worth supplementing it with email lists. We did not find a step-by-step menu path on these pages, so we do not invent one: the settings are switched on in the campaign following Google’s instructions, and if that is beyond you, it is a job for a contractor.
What Google’s help does not say
The words “LTV” or “long-term value” are not on that page. Google speaks about “new customer value” and “high value” customers, and we do not attribute more to it. The help also gives neither a time window for counting nor a payback threshold. So “six months or a year” in this text is our practical advice, not a norm from Google.
Where the limit is
LTV does not justify any price for a person. There are two cases where you need to count differently.
They buy rarely. Windows, renovation, a kitchen, a garage: a person returns once in several years or never. There is no “annual value” here, count one purchase. The acquisition cost should be clearly lower than the profit from it.
They buy often. Here the price of a new buyer should be lower than the profit they will give in six months or a year, and not by a hair. From our practice, a margin is needed for two reasons. First, repeat purchases are not guaranteed, and even an honest average may not repeat next year. Second, money moves in the opposite order: you pay for the ad today, and collect the profit over a year. If there is no money for that year, payback on paper will not help.
What to do now
Write down three numbers: profit from one purchase, the average number of purchases per year, and the price of a new person from advertising. Multiply the first two and compare with the third. If the result is higher with a margin, the ad works. If lower or almost equal, look for the weak spot: acquisition price, repeat purchases or profit per receipt.
Summary
“Ads do not pay off” often means “we counted only the first purchase”. For a business with repeat purchases, the right measure is profit from a new customer over six months or a year. Count newcomers, otherwise the result comes out inflated, and those who sell once in several years need only one purchase.
If someone has already bought from you and the ad still chases them, we wrote about it in the article “They Bought From You, but the Ad Keeps Chasing Them”. Why an ad account and a CRM show different lead counts, and which gap is normal, is in our article on that mismatch. How we run campaigns is in the sections on search advertising and Performance Max.
Not sure what a new customer costs you? Want a free audit of your Google Ads account? Send us a request — we will look at your account and tell you what is going on with the numbers.