How to Identify Your Top 20% Customers by LTV

Your top 20% of customers by lifetime spend often drive 60% to 80% of revenue, and you can find exactly which customers those are from a single Shopify order export. The point isn't the trivia that revenue is concentrated. It's what you do next: profile the shared traits of that top group, then point acquisition at finding more people who look like them at the moment they place a first order.

This guide shows how to find your own revenue-concentration cutoff with a cumulative-revenue table, how to profile the top cohort on four traits you already have, and how to feed that profile back into acquisition so you're cloning your best customers instead of guessing.

Why the 80/20 Rule Is a Starting Point, Not the Answer

The Pareto principle says roughly 80% of outcomes come from 20% of causes. In a real D2C store the split is rarely a clean 80/20. It might be 72/18, or 65/25, or in a subscription-heavy catalog closer to 85/15. The exact numbers are set by your repeat rate and how skewed your basket sizes are, not by a rule of thumb.

That matters because the cutoff is a business decision, not a constant. If 18% of your customers drive 70% of revenue, treating "the top 20%" as your VIP tier over-includes people who barely clear the line. Find the actual inflection point in your data first. Then decide where to draw the tier.

Build the Cumulative-Revenue Table, Step by Step

Everything here runs from one file: a Shopify orders export (Orders, Export, all orders, CSV) covering at least the last 12 months, ideally 24. The following steps take you from that raw export to a cutoff you can defend.

Step 1: Roll Orders Up to One Row Per Customer

The export is one row per order. You want one row per customer with their total spend. In a spreadsheet, make a pivot table with customer email as the row and sum of order total as the value. Use email, not name, because names collide and guest-checkout spellings drift.

Strip refunds and cancellations first, or you'll credit a customer for revenue you gave back. If your export has a Financial Status column, keep only paid and partially_refunded, and subtract the refunded amount where you can. Net revenue is the honest basis for this exercise.

Step 2: Sort Descending and Add a Cumulative Column

Sort the customer list by total spend, highest first. Add a running-total column that accumulates spend down the list, and a second column expressing that running total as a percentage of total revenue. Add a third column for each customer's rank as a percentage of all customers.

You now have the two numbers that define concentration: what share of customers you've counted so far, and what share of revenue they account for. Reading down the list, you're looking for where revenue share starts to flatten out.

Step 3: Find the Cutoff Where the Curve Bends

Here is a worked example from a 4,000-customer store doing 1.2M in trailing-twelve-month net revenue. Customers are ranked by spend and grouped into deciles.

Customer decile (by spend)	Customers	Revenue in band	Cumulative revenue	Cumulative % of revenue
Top 10%	400	$528,000	$528,000	44%
Top 20%	400	$240,000	$768,000	64%
Top 30%	400	$150,000	$918,000	76%
Top 40%	400	$108,000	$1,026,000	85%
Top 50%	400	$78,000	$1,104,000	92%
Bottom 50%	2,000	$96,000	$1,200,000	100%

Read the cumulative column. The top 20% (800 customers) drive 64% of revenue. The jump from the top 10% to the top 20% adds 20 points of revenue; the next decile adds only 12, then 9, then 7. The curve is bending hardest between the 20% and 30% marks. That bend is your cutoff. In this store, "top customers" is the top 20%, and the average one is worth $960 versus $135 for everyone below the line.

Your own numbers will differ. The method is the same: draw the line where each additional decile of customers stops adding much revenue.

Profile What the Top Cohort Shares

A list of your best customers is useful. Knowing what they had in common at their first order is what makes it actionable, because first-order traits are the only ones you can target in acquisition. Pull these four for the top cohort and compare each against the bottom half.

• First product bought. The single SKU that most often starts a high-value relationship. Frequently a specific bestseller or a bundle, not your cheapest tripwire product.
• Acquisition channel or first-order UTM. Which source delivered them. If your top cohort skews to email, referral, or one specific paid campaign, that's a spend signal.
• First-order AOV. Top customers often, though not always, start with a larger first basket. If yours do, a low free-shipping threshold may be selecting for low-value customers.
• Time to second order. How fast they came back. A short gap to the second purchase is one of the strongest early predictors of high lifetime value.

In the example store, the top cohort's shared profile was clear: 61% started on one of two "starter kit" bundles, 44% came from email or referral rather than cold paid, first-order AOV was $74 versus $41 for everyone else, and 55% placed a second order within 45 days. That's a customer you can describe, and describe means you can target.

Feed the Profile Back Into Acquisition

The payoff is turning the profile into targeting decisions. Cloning your best customers means biasing acquisition toward the traits they shared, so you're buying more of the people who become the top 20% and fewer who never clear the line.

Do four concrete things with the profile. First, build a lookalike or advantage audience from your top-cohort customer list, not from all purchasers, so the model learns from your best customers rather than your average. Second, lead acquisition creative and landing pages with the first product that starts high-value relationships, instead of your cheapest entry SKU. Third, shift budget toward the channels that over-index in the top cohort and pressure-test the ones that only deliver bottom-half customers. Fourth, if top customers start with a larger basket, set your free-shipping threshold and bundle offers to nudge first-order AOV upward rather than down.

Then measure whether it worked by cohort. Tag new customers by acquisition month and watch whether the newer cohorts concentrate revenue faster than older ones did. If your clone-them changes are working, a larger share of each new cohort should cross into top-customer territory over time. For the underlying metric, see the lifetime value glossary page, and to keep acquisition economics honest, read what is a good LTV to CAC ratio.

Frequently Asked Questions

How do I find my top 20 percent of customers?

Export your Shopify orders, roll them up to one row per customer by summing net order value, then sort descending. Add a cumulative-revenue column and a cumulative-customer-percentage column. Read down until you find where each added decile of customers stops adding much revenue. The customers above that bend are your top cohort, whether that lands at 15, 20, or 25 percent.

Is the 80/20 rule accurate for ecommerce revenue?

It's directionally right but rarely exact. Most D2C stores see something between 65/25 and 85/15 rather than a clean 80/20, set by repeat rate and how skewed basket sizes are. Subscription-heavy catalogs concentrate harder; one-time-purchase catalogs concentrate less. Build your own cumulative-revenue curve instead of assuming 80/20, because the real cutoff drives who you treat as a VIP.

What traits should I look at for high-value customers?

Focus on traits present at the first order, since those are the only ones you can target in acquisition. The four most useful are first product bought, acquisition channel or first-order UTM, first-order AOV, and time to second order. Compare each against your bottom half. A short gap to the second purchase and a specific starter product are usually the strongest predictors of high lifetime value.

How do I use my best customers to improve acquisition?

Build lookalike or advantage audiences from your top-cohort customer list rather than all purchasers, so the model learns from your best customers. Lead creative with the first product that starts high-value relationships, shift budget toward the channels that over-index in the top cohort, and set free-shipping thresholds to nudge first-order AOV up. Then track whether newer acquisition cohorts concentrate revenue faster.

Should I remove dormant customers from the top cohort?

Remove them from the clone-them profile, not from the analysis. A customer can rank in your top 20 percent on lifetime spend while having gone quiet months ago. Add a last-order-date column and flag anyone past your typical repurchase window. Those customers belong in a win-back flow. Building acquisition lookalikes from lapsed high-spenders teaches the model to find people who also churn.

Conclusion

Your best customers are already in your order export. Find the concentration cutoff with a cumulative-revenue table, profile what the top cohort shared at their first order, then aim acquisition at cloning those traits. Do it as a repeatable quarterly pass, not a one-off, because the profile shifts as your catalog and channel mix change.

To go further on segmentation, see the customer segmentation guide and the RFM segmentation on Shopify walkthrough, which turns this same export into recency-frequency-monetary tiers.