What Actually Drives LTV: A Framework for Repeat Purchase
Learn what drives LTV in performance marketing and how to build repeat purchase systems using post-purchase flows, cohort tracking, and retention strategy.
Most performance marketing conversations stop at the first conversion.
You optimized the creative. Dialed in the targeting. Hit your CPA target. The campaign is working.
But if that customer never buys again, you didn't build a business. You rented one transaction.
Lifetime value is the metric that separates brands that compound from brands that grind. And yet most performance marketing systems — including the ones agencies build for their clients — are almost entirely structured around acquisition. Retention is treated as a separate function, often owned by a different team, running different tools, with different goals.
That structural separation is one of the most expensive mistakes in eCommerce.
This is the framework I use at Impremis to close that gap and build repeat purchase behavior into the marketing system at every layer — creative strategy, channel architecture, the data models that drive budget decisions.
Image brief: Three columns (Frequency × AOV × Lifespan) with a multiplication symbol and a single LTV total. alt: "Three-driver LTV framework." caption: "LTV is three numbers multiplied. Move the right one for your category."
Why LTV improvements are more valuable than CAC reductions
There's a common instinct in performance marketing to focus relentlessly on lowering CAC. CAC optimization is real work that produces real results.
But the math on LTV improvement is often more compelling.
If a brand acquires 1,000 customers at $40 CAC and those customers have an average LTV of $120, the business generates $80,000 in net contribution from that cohort before overhead.
Lower CAC to $35 on the same cohort? You save $5,000. Meaningful, incremental.
Increase LTV from $120 to $160 through better retention and repeat purchase mechanics? Same cohort generates $120,000. A $40,000 improvement from the same acquisition investment.
LTV leverage compounds differently than CAC optimization. At scale, it can change the fundamental economics of a business more dramatically than any media efficiency gain. It's also the buffer that protects you from the CAC trap when you scale.
The three drivers of real LTV growth
LTV is not one metric. It's the product of three variables: purchase frequency, average order value, and customer lifespan. Most brands try to move all three at once and end up moving none of them meaningfully.
The cleaner approach: identify which lever has the most room to move for your specific customer base, and build the system around that lever first.
Driver 1: Purchase frequency
Purchase frequency is the most scalable LTV lever for consumable, replenishable, or habitual product categories — supplements, skincare, coffee, pet food, apparel basics.
If your average customer buys 1.8 times per year and your category benchmark is 3.2, you have a frequency gap worth investigating. The gap is almost always explained by one of three things: a weak post-purchase sequence, no active reason for the customer to return, or a product experience that didn't generate enough satisfaction to create repurchase intent.
Fixing the frequency gap requires a post-purchase system that maintains engagement between transactions. More on that below.
Driver 2: Average order value
AOV growth is a product and merchandising problem as much as a marketing problem. But marketing can influence it significantly through bundling, upsell sequencing, and threshold-based offers.
The most effective AOV tactic I've seen across high-volume accounts is the post-purchase upsell, presented immediately after the initial transaction before the customer has left the purchase mindset. A well-executed post-purchase upsell that converts at 15–20% on a $30 add-on adds meaningful revenue to every acquisition without adding any media spend.
Driver 3: Customer lifespan
Lifespan is primarily a function of churn. A customer who buys twice is not retained. A customer who buys five times over two years is retained.
Most brands focus on reactivating lapsed customers after they've already churned. The more efficient approach is to identify the behavioral signals that precede churn and intervene before it happens.
Customers who don't engage with any email in the 60-day window after a second purchase are significantly more likely to never return. Customers who leave a negative review within 30 days rarely come back. These are predictable signals that should trigger specific interventions, not generic discount blasts.
Building the repeat purchase system
A functional LTV system has four components that work together. Most brands have one or two of these. Brands with strong retention have all four operating in sync.
Component 1: Cohort-level LTV tracking
You cannot improve what you cannot see. The foundation of any LTV program is cohort tracking that lets you see how different acquisition cohorts perform over time, and identify which channels, campaigns, and creative types are bringing in customers with meaningfully different retention profiles.
This is one of the most underdeveloped analytics capabilities in mid-market eCommerce. Brands track blended LTV averages but rarely break it down by acquisition source, acquisition creative type, or product entry point.
That breakdown matters because not all customers are equally valuable. A customer acquired through a TikTok Shop creator review and one acquired through a Google Shopping ad may have very different retention profiles and purchase frequencies. If you're allocating acquisition budget purely on first-order CPA, you're likely over-investing in channels that bring in low-LTV customers and under-investing in channels that bring in high-LTV customers.
Component 2: The post-purchase sequence
The 72 hours after a customer's first purchase is the highest-leverage window in the entire customer relationship. Engagement rates are high, brand sentiment peaks, and the customer is open to information that reinforces their purchase decision.
Most brands waste this window with a generic order confirmation and a shipping notification.
A high-performing post-purchase sequence does four things:
- Confirms and reinforces the purchase decision immediately, reducing buyer's remorse and returns
- Delivers product education that increases the likelihood the customer uses the product correctly and gets results
- Introduces complementary products or use cases that expand the customer's relationship with the brand
- Creates a reason for the customer to engage with the brand's community, content, or social channels
This sequence runs through email and SMS and should be built before you scale acquisition spend. Every dollar you spend acquiring a customer that then gets no post-purchase nurture is a dollar with a structurally lower ROI than it could have. (More detail on the email side: the email framework.)
Component 3: Replenishment and reorder architecture
For brands with replenishable products, the reorder system is the most direct LTV driver available.
The mechanics are straightforward: identify the average consumption window for your product, and build a triggered reorder campaign that activates approximately two weeks before that window closes. The message is practical, not promotional. You're not running a sale — you're solving a problem the customer is about to have.
Subscription and auto-replenish models extend this logic further. The highest-LTV customers at most consumable DTC brands are on subscription. Getting a customer onto a subscription during the post-purchase window, when they're most open to the idea, is worth significantly more than any loyalty-points structure or discount ladder.
Component 4: Win-back campaigns with real segmentation
Most win-back campaigns are a single email with a discount code sent to every lapsed customer regardless of their purchase history, category, or churn signal.
That approach produces mediocre results because it treats all lapsed customers as the same problem.
A segmented win-back identifies at minimum three distinct lapsed groups:
- High-value lapsed: Customers who bought multiple times, had high AOV, went quiet. This group warrants a high-effort, high-investment win-back because their expected LTV if reactivated is significant.
- Single-purchase lapsed: Customers who bought once and never returned. Win-back economics are thinner. Messaging should focus on identifying what prevented a second purchase — using a direct question or a short survey — rather than leading with a discount.
- Category-specific lapsed: Customers who bought from one product category but never explored others. The win-back message is a category introduction, not a discount.
The CEO-level framing: LTV is a business model question
Agencies that present LTV improvement as a retention tactic are missing the larger point.
LTV is the metric that determines how much you can afford to spend to acquire a customer. A brand with a 90-day LTV of $80 has a fundamentally different CAC ceiling than a brand with a 90-day LTV of $200, even if they sell similar products at similar price points.
When you improve LTV, you expand the acquisition economics. You can outbid competitors on paid media while maintaining healthier margins. You can take risks on new channels and new creative approaches that wouldn't pencil out at lower LTV.
This is why LTV improvement should be positioned to clients not as a retention initiative but as an acquisition advantage. It's the work that makes everything else in the marketing system more efficient.
Build the retention infrastructure first. Then scale the acquisition spend on top of it.
FAQ
What's a healthy 90-day repurchase rate for first-time customers? Median DTC sits around 18–25%. Strong programs run 30–40%+. Below 15% means the post-purchase system is the bottleneck, not acquisition.
Should I track LTV by channel or by creative? Both. Channel tells you where to allocate budget. Creative tells you which messaging brings in higher-retention customers — and that should feed back into your creative testing system.
Is subscription always the right move for replenishable products? Almost always. Subscription customers consistently produce 2–3x the LTV of one-time buyers in the same category. Even a small subscription opt-in rate (10–15%) materially improves cohort economics.
How do I prioritize the four LTV components if I'm starting from zero? Post-purchase sequence first (highest impact, lowest build cost). Cohort tracking second. Replenishment third. Win-back last.
Closing
If your current system is primarily acquisition-focused, the highest-leverage first step is not a loyalty program or a win-back campaign.
It's a post-purchase sequence that actually works.
Audit your current post-purchase flow. Measure open rates, click rates, second-purchase conversion within 90 days. If your 90-day repurchase rate for first-time customers is below 20%, you have a retention infrastructure problem no amount of acquisition optimization will solve.
Fix the post-purchase system. Build the cohort tracking. Then scale.
LTV is not a metric you report on after the fact. It's a system you build before you scale.
Keep reading
Pieces I've written on related topics that pair well with this one:
- The 90-Day Cohort Analysis That Predicts Whether Your Paid Media Can Actually Scale — Most brands scale paid media using blended LTV averages that hide which channels produce customers worth keeping. Here's the 90-day cohort framework.
- How to Price Performance Marketing Services Without Destroying Your Margin — Performance marketing agencies underprice because they compare rates instead of building cost models.
- The Creative Brief Template I Use for Every Ad Campaign — A proven creative brief framework used at Impremis to improve ad performance, align teams, and scale winning campaigns across paid media channels.
- The Full-Funnel Media Plan: Awareness Pays the Conversion Layer — Learn how full-funnel media planning connects awareness spend to conversion performance, improving ROAS, lowering CPAs, and scaling eCommerce growth.
- What I Got Wrong About Hiring Media Buyers — Most media buyer hires fail because agencies optimize for platform skills instead of judgment, communication, and commercial thinking.