Skip to main content
iSample Report·Sage & Stone (demo data)
ADZETA

Prediction Opportunity Report

Sage & Stone

11,523customers analyzed
16 monthsof history
36%repeat purchase rate

+$86K

projected annual impact

Based on $45K/mo ad spend, optimized by predicted LTV

Moderate Opportunity

Positive correlation

High (7.9× range)

Your first-purchase data already captures a positive signal about future customer value, and there's a large 7.9× spread between your best and worst customers. Predictive modeling can refine what you already see — sharpening bid strategies to capture the remaining uplift from more precise targeting.

Customer Value Pattern

First-order value vs. 12-month LTV

Tap a point to see details

Customer
Trend Line

There's a visible upward trend, but significant scatter around it — first-purchase data captures part of the picture but misses a lot.

14 outlier customers beyond this range.

Value Divergence Over Time

How customer value gaps compound

The ratio between your best and worst customers over time. Changes reveal whether early identification becomes more or less critical.

30 days3.8x
90 days5.8x
180 days6.9x
365 days7.9x↑ widening
The value gap in dollars
Cumulative revenue per customer, top 20% vs bottom 20%
Top / Bottom 20% customersValue gap
The gap between your best and worst customers widens over time — from 3.8x to 7.9x. The compounding effect of repeat purchases means early identification is critical.

High-Value Customer Profile

What sets your best customers apart

Side-by-side comparison of your top 20% vs bottom 20% customers.

7.9× more valuable

Top 20% vs Bottom 20% by first order value

MetricTop 20%Bottom 20%Difference
Avg First Order$115.4$26.3+$89
Days to 2nd Purchase217251d
Repeat Rate75%20%+55pp
Avg Orders4.51.33.5x
Avg 12-Mo LTV$382.5$48.77.9x

Your top 20% shows distinct first-order patterns that pLTV modeling can identify in real-time to optimize bids toward high-value acquisition. Top category: Serums & Actives. Best source: Instagram Ads.

Discount dependency alert: 34% of first orders used a discount code, and those customers have 23% lower LTV than full-price buyers ($148.6 vs $192.4). Your discount strategy may be attracting customers who buy once and churn — exactly the pattern pLTV modeling can help you avoid.

Want to see this working on your ad spend?

Start a free 60-day experiment. We'll connect your data and test pLTV bidding on a conservative portion of spend.

Revenue Concentration

Revenue distribution by quintile

Customers grouped by first order value, showing revenue share per group

44%
25%
16%
11%
6%

Top 20%

$383/cust

21-40%

$219/cust

41-60%

$142/cust

61-80%

$92/cust

Bottom 20%

$49/cust

Revenue is moderately concentrated, with the top 20% generating 44% of total value. There's meaningful dispersion across customer groups.

Marketing Efficiency

What pLTV-optimized bidding could unlock

Adjust your monthly ad spend to see personalized projections.

$

Projected Annual Impact

+$86K

+16% ROAS improvement

Current Wasted Spend

$10,080

22% of monthly budget

Current Avg LTV

$174

Projected Avg LTV

$202

Recommended Experiment

Medium ($30K-$75K) · 25-30% of monthly spend

Test Budget/Mo

$12K

Duration

56 days

Expected Conv.

589

Detectable Lift

20%+

Free for qualified brands

Statistical basis: 95% confidence, 80% power, 8-week test duration

Projections based on 16% ROAS improvement from pLTV-optimized bidding. Actual results depend on market conditions and implementation.

Industry Benchmark

How you compare

Your brand vs. the industry average across the two dimensions that determine prediction opportunity.

Comparing against:
Select an industry above to see how you compare to benchmarks

Data Readiness

Data quality assessment

How well your data supports predictive LTV modeling.

6/6 checks passingStrong
History

16mo

≥12mo required

Customers

11.5K

≥1K required

12mo Cohort

7.5K

≥500 required

Repeat Rate

36%

≥30% ideal

Category

91%

≥80% coverage

Source

84%

≥80% coverage

Your data fully supports predictive modeling.

Recommended Path Forward

Based on your data analysis

Your data supports a pLTV experiment

Your customer patterns support a controlled experiment

Moderate Opportunity OpportunityStrong Data

What to Expect

  • Free 60-day experiment to validate improvement potential
  • Test pLTV bidding on a conservative portion of spend
  • Progressively enhance model accuracy over time

Next Steps

  1. 1Discovery call with your ad & data teams
  2. 2Connect AdZeta to your data sources & ad networks
  3. 3Launch free 60-day experiment on recommended budget

Schedule a call to explore how pLTV can improve your acquisition efficiency

Ready to optimize your LTV strategy?

Schedule a call to explore how pLTV can improve your acquisition efficiency. The 60-day experiment is free for qualified brands.

SOC 2 compliant · Your data stays yours