The 4 Numbers That Matter on a VSL Launch (And Why Play Rate Isn't One)
Every VSL dashboard I've ever seen leads with play rate. It's right there at the top - big number, green arrow, looks great in screenshots. Last month a client showed me a 91% play rate and asked why the VSL wasn't converting. The answer was in the retention curve: 55% of viewers dropped off before the 30-second mark. The hook was failing. The 91% play rate just meant the page design was fine - the thumbnail and auto-play settings were doing their job. The video itself was hemorrhaging viewers before it got to the pitch.
After watching this pattern repeat across dozens of VSL launches, I've narrowed the numbers that actually predict profitability down to four. None of them is play rate.
The four metrics that predict VSL profitability are: 30-second retention rate (does the hook hold?), midpoint retention (does the body sustain interest?), revenue per viewer (how much does each viewer generate?), and cost per qualified view (how much are you paying for viewers who actually reach the CTA?). Play rate tells you about your page, not your video.
Why is play rate a vanity metric for VSLs?
Play rate measures the percentage of page visitors who start watching the video. It's a function of page design, thumbnail quality, and auto-play settings - not the VSL's ability to sell. A 95% play rate with 0.3% conversion tells you the page works and the video doesn't. A 60% play rate with 2.1% conversion tells you the page needs work but the video is a money machine.
Play rate is the metric that gets the most attention because it's the easiest to measure and the most flattering to report. It's also the least predictive of revenue. Here's why:
- Auto-play inflates it to meaninglessness. If your player auto-plays on load (which most do), play rate approaches 100% by definition. It's measuring page loads, not engagement.
- It doesn't differentiate between a 2-second view and a 40-minute view. Someone who bounces after 3 seconds and someone who watches to the CTA both count as "played." Those are not the same viewer.
- It's a page-level metric masquerading as a video metric. Play rate is determined by above-the-fold layout, load speed, thumbnail, and player visibility. None of those tell you anything about whether your script, offer, or presentation is working.
Play rate is the most-cited and least-useful VSL metric: a 90% play rate on traffic that converts at 0.4% means you have a hook problem, not a play problem.
Stop putting it in your dashboard's hero slot. Here are the four numbers that actually deserve that space.
30-Second Retention Rate
What it tells you: Whether your hook is working. The first 30 seconds of a VSL are a filter - they decide who stays and who bounces. Everything downstream depends on this number.
Benchmark: 60-75% for paid traffic to a 30-45 minute VSL. Top decile hits 80%+. Below 55% and your hook needs immediate work.
Why does 30-second retention matter more than any other metric?
Because it's the strongest leading indicator of conversion. Viewers who survive the first 30 seconds are 4-6x more likely to reach the CTA than those who don't. The hook sets the ceiling for everything downstream - no amount of great copywriting at minute 20 helps if 45% of viewers left before the half-minute mark.
The 30-second mark is where the viewer's brain makes a commitment decision. Before 30 seconds, they're evaluating: "Is this worth my time?" After 30 seconds, they've invested enough attention that the cost of leaving increases. It's the psychological equivalent of sitting down in a restaurant versus walking past it.
What a healthy 30-second retention curve looks like:
| 30-second retention | Diagnosis | What to do |
|---|---|---|
| 80%+ | Hook is excellent. Don't touch it. | Optimize downstream - focus on midpoint retention and CTA. |
| 65-80% | Hook is solid. Normal range for paid cold traffic. | Test hook variations but prioritize other levers first. |
| 55-65% | Hook is underperforming. Losing a third of viewers before the pitch starts. | Rewrite the first 30 seconds. Test 3-4 hook structures. |
| Below 55% | Hook is broken. Nothing else matters until this is fixed. | Stop traffic. Rebuild the hook from scratch. Re-launch. |
If your VSL retention curve drops 40% in the first 60 seconds, your offer isn't the problem - your hook is.
Midpoint Retention
What it tells you: Whether the body of your VSL holds attention through the teaching/proof section. The midpoint is where most VSLs transition from "here's your problem" to "here's the solution." If you're losing viewers here, the story isn't compelling enough.
Benchmark: 25-40% for a 30-45 minute VSL on paid cold traffic. Top performers hit 45%+. Below 20% and the body needs structural work.
What does midpoint retention tell you that the hook doesn't?
Midpoint retention separates hook quality from content quality. A VSL with 75% at 30 seconds and 15% at the midpoint has a great hook but boring content. A VSL with 60% at 30 seconds and 35% at midpoint has a decent hook and excellent content. The second VSL will almost always convert better, because the viewers who reach the CTA are deeply engaged.
The retention curve between the hook and midpoint is the most diagnostic section of the entire video. Three patterns to watch for:
- Gradual decline (1-2% per minute). Healthy. This is what a well-paced VSL looks like. Viewers are naturally filtering out, but the engaged ones stay.
- Cliff at minute 5-8. The hook-to-body transition is failing. The hook promised something the body isn't delivering - either the tone shifts, the pacing slows down, or the content gets generic.
- Plateau then cliff at midpoint. The "problem" section worked but the "solution reveal" disappointed. Viewers stuck around for the diagnosis but left when the prescription felt weak or salesy.
Each pattern has a different fix. That's why midpoint retention matters independently of the hook metric - it tells you where the content problem is, not just that one exists.
Revenue Per Viewer (RPV)
What it tells you: The economic value of each person who watches your VSL. This is the single most important number for deciding whether to scale traffic. If RPV exceeds cost per viewer, you're profitable. If it doesn't, more traffic just means more losses.
Example: A $997 product with 1.5% conversion rate = $14.96 RPV. If your cost per viewer is $8, you're making $6.96 profit per viewer. Scale aggressively.
How do I calculate revenue per viewer correctly?
Divide total revenue attributed to the VSL by unique viewers (not plays, not page views). Use your payment processor (Stripe, PayPal) as the revenue source - not the ad platform, which is undercounting. Use your VSL player's unique viewer count, not your ad platform's click count, because not every click becomes a view.
The formula is simple. Getting the inputs right is the hard part:
Revenue source: Stripe or your payment processor. Not Meta's "conversion value," which is suppressed by the pixel undercount problem. You need the real number.
Viewer count: Unique viewers from your VSL player, not clicks from the ad platform. A viewer who watches twice is one viewer, not two. A click that bounced before the video loaded is not a viewer.
Time window: Match the attribution window to your buying cycle. For a $47 product, 24 hours is enough. For a $2,000 product, use 7-14 days - some buyers come back after watching the VSL twice.
Revenue per viewer is the number that tells you whether to scale or fix. Everything else is diagnostic. This one is economic.
Most VSL marketers track conversion rate instead of RPV. Conversion rate is useful but incomplete - it doesn't account for average order value, upsells, or downsells. RPV captures the full economic picture in one number.
Cost Per Qualified View (CPQV)
What it tells you: How much you're paying for viewers who actually see your offer. Not viewers who bounced at 10 seconds. Not page visitors. Viewers who made it to the point where they could say yes.
Why it matters: CPQV connects your ad spend to the viewers who can actually convert. If RPV is $15 and CPQV is $20, you're paying $20 to generate $15 - unprofitable. If CPQV is $8, you're profitable at scale.
Why should I track cost per qualified view instead of cost per click?
Cost per click measures ad platform efficiency. Cost per qualified view measures funnel efficiency. A $2 CPC means nothing if 70% of viewers bounce in the first 15 seconds - you're paying $2 for traffic that never sees your offer. CPQV filters for the viewers who actually reached the CTA, giving you the true cost of each potential buyer.
The math makes this clear:
| Scenario | Ad spend | Clicks | CPC | Viewers reaching CTA | CPQV |
|---|---|---|---|---|---|
| Campaign A | $5,000 | 2,500 | $2.00 | 625 (25%) | $8.00 |
| Campaign B | $5,000 | 1,667 | $3.00 | 833 (50%) | $6.00 |
Campaign A has the lower CPC. Campaign B has the lower CPQV. Campaign B is delivering qualified viewers at 25% less cost, despite the higher CPC. If you optimize on CPC alone, you'd scale the wrong campaign.
CPQV is the bridge between your ad spend and your RPV. When CPQV < RPV, you're profitable. When CPQV > RPV, you're not. That single comparison tells you more than any dashboard full of vanity metrics.
How these four metrics work together
The four metrics form a diagnostic chain:
- 30-second retention tells you if the hook is working.
- Midpoint retention tells you if the content holds.
- Revenue per viewer tells you the economic value of each viewer.
- Cost per qualified view tells you the cost of each potential buyer.
When something breaks, you diagnose in order. Low RPV? Check midpoint retention - if viewers aren't reaching the offer, no amount of offer optimization helps. High CPQV? Check 30-second retention - if the hook is leaking, you're paying for views that never become qualified. The chain always starts at the top.
The profitability equation:
If RPV > CPQV, scale traffic. Every additional qualified viewer is profit.
If RPV < CPQV, don't scale - diagnose. Work backward from CPQV: is the ad targeting wrong (too expensive per click), or is the VSL leaking (too few clicks become qualified views)?
What I'd do if I were launching a VSL next week
Set up tracking for these four numbers before you buy a single click. Not after launch. Not after the first $5,000. Before. Here's the specific setup:
- Configure retention tracking at 30 seconds and at the midpoint timestamp. Your VSL player should give you retention curves with specific timestamp milestones. If it doesn't, you're flying blind.
- Set up a CTA-reached event. Fire a custom event when the viewer reaches the timestamp where your CTA first appears. This is your "qualified view" marker.
- Pull revenue from Stripe, not from your pixel. The pixel is undercounting by 30-50%. Use Stripe (or your payment processor) as the source of truth.
- Build a weekly RPV vs CPQV comparison. One number versus one number. Profitable or not. That's the whole report.
The hardest part isn't the math. It's the discipline to ignore the vanity metrics that every dashboard puts front and center. Play rate, total views, average watch time - they all feel important. They're not. The four numbers above tell you everything you need to decide: scale, fix, or kill.
Track the metrics that matter
VSLStats gives you retention curves with timestamp milestones, revenue attribution per viewer, and qualified-view tracking. Try any plan for $1.
Start $1 Trial →