Slippery slide copywriting: how Joseph Sugarman's rule keeps VSL viewers watching

Slippery slide copywriting is Joseph Sugarman's idea that every part of an ad has one job: get the next part read, so the reader slides from the headline to the order without finding a place to stop. In a VSL you don't have to guess whether the slide works. The retention graph shows exactly where viewers hit a ledge and leave.

Who was Joseph Sugarman?

Joseph Sugarman (1938–2022) was an American direct marketer and copywriter. He founded JS&A Group in 1971, a mail-order company that sold consumer electronics such as pocket calculators and digital watches through print ads, and in 1986 he introduced BluBlocker sunglasses, which were marketed mainly through infomercials he appeared in himself (Wikipedia). MediaPost's obituary credits JS&A with pioneering credit card orders through a toll-free 800 number (MediaPost).

He wrote several books on selling with words, including Advertising Secrets of the Written Word (1998), Triggers (1999) and The Adweek Copywriting Handbook, which Wiley published in December 2006. The handbook's table of contents includes a chapter called "The Slippery Slide", one called "The First Sentence" and one called "Seeds of Curiosity" (Wiley table of contents).

Joseph Sugarman, in dates

1. 1938Born
2. 1971Founds JS&A GroupA mail-order company that sold consumer electronics, such as pocket calculators and digital watches, through print ads.
3. 1986Introduces BluBlocker sunglassesMarketed mainly through infomercials he appeared in himself.
4. 1998Book: Advertising Secrets of the Written Word
5. 1999Book: Triggers
6. Dec 2006Book: The Adweek Copywriting HandbookPublished by Wiley. Its chapters include "The Slippery Slide", "The First Sentence" and "Seeds of Curiosity".
7. 2022Dies

His ads sold his own company's products, so they were judged by orders, not applause: the same scoreboard a VSL lives on.

What is the slippery slide in copywriting?

The slippery slide describes how good copy feels from the inside: once you start, you keep going, because nothing gives you a reason to stop. AWAI, summarizing Sugarman's handbook, describes it as copy compelling enough that the prospect keeps moving through the message (AWAI). The order is the bottom of the slide. Everything above it is either slippery or a ledge.

The mechanism is momentum. Each line spends a little of the audience's attention and has to pay it back with a reason to continue: a new fact, a turn in the story, a question, a promise of something better just ahead. When a line takes attention without paying it back, the reader stops. In print they might skim and climb back on. Often they don't.

In practice, a slippery message rests on three habits:

1. An effortless start. The first line is short and easy to begin, so starting costs almost nothing.
2. Constant payoff. Every few lines hand over something: a fact, an answer, a turn.
3. Curiosity carried across breaks. Each section ends with a reason to start the next one. Sugarman gave curiosity its own chapter, and in a VSL the breaks between sections are exactly where people stop.

Why the slippery slide matters more in a VSL than on a page

A sales letter forgives a ledge. A bored reader can skim three paragraphs, catch a subhead and get back on. A VSL viewer has two choices at every second: keep watching or leave. Many VSL pages also hide the seek bar, which removes the option of skipping ahead. So every ledge in a video costs you viewers for good, and the pitch only reaches the people who slid all the way down.

Small leaks compound. Say 1,000 people start a 20-minute VSL and you lose 4% of the remaining audience every minute. After 20 minutes, about 442 are still watching when the offer appears (1,000 × 0.96 to the 20th power). Cut the leak to 3% a minute and about 544 reach the offer. That's roughly 100 more people seeing the price, from the same traffic, without touching the ad or the offer.

How a small leak compounds

Illustrative: 1,000 viewers start a 20-minute VSL; how many are still watching when 4% or 3% of those left leave every minute

• Losing 4% a minute
• Losing 3% a minute

02505007501,000048121620442544

Minutes into the video

Show the numbers

Minutes into the video	Losing 4% a minute	Losing 3% a minute
0	1,000	1,000
2	922	941
4	849	885
6	783	833
8	721	784
10	665	737
12	613	694
14	565	653
16	520	614
18	480	578
20	442	544

That's why slipperiness is a revenue lever, not a matter of style. Every stage after the pitch, from button clicks to orders, scales from the number of viewers who arrive there.

How do you measure slipperiness? Read the retention graph

Your retention graph is the slide drawn as a line. The horizontal axis is time in the video; the vertical axis is the share of viewers still watching. A slippery VSL drops in the first seconds, as people sample it, then settles into a gentle, steady slope. Ledges show up as cliffs: short stretches where the line falls much faster than it does around them.

A workflow that finds ledges:

1. Collect at least a few hundred plays on one version (a rule of thumb) so the curve isn't noise.
2. Find the three steepest drops after the first 30 seconds.
3. Note the exact timestamps and open the script at those lines.
4. Name each ledge using the ledge finder below and write one fix per ledge.
5. Change one ledge per test, so you know which fix worked.

Ledge finder

Six ledges: what the graph shows, the usual cause and the fix

1Where does the line fall much faster than it does around it?

In the first 15–30 secondsDuring a technical explanationA long, steady leak in a storyWhen the price first appearsRight after a big revealDuring a static stretch

Best fit

Slow start

A steep fall in the first 15–30 seconds. Usual cause: a logo, greeting or credentials before the hook. Fix: open on the hook line; move credentials later.

Best fit

Jargon wall

A cliff during a technical explanation. Usual cause: terms the viewer doesn't know. Fix: one idea per sentence; explain with a comparison.

Best fit

Unpaid promise

A long, steady leak during a story. Usual cause: a story with no visible link to the viewer's problem. Fix: tie each story beat back to the viewer's problem.

Best fit

Early pitch signal

A cliff when a price or "order now" first appears. Usual cause: selling before the case is made. Fix: hold the pitch until the argument is complete.

Read more: Delayed CTA Button: When to Show the Buy Button

Best fit

Closed loop

A drop right after a big reveal. Usual cause: the main question was answered and nothing new was opened. Fix: open the next question before closing the current one.

Best fit

Dead visuals

A slow leak during a static stretch. Usual cause: one slide or one framing held too long. Fix: change something on screen every few seconds.

Read more: VSL Editing: How to Edit a Sales Video for Retention

In RevvyScales, the retention chart is second by second with the pitch marked, so you can line up a cliff with the exact script line and see how many viewers are left when the offer starts (reading your analytics). For a full walkthrough of curve shapes, see how to read a video retention graph.

Seeds of curiosity and open loops in a VSL script

A seed of curiosity, in script terms, is a short line at the end of a section that makes the next section feel necessary. An open loop is the bigger version: a question you raise early and answer later, which gives the viewer a reason to stay through everything in between.

Here's a section of a script for a fictional wireless meat thermometer, PitProbe, first without seeds:

Every sentence is a feature, nothing pulls the viewer forward, and the discount at the end is a pitch signal that arrives before the viewer cares. Now with seeds:

Each line pays something and asks for the next. The open loop ("exactly when it happens") gets paid off later, and a new loop should open before it closes.

Rules for loops that don't backfire:

• Pay every loop. An open loop is a promise. Never answering it feels like a trick, and the buyers who notice are the ones who ask for refunds.
• Nest, don't stack. Open the next loop just before you close the current one, so there's always one question in the air. Ten unanswered questions at once feel like stalling.
• Make the loop about the viewer. "I'll show you why your brisket dries out" pulls harder than "I'll tell you my story later."
• Pay off in installments. A question raised at minute 2 and answered at minute 19 needs smaller payoffs along the way, or viewers decide it isn't coming.

The first loop has to work within seconds; for openings by awareness level, see VSL hook examples.

Transitions: where viewers fall off

Most cliffs sit at the joints of a VSL: story to discovery, discovery to mechanism, mechanism to proof, proof to offer. A joint is where the viewer silently asks where this is going. If the script doesn't answer, they leave. A good transition answers the question before the viewer asks it.

Four transitions that answer "where is this going?"

Example lines for PitProbe, a fictional wireless meat thermometer

ConsequenceContrastQuestionProof

Consequence

Shows why the last point matters.

"Which meant every cook I'd ever done was a guess."

Contrast

Sets the next idea against the last.

"That's what the cookbooks say. Here's what the meat actually does."

Question

Voices the viewer's next question.

"So how do you see inside the meat without opening the lid?"

Proof

Promises evidence for the claim just made.

"You don't have to take my word for it. Watch this cook."

A quick test: read the last sentence of each section and the first sentence of the next one aloud, back to back. If they don't sound like they belong together, you've found a joint that needs work.

Remove friction: write for the ear

Copy that reads well can still sound slow. Viewers can't reread a sentence they missed, so a VSL script needs shorter sentences than a letter, one idea per sentence and concrete words. When you use a number, say what it means in the viewer's terms.

Runtime math keeps you honest. Presenters typically speak somewhere around 140 to 160 words a minute, so a 3,000-word script runs roughly 19 to 21 minutes. If your retention graph shows viewers leaving during a seven-minute backstory, that's about 1,000 words you can cut or compress before you touch anything else.

Friction worth cutting:

• Greetings, logos and housekeeping in the first ten seconds.
• Credentials before the viewer cares who you are. Earn the interest first, then explain why you're the one to listen to.
• Synonyms for the same thing. If it's "the sensor" in minute 3, don't call it "the probe tip" in minute 5.
• Any sentence the presenter stumbles over in a table read.

Edit for momentum

The edit is the second half of the slide. A script can be slippery on paper and still stall on screen if the picture sits still while the voice explains.

• Cut air. Trim pauses, breaths and repeated takes.
• Change the picture on purpose. At every point where your graph showed a ledge, plan a visual change: b-roll that shows the claim, a new framing, a key word on screen.
• Show, then tell. For a product you can demonstrate, the demo is the most slippery thing you have. Put it early.
• Caption for silent viewers. Many people start watching with the sound off. Captions keep them sliding until they tap for sound.

If your page autoplays muted, the slide really starts when the viewer taps for sound. A player that restarts from 0:00 on unmute makes sure the first words they hear are your hook, not the middle of a sentence. RevvyScales does this on every VSL (player conversion features). For the full editing playbook, see VSL editing.

Where should the slide end? At the pitch, not before

In Sugarman's picture, the reader's only stop is the order. In a VSL that means the offer, the guarantee and the button appear when the argument is complete. A price that flashes on screen at minute 3 is the biggest ledge you can build, because it invites the viewer to judge the price before they understand the value.

Two practical rules:

• Reveal the button with the price. Shown long before the offer, the button invites early clicks into a price nobody was ready for. Shown long after, it makes ready buyers wait, and some drift away. See when to show the buy button.
• Check the graph at the pitch. If most viewers are gone before the offer starts, the slide is broken upstream, and a better offer won't fix it.

A slide is not a trap. Viewers should always be able to pause, leave or come back. Fake "don't close this page" warnings and invented countdowns may hold a few viewers longer, but they invite complaints, chargebacks and ad-policy trouble. Slipperiness comes from relevance and momentum, not pressure.

How to test whether a fix made the slide more slippery

Treat each ledge fix as an experiment. Split traffic between the current version and the fixed one, change nothing else, and judge in this order:

Judge a ledge fix in this order

1. 1Retention at the ledgeDid the cliff get smaller?
2. 2Pitch reachDid more viewers arrive at the offer?
3. 3Revenue per visitorDid the extra viewers buy?

A worked example: version A loses viewers sharply at 4:10, where a 50-second credentials section starts. Version B moves the credentials after the demo. Over 3,000 visitors each, retention at 5:00 rises from 58% to 64% of viewers, pitch reach rises from 31% to 35% of all visitors, and revenue per visitor goes from $1.18 to $1.29. The improvement holds at every stage, so B becomes the new control. If retention had improved but revenue per visitor hadn't, you'd look for a promise the offer doesn't keep.

What the slippery slide doesn't tell you

• It's a principle, not a structure. The slide tells you to remove ledges; it doesn't tell you what to say. You still need an opening that fits what your viewer already knows (see the five levels of awareness) and a believable mechanism.
• Slippery isn't the same as long. Fixing a ledge often shortens the video. Don't add length to look thorough; padding is a ledge.
• Clarity beats cleverness. A loop that confuses is a ledge. If viewers can't tell what the video is about within the first minute, curiosity turns into frustration.
• The graph shows where, not why. A cliff tells you where people left; watching that section cold and asking buyers tells you why.

How to apply this to your VSL tomorrow

Work through these on your own VSL; the ticks stay in this browser:

Make your VSL more slippery

• Open your retention graph and mark the three steepest drops after the first 30 seconds.
• Find the script lines at those timestamps and label each ledge: slow start, jargon, unpaid promise, early pitch signal, closed loop or dead visuals.
• Rewrite the first ten seconds so the first words the viewer hears are the hook.
• Add a seed of curiosity to the last line of every major section.
• Read each section joint aloud and fix any transition that doesn't answer "where is this going?"
• Cut every sentence the presenter stumbles over and every synonym that renames the same thing.
• Mark a visual change at each ledge for your editor.
• Check that the button and the price appear together, once the argument is complete.
• Run the fixed version against the current one and judge by retention at the ledge, then pitch reach, then revenue per visitor.

0 of 9 done

To see where all of this sits in a complete script, start with the VSL script guide.