Reuben Cohen

[Reuben Cohen @MRRubyCohen]

@ecomrat @ecomcowboy @DTCMidas @jforjacob @conortrains that band is usually where first-order contribution and conversion can still coexist. under it you're betting the LTV, over it you're betting the conversion rate.

[Reuben Cohen @MRRubyCohen]

@ArbMails the pause/skip people are the ones that need a different letter. blended "active subscribers" treats a two-skip customer like a healthy one.

[Reuben Cohen @MRRubyCohen]

@halalmails the 30% also trains the January list. they don't come back at full price because november taught them not to. a cheap gift only works if they actually value it ; most GWP tests never check that.

[Reuben Cohen @MRRubyCohen]

@linafahizul beauty cadence is a guess dressed up as a plan. if usage isn't predictable, a 2 or 3 pack on order one is cleaner than a 30-day rebill they didn't ask for.

[Reuben Cohen @MRRubyCohen]

@rahijain save rate is a button metric. if they never take another charge, you didn't save them. you delayed the cancel by a week.

[Reuben Cohen @MRRubyCohen]

@AdamKitchen_co not that surprising if the bundle just pulled forward the next two orders. 180-day LTV looks worse because you already sold month 2 and 3 on day one.

[Reuben Cohen @MRRubyCohen]

@JackHenryPaxton @dtcprophet default the single and leave the 2-pack next to it. default the bundle and you buy AOV with people who would have paid $25 and now bounce.

[Reuben Cohen @MRRubyCohen]

@shanerostad @dtcprophet right. you didn't raise AOV, you borrowed month two and three and spent them on today's CAC. looks fine until the reorder never arrives.

[Reuben Cohen @MRRubyCohen]

@DTCMidas the audience shrink is the expensive part. AOV went up, the bid went with it, and you're suddenly shopping a thinner slice of the same catalog.

[Reuben Cohen @MRRubyCohen]

@dtcprophet the $79 sub is the giveaway. that's not an AOV lift, that's asking a stranger to prepay three months of something they haven't tried. conversion takes the hit before LTV ever shows up.

[Reuben Cohen @MRRubyCohen]

And processors keep the original processing fee on a refund. Stripe's published policy spells that out. Every reversed $97 leaves about $3 with your processor, so the timer's refund tail costs you on the amount and on the fee.

[Reuben Cohen @MRRubyCohen]

Illustrative: 1,000 buyers of a $47 digital product with a one-click $97 upsell. No timer, 8% take and 6% of those upsells refunded. Add a ten-minute countdown and say take goes to 12%, refunds to 14%. Both numbers move, and most timer tests show you only the first.

[Reuben Cohen @MRRubyCohen]

Your read is net revenue per front-end buyer with the refund window closed, and on digital products that window runs weeks past the sale. A take rate you measure on day two is a forecast.

[Reuben Cohen @MRRubyCohen]

The heavy bucket is where next year's cancels come from, because a subscriber who has skipped twice has already rehearsed leaving. Split the skip report by subscriber rather than by cycle, and move the two-in-three cohort onto a longer cadence before the third skip.

[Reuben Cohen @MRRubyCohen]

A subscriber who skips two cycles in every three pays you 4 times a year on a $40 box: $160, or $13.33 per subscriber-month. One who skips 20% of cycles pays 9.6 times, $384, $32 a month. Both read as active on your dashboard.

[Reuben Cohen @MRRubyCohen]

Blended skip rate hides the shape. Thirty subscribers skipping two cycles in three, with seventy skipping none, gives the same 20% blended rate as all hundred skipping one in five. Both produce 960 orders this year. They are not the same book.

[Reuben Cohen @MRRubyCohen]

Follow-up consults repeat the consult cost every time the prescriber reviews again. And decline rate moves everything: go from 10% to 25% and your cost per patient runs from $166.67 to $200, a 20% worse payback with no ad changed. Divide by approved patients, not by orders.

[Reuben Cohen @MRRubyCohen]

A $150 cost per checkout at a 20% prescriber decline rate is a $187.50 cost per patient. Most telehealth payback models divide by checkouts. Ecommerce has one gate, the customer buys. Telehealth has two: the customer pays, then a clinician decides.

[Reuben Cohen @MRRubyCohen]

What these models leave out. Refund a declined intake and most processors keep the fee, so 20 declines per 100 checkouts at $6 is $120 charged straight to acquisition. Refunds when a clinician ends treatment mid-cycle belong in the model as a monthly rate.