Take out your phone. Hold it in one hand, the way a shopper holds it with a coffee in the other. Open your store. That is what most of your holiday customers will see.
The phone is the main till
Adobe forecasts $275.1 billion in US online holiday sales for 1 November to 31 December, up 6.7%. Mobile is expected to take 57.4% of it, up from 56.4% last year. On Thanksgiving Day the share is forecast at 63%, up from 61.6%.
Adobe also expects AI-referred traffic to US retail sites to rise 159% on Thanksgiving. Those shoppers arrive from an assistant's answer, and some will have one question left.
The forecast says nothing about live calls. What follows is a test I am proposing, not a finding.
Tap one: find the call
Open one product page on your own phone, on mobile data. Start a timer.
How many seconds pass before you see a way to talk to a person? Count it from the moment the page loads, and do not scroll before you look.
If the answer is "after the gallery, the reviews and the size chart," a shopper in a queue will not get there. The option should sit where the price and the buy button already are.
Tap two: talk without typing
Start the call. Note whether you had to type anything: a name, an email, a phone number.
A shopper on a phone is in a queue or on a sofa with relatives. They will not fill in a form to ask a question. On a BeamCart call there is no account and no app to install, the microphone starts and the camera is optional, so the shopper can talk without appearing on video.
Check two more things. Does the product stay on screen while you talk? Can you still see the buy button without ending the call?
Tap three: buy without leaving
Ask for a suggestion, then add it to the cart and pay with the call still open. Count the taps from "add" to "paid."
Payment is part of this test. Adobe forecasts $21.3 billion in buy now, pay later spending this season, so the pay step on a phone should offer what your shoppers actually use.
Run the one-hand test
Thanksgiving falls on 26 November, and Cyber Monday on 30 November.
First, run the three taps yourself and write down three numbers: seconds to find the call, characters typed before talking, taps from add to paid.
Second, ask someone who did not build the page to do the same with one hand, holding a cup in the other. Watch without helping.
Third, fix the worst number first. A live call is only useful if a shopper can reach it, and on a phone that means reaching it with a thumb.
Adobe's figures are a forecast of US online sales, as reported by Digital Commerce 360. They describe where shoppers will be, not what will work for your store. Your three numbers do that.
Source
Digital Commerce 360, "Adobe projection: 2026 online holiday sales to reach $275 billion" (29 September 2026, reporting Adobe's forecast for US online sales from 1 November to 31 December 2026: $275.1 billion, up 6.7%; Cyber 5 $47.5 billion; mobile 57.4% of the season, up from 56.4%, and 63% on Thanksgiving Day, up from 61.6%; AI traffic to US retail sites up 159% on Thanksgiving; buy now, pay later $21.3 billion for the season). The article does not name Adobe's data source, so the figures are as reported there, and they are a forecast for US online sales only. They say nothing about live calls. The three taps, the three counts and the one-hand test are editorial applications, not findings. The statements about the BeamCart call (no account, no app, microphone starts and camera optional) come from the local What is live shopping page. Photos: Wikimedia Commons, CC0 (hand on a smartphone beside a coffee cup, www.Pixel.la; "Thanksgiving Day: Ways and Means", Winslow Homer, Harper's Weekly, 27 November 1858, Metropolitan Museum of Art; hand holding a smartphone, NeONBRAND).
