Shoes on TikTok Shop sell through video at mid-range prices in a cooling market

1. Introduction

Brands choosing what to list, and creators choosing what to film, face two questions in a category such as Shoes: where is demand moving, and which products and videos convert attention into sales? Prior work on short-video popularity finds that early engagement and creator characteristics predict later popularity, and that attention is long-tailed [1, 2, 3]. Those studies predict views, not sales. Shoppable video adds price, commission and competition, which matter for commercial decisions. Evidence on advertising returns also warns that observational data rarely identifies what drives sales [4]. This paper describes what sells in Shoes on TikTok Shop and how it relates to price, commission, product age, seasonality and video style, using correlation language throughout.

2. Data

The window runs from 2 September to 1 October 2026. We tracked 434 Shoes videos, and our AI analyst watched and analysed 183 of them in depth; this is a count of videos, not views. Marketplace figures (prices, commission, product age) rest on the 100 top-selling Shoes products. Sales-per-view comparisons use a market benchmark drawn from 3,342 videos analysed in depth.

Daily rankings of top-selling TikTok Shop videos in the US were collected from several independent sources and merged into one record per video. Where sources overlap, their sales estimates are cross-checked against each other, and videos whose estimates disagree by more than half are flagged. An AI analyst watched each video and recorded its opening, format, angle, production style, use of AI, hook source, length, pacing and what appears in the first second, without seeing how the video sold. Sales per view is a video group’s revenue per view divided by the market’s, shrunk toward the average for small groups, with 90% ranges. Weekly category figures for the US (estimated revenue, its split between video, LIVE and the shop tab, affiliate revenue, shops and active products, and a year of weekly revenue) and daily rankings of each category’s top-selling products (unit price, commission rate, launch date and revenue by channel) were collected from several independent sources. Medians of price and commission are weighted by revenue; momentum compares the latest weeks with the same number of weeks before. This paper covers 2026-09-02 to 2026-10-01: 434 videos tracked in Shoes, 183 of them watched and analysed; the market baseline uses all 3,342 watched videos; 100 top-selling products in Shoes. Predictions come from a logistic model retrained daily and tested on recent weeks it had not seen.

3. Methods

Sales per view is a group of videos' revenue per view divided by the market's, so 1 is the market average. Estimates for small groups are shrunk toward the average by empirical Bayes [5] and carry 90% intervals on the log scale. The predictive model is a logistic model of top-selling videos, retrained daily and tested on recent weeks it had not seen. It is reported as AUC [6] with a bootstrap range [7].

4. Results: momentum, channels and competition

Shoes accounted for 2.9% of market revenue in the latest four weeks and ranked 13th of 28 categories. Revenue in the latest four weeks was 1.4% below the previous four. Over 13 weeks it was 10.4% below the previous 13, against an 11.1% fall for the whole market, so the category has cooled roughly in line with the market rather than faster.

In the latest week, 72.5% of Shoes revenue came from shoppable video, 13.9% from LIVE and 13.6% from the shop tab. The market splits 64.9%, 17.1% and 18.0%. Affiliate creators were credited with 82.6% of Shoes revenue, so the category leans more on video and creators than the average category. Live-stream selling is associated with trust in social commerce [8], but it is a smaller channel here.

The category had 3,394 shops and 30,771 active products. Average revenue per shop was $4,985, against $2,401 in the median category, and the median category had 4,461 shops. The top ten products took 28.8% of top-product revenue, identical to the median category, and the Herfindahl–Hirschman index among top products was 178 [9]. This is low concentration: sales are spread across many products, consistent with a long tail of niche items [10].

5. Results: price, commission and product age

The revenue-weighted median price of top-selling products was $49, with the middle half between $39 and $69. The $35–60 band held 47 of the 100 products and 54.4% of revenue, and revenue per product there was 1.16× the average. The $10–20 and $20–35 bands sold at 0.67× and 0.69× the average per product, the $60–100 band at 0.90×, and products at $100 and over at 1.06× (13 products). Price endings can affect sales [11], but we did not test endings here.

The revenue-weighted median commission was 10%. Products at 10–15% commission (45 products) took 52.0% of revenue at 1.10× the average per product; those at 5–10% (also 45) took 43.7% at 0.92×. The under-5% and 15–20% bands hold only 3 and 2 products, so we do not interpret them. Commission levels vary for reasons that include the seller's margin and the need to attract affiliates [12]. The share of product revenue sold through video was 86–88% in the three bands with enough products to read.

Products launched in the last 90 days produced 17.8% of top-product revenue. Products under 30 days old (6) and 30–90 days old (16) sold at 0.82× and 0.80× the average per product. Products aged 3–6 months (11) sold at 1.44× and those aged 1–2 years (25) at 1.22×, while products over two years old (14) sold at 0.77×. Several bands are small, so these gaps are indicative only.

6. Results: hooks, formats and production

Unboxing-reveal hooks sold at 1.34× the market average per view (90% range 1.01–1.79, 11 videos, medium confidence). Relatable point-of-view hooks, the most common (51 videos), took 35.5% of watched-video sales at 1.06× (0.89–1.25). Bold-claim hooks (46 videos) took 24.9% at 0.94× (0.79–1.12). Trend-or-sound hooks (0.81×, 0.61–1.08) and straight-into-the-demo openings (0.80×, 0.60–1.06) sat below average, but their ranges include 1. Only the unboxing reveal has a range wholly above 1.

By format, unboxing sold at 1.41× (1.04–1.91) but from only 9 videos, so confidence is low. Voice-over B-roll (45 videos, 33.3% of sales) sold at 1.14× (0.95–1.36), review or testimonial at 1.08× (0.90–1.30) and demo or tutorial at 0.90× (0.74–1.09). Filmed live action took 95.5% of sales at 1.04×; photo slideshows (6 videos) sold at 0.77× (0.55–1.07), a thin base.

Our model of top-selling videos reached an AUC of 0.64 (90% range 0.57–0.70), trained on 2,667 rows. It was not restricted to analyses that never saw sales (cleanOnly is false), so it is a descriptive ranking aid rather than a clean test. Its strongest positive associations were filmed live action (odds ratio 2.54), a visual pattern-break combined with a review or testimonial (1.92) and no visible AI (1.71). Education content (0.59) and the product appearing in the first second (0.62) were associated with lower odds.

7. Discussion

For brands, the evidence points to a category where mid-priced products ($35–60) and a 10–15% commission are where revenue per product is highest, and where video and affiliates carry most sales. This is association: higher commission may reflect product margin or seller quality rather than attract more sales. Newer products are not shut out (17.8% of revenue from products under 90 days old), but the youngest bands sold below the average per product.

For creators, hook and format effects are modest and uncertain. Most 90% ranges include 1, so we would not treat any style as a rule. Unboxing and reveal-style openings are the most consistent signal, but the formats rest on 9–11 videos. The model's modest AUC reinforces that a video's style explains only part of why it sells, in line with prior work where creator features matter most [2]. Product, price and creator reach are plausible confounders we cannot separate.

Seasonally, Shoes revenue against a normal month (1 = normal) ranged from 0.80 in August to 1.31 in March and 1.21 in June. November 2025 was 1.08 and December 0.91, so the past year shows no clear holiday lift. With one year of history, season cannot be separated confidently from trend [13]. Halloween began on 25 September, Black Friday / Cyber Monday starts in 40 days (10 November) and holiday gifting in 45 days (15 November).

8. Limitations

Most videos in the sample are top-ranked, so findings mostly separate strong sellers from good ones rather than from all videos; typical and weak videos are being added. Product figures cover the top-selling products we track in each category, not every listing, so they describe what sells best rather than everything on sale. Results are associations, not causes. Paid promotion, creator audience size and product price are not fully controlled for. Revenue, views and sales are estimates, not figures reported by TikTok. This is a single 30-day window for one category, and several groups are small: all formats other than the largest few and the photo-slideshow production style rest on fewer than 10 videos, and commission bands under 5% and 15–20% hold only 3 and 2 products. Seasonality covers one year, so holiday effects are not separable from trend. The model was not restricted to analyses that never saw sales, and all results are associations, not causal effects.

9. Conclusion

Shoes on TikTok Shop is a mid-sized category (13th of 28) that has cooled about as much as the market over 13 weeks. Sales run mainly through video and affiliates, are spread across many products, and are strongest in the $35–60 band. Among video styles, only unboxing-reveal openings show a range wholly above the market average, from small samples, and the model's ability to rank top videos is modest. Brands and creators should treat these patterns as starting points to test, not as guarantees.