ContentIQ working paperCIQ-WP-2026-20CategoryVersion 11 Oct 2026
Phones & Electronics on TikTok Shop: a momentum rebound, high-ticket revenue and formats that sell per view
606 videos tracked, 203 analysed in depth, 100 top-selling products, 2 Sept–1 Oct 2026
MediaLabs Research · in collaboration with Coherence Limited
Key findings
- 01+13.4%revenue in the latest 4 weeks against the previous 4, after −15.9% over 13 weeks
- 025.5%share of all tracked marketplace revenue, ranking 6th of 28 categories
- 031.6×sales per view for talking-head videos (90% range 1.05–2.32), against the market average
- 04$65revenue-weighted median price, with the middle half of revenue between $35 and $179
- 0530.8%share of top-product revenue from products launched in the last 90 days
- 060.64AUC of the top-seller model (90% range 0.57–0.70), so only modest predictive power
Contents
Abstract
We analyse what sells in Phones & Electronics on TikTok Shop, the sixth-largest of 28 categories by revenue. Over the latest four weeks its revenue was 13.4% above the previous four, though it is still 15.9% below the previous 13 weeks. Products priced at $60 and over earn the most revenue per product (1.17–1.27× the average product), and talking-head and review formats sell at about 1.6× the market average per view, with wide ranges. These are associations from a thin, single-window sample, and a predictive model of top-selling videos has only modest accuracy (AUC 0.64).
1. Introduction
Brands choosing what to list and creators choosing what to film need to know two things: where in a category the money is, and which kinds of video turn views into sales. This paper asks both questions for Phones & Electronics on TikTok Shop: how the category is moving, where its sales happen, which prices, commission rates and product ages are associated with revenue, how competitive it is, and which openings and formats sell.
Prior work on short-video popularity shows that attention is long-tailed and that early popularity predicts later popularity [1]. Recent prediction benchmarks find that creator features matter most and that multimodal content signals add comparatively little [2, 3]. Research on social commerce suggests live-stream selling builds trust and engagement [4], that affiliate commissions are set to balance incentives [5], and that online retail sells a long tail of niche products beyond the bestsellers [6]. Evidence on sales per view for a specific category is scarcer, which this paper addresses.
2. Data
The window runs from 2 September to 1 October 2026. We tracked 606 videos in Phones & Electronics, and our AI analyst watched and analysed 203 of them in depth; this is a count of videos, not of views. The marketplace figures rest on 100 top-selling products in this one category, compared with all 28 categories tracked.
For each analysed video we recorded the opening (hook), the format and the production style, along with whether it appeared boosted with ads. For the marketplace we used weekly revenue, sales channels, product prices, affiliate commission rates, launch dates, shop and product counts and monthly revenue. The latest marketplace week ends 29 September 2026.
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: 606 videos tracked in Phones & Electronics, 203 of them watched and analysed; the market baseline uses all 3,342 watched videos; 100 top-selling products in Phones & Electronics. 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, where 1 is the market average. Small groups are shrunk toward the average by empirical Bayes [7], and we report 90% intervals computed on the log scale. A predictive model of top-selling videos is a logistic model retrained daily and tested on recent weeks it had not seen (time-split). We report AUC [8] with a bootstrap range [9].
Momentum compares the category’s estimated revenue in the latest 4 and 13 weeks with the same number of weeks before, using weekly revenue over the past year; shares are of all categories tracked. Channels split the latest week’s revenue into shoppable video, LIVE and the shop tab, plus the share credited to affiliate creators. Prices, commission rates and product age are computed for the top-selling products, with revenue-weighted medians and revenue per product by band against the average product.
Competition is measured by shops and active products, average revenue per shop, and concentration among tracked top products as the top ten’s share and the Herfindahl–Hirschman index [10]. Seasonality compares each month’s revenue with the category’s average month over the past year (1 = a normal month), leaving out partial months; separating season from trend follows [11].
4. Results: openings that sell
Among the openings, Price or deal sells at 1.4× the market average per view (90% range 1.00–1.97; 17 videos, 9.7% of watched-video sales). Question openings sit at 1.37× but the range is wide (0.86–2.21; 10 videos). Bold claim (1.24×, 24 videos) and Problem call-out (1.17×, 36 videos, high confidence) are above average, but their ranges include 1.
Comparison openings sell below average at 0.67× (0.53–0.85; 21 videos), even though they account for 15.4% of watched-video sales. Unboxing reveal sits at 0.70× (0.46–1.07; 11 videos). Relatable POV is the most common opening (39 videos, 22.5% of sales) and sells near but below average at 0.93× (0.71–1.21). Only the Price or deal and Comparison ranges exclude 1, and most groups hold fewer than 25 videos.
| Value | Videos (n) | Share of sales | Sales per view | 90% range |
|---|---|---|---|---|
| Price or deal | 17 | 9.7% | 1.40× | 1.00–1.97 |
| Question | 10 | 7.3% | 1.37× | 0.85–2.21 |
| Bold claim | 24 | 14.3% | 1.24× | 0.99–1.55 |
| Straight into the demo | 11 | 5.2% | 1.20× | 0.90–1.60 |
| Problem call-out | 36 | 20.2% | 1.17× | 0.93–1.49 |
| Relatable POV | 39 | 22.5% | 0.93× | 0.71–1.21 |
| Unboxing reveal | 11 | 5.4% | 0.70× | 0.46–1.07 |
| Comparison | 21 | 15.4% | 0.67× | 0.53–0.85 |
Note. Sales per view is relative to the market (1.00 = average), shrunk toward the average for small groups (empirical Bayes); ranges are 90% intervals on the log scale.
5. Results: formats and production
Talking head (1.56×, range 1.05–2.32; 15 videos) and Review or testimonial (1.55×, 1.05–2.30; 13 videos) have the highest sales per view, and the lower ends of both ranges are only just above 1. Unboxing is 1.35× with a range that includes 1 (0.86–2.11; 14 videos). Voice-over B-roll (1.48×) has only 5 videos and low confidence, so we do not interpret it.
Demo or tutorial is the largest format, with 92 videos and 43.1% of watched-video sales, but sells slightly below average per view at 0.90× (0.79–1.03). Skits sell at 0.71× (0.54–0.92; 27 videos), the clearest underperformer among formats with a meaningful sample.
All 127 videos with a recorded production style were filmed live action (0.996×, 0.86–1.15), so this sample cannot compare production styles. The predictive model reached an AUC of 0.64 (90% range 0.57–0.70) on 2,667 training rows. It was not trained only on analyses that never saw sales (cleanOnly is false), so it should be read as descriptive. Its strongest signals were filmed live action (odds ratio 2.54), a visual pattern-break combined with a review (1.92) and no visible AI (1.71), while education content (0.59) and product in the first second (0.62) were associated with lower odds.
| Value | Videos (n) | Share of sales | Sales per view | 90% range |
|---|---|---|---|---|
| Talking head | 15 | 12.5% | 1.56× | 1.04–2.32 |
| Review or testimonial | 13 | 7.4% | 1.55× | 1.05–2.30 |
| Voice-over B-roll | 5 | 2.8% | 1.48× | 0.93–2.35 |
| Unboxing | 14 | 8.6% | 1.35× | 0.86–2.11 |
| Other | 5 | 2.8% | 1.31× | 0.86–2.00 |
| Demo or tutorial | 92 | 43.1% | 0.90× | 0.79–1.03 |
| Comparison | 10 | 4.4% | 0.89× | 0.62–1.27 |
| Skit | 27 | 18.4% | 0.71× | 0.54–0.92 |
Note. Sales per view is relative to the market (1.00 = average), shrunk toward the average for small groups (empirical Bayes); ranges are 90% intervals on the log scale.
6. Results: market, prices and season
Phones & Electronics held 5.5% of tracked marketplace revenue over the latest four weeks, ranking 6th of 28. Revenue rose 13.4% against the previous four weeks, but over 13 weeks it fell 15.9%, against a market decline of 11.1%. In the latest week 64.4% of revenue came through shoppable video, 18.6% through LIVE and 17.1% through the shop tab, close to the market mix (64.9%, 17.1%, 18.0%). Affiliate creators were credited with 75.8% of revenue.
The revenue-weighted median price is $65, and the middle half of revenue lies between $35 and $179. Products at $100 and over (30 products) took 35.2% of top-product revenue at 1.17× the average product, and the $60–100 band (14 products) earned 1.27× on 17.7%. The $10–20 band had 21 products and 15.6% of revenue at 0.74× the average product. The median commission is 8%. Products paying 10–15% earned 1.03× the average, and the 20–30% band, with only two products, earned 1.87×, too few to generalise. Products at under 5% commission sold only 32.1% of their revenue through video, against 78.9% in the 5–10% and 10–15% bands.
Products launched in the last 90 days produced 30.8% of top-product revenue. The 30–90 day band (19 products) earned 1.39× the average product, while the youngest band (under 30 days, 7 products) earned 0.63×. The category has 8,631 shops (median category: 4,461) and 63,555 active products. Average revenue per shop is $2,516, against $2,401 for the median category. Concentration among top products is moderate: the top ten take 30.1% of top-product revenue (the same as the median category), with an HHI of 178.
Monthly revenue against a normal month ranged from 0.81 in November 2025 to 1.36 in June 2026, with March at 1.23. July (0.87) and August (0.85) were below normal, and December 2025 was 1.00. Black Friday / Cyber Monday starts in 40 days, holiday gifting in 45 days and New Year resolutions in 92 days. Last year’s November and December showed no clear peak.
| Band | Products (n) | Share of revenue | Revenue per product vs average |
|---|---|---|---|
| Under $10 | 1 | 0.8% | 0.79× |
| $10–20 | 21 | 15.6% | 0.74× |
| $20–35 | 11 | 8.5% | 0.77× |
| $35–60 | 23 | 22.3% | 0.97× |
| $60–100 | 14 | 17.7% | 1.27× |
| $100 and over | 30 | 35.2% | 1.17× |
Note. Among the top-selling products we track in the latest week; revenue per product is against the average product in the same set (1.00 = average).
7. Discussion
For brands, the evidence points to high-ticket products as the main revenue pool: products at $60 and over earn more revenue per product than the average product. Newer products (30–90 days old) also earn more than average, which suggests openings for new listings, although the youngest products do not yet. The category is crowded, with about twice as many shops as the median category, but revenue per shop is similar and concentration is typical. Commission looks like a lever for video reach more than for revenue: low-commission products sell much less through video.
For creators, talking-head and review formats and price-led or question openings are associated with higher sales per view, while comparison openings and skits are associated with lower. Demos are the volume format, with 43.1% of watched-video sales, but they sell slightly below the average per view. This is consistent with prior work showing that popularity is hard to predict from content alone [2, 3], and the model’s modest AUC agrees.
Alternative explanations are plentiful. Formats may differ in the products, price points or creators they attract, which the data cannot separate. Ads boosting and creator audience size also affect sales per view, and measuring advertising returns from observational data is notoriously hard [12]. The 13-week decline may reflect a strong spring and early summer rather than a lasting trend, and the 13.4% four-week recovery could be a short-term swing.
8. Conclusion
Phones & Electronics is a large, mid-ranked category that has rebounded recently, selling mainly through affiliate-driven video and concentrated in higher-priced products. Within video, talking-head, review and price-led content sell at above-average rates per view, with wide ranges, while comparison and skit content sell below. These are associations, not causal effects. Brands and creators should treat them as hypotheses to test on their own products, particularly ahead of the Black Friday and holiday gifting moments.
Methodology
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: 606 videos tracked in Phones & Electronics, 203 of them watched and analysed; the market baseline uses all 3,342 watched videos; 100 top-selling products in Phones & Electronics. Predictions come from a logistic model retrained daily and tested on recent weeks it had not seen.
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. Each hook and format group contains few videos (5–92), and many 90% ranges include 1, so most differences are suggestive only. Only 203 of 606 tracked videos were analysed in depth, and the sample covers a single 30-day window in one category. All analysed videos share one production style, so production cannot be compared. Seasonality rests on about ten months from one year, which cannot separate season from trend, and the model was not trained on clean-only analyses and has modest accuracy. The price, commission and age bands rest on 100 products, and bands with 1–2 products (such as 15–20% and 20–30% commission) are too thin to interpret.
References
- [1]Szabo, G., & Huberman, B. A. (2010). Predicting the popularity of online content. Communications of the ACM, 53(8), 80–88. cacm.acm.org/research/predicting-the-popularity-of-online-content
- [2]Ye, L., Zhang, Y., Wu, Y., et al. (2025). MVP: Winning solution to SMP Challenge 2025 video track. arXiv:2507.00950. arxiv.org/abs/2507.00950
- [3]Lu, J., Wang, W., Xiao, M., et al. (2024). M3TR: Temporal retrieval enhanced multi-modal micro-video popularity prediction. arXiv:2411.15455. arxiv.org/abs/2411.15455
- [4]Wongkitrungrueng, A., & Assarut, N. (2020). The role of live streaming in building consumer trust and engagement with social commerce sellers. Journal of Business Research, 117, 543–556.
- [5]Libai, B., Biyalogorsky, E., & Gerstner, E. (2003). Setting referral fees in affiliate marketing. Journal of Service Research, 5(4), 303–315.
- [6]Brynjolfsson, E., Hu, Y. J., & Smith, M. D. (2003). Consumer surplus in the digital economy: Estimating the value of increased product variety at online booksellers. Management Science, 49(11), 1580–1596.
- [7]Efron, B., & Morris, C. (1975). Data analysis using Stein’s estimator and its generalizations. Journal of the American Statistical Association, 70(350), 311–319.
- [8]Fawcett, T. (2006). An introduction to ROC analysis. Pattern Recognition Letters, 27(8), 861–874.
- [9]Efron, B., & Tibshirani, R. J. (1993). An Introduction to the Bootstrap. Chapman & Hall.
- [10]Rhoades, S. A. (1993). The Herfindahl–Hirschman index. Federal Reserve Bulletin, 79, 188–189.
- [11]Cleveland, R. B., Cleveland, W. S., McRae, J. E., & Terpenning, I. (1990). STL: A seasonal-trend decomposition procedure based on loess. Journal of Official Statistics, 6(1), 3–73.
- [12]Lewis, R. A., & Rao, J. M. (2015). The unfavorable economics of measuring the returns to advertising. The Quarterly Journal of Economics, 130(4), 1941–1973.
Cite as
MediaLabs Research, in collaboration with Coherence Limited (2026). Phones & Electronics on TikTok Shop: a momentum rebound, high-ticket revenue and formats that sell per view. ContentIQ Working Paper CIQ-WP-2026-20, version 1. https://medialabs-co.com/research/category-phones-and-electronics
Analysis and data: Coherence Research, Coherence ContentIQ. Published by MediaLabs in collaboration with Coherence Limited.
Version history
- Version 11 Oct 2026This version