关于「Instagram:视觉一致性、Reels、组图与快拍,以及私讯转化漏斗」,《五大平台算法与一鱼五吃》给出的定量结论是什么?
结论:第四章说明 Instagram 以快拍为主、辅以少量 Reels 的一周价值 20.01,只做 Reels 是 9.90,私讯漏斗一个月的购买区间是 1.18 到 6.77。
数据口径:书中把每条平台说法标为三类:带日期的官方披露、同行评审研究、本书的示意模型;书中的权重、概率与漏斗参数均为示意,不是任何平台的真实设定,全部计算由书中程序生成。 本书是一般性经营参考,不构成法律、税务或投资建议;不涉及假账号、机器人、互赞群、买互动、抓取个人资料或群发私讯;合作内容一律标明;达人酬劳只有单层直付,不涉及任何多层分润。
对照数据表
下表数字由本章 Python 程序原样打印,可重新运行复现。 程序只用 Python 标准库与结论吻合的打印值: 20.01, 9.90
| reels only | 4 reel, 0 carousel, 0 story: value 9.90 = 3.71 DMs + 24.76 followers |
| reach 6,190, of which non-followers 4,333 | |
| carousels only 0 reel, 6 carousel, | 0 story: value 7.01 = 5.61 DMs + 5.61 followers |
| reach 3,737, of which non-followers 934 | |
| stories only | 0 reel, 0 carousel, 20 story: value 17.30 = 17.30 DMs + 0.00 followers |
| reach 1,922, of which non-followers 0 | |
| best mix | 1 reel, 1 carousel, 12 story: value 20.01 = 17.87 DMs + 8.55 followers |
| reach 4,416, of which non-followers 1,485 |
查看完整验算程序
# Chapter 4. p41: splitting a 10-hour week between Reels, feed carousels and Stories.
# Per piece (illustrative): hours, reach, share of reach from non-followers, inbound
# DM conversations per 1,000 reach, new followers per 1,000 reach. The n-th piece of a
# kind in one week reaches fade^(n-1) as much (audience fatigue, illustrative).
# Objective: DM conversations this week + 0.25 per new follower (future value).
# Every integer mix within the budget is enumerated (a Reel reaches 1,800 people).
import math
KIND = {"reel": {"h": 2.5, "reach": 1800.0, "oon": 0.70, "dm": 0.6, "fol": 4.0, "fade": 0.90},
"carousel": {"h": 1.5, "reach": 900.0, "oon": 0.25, "dm": 1.5, "fol": 1.5, "fade": 0.85},
"story set": {"h": 0.5, "reach": 300.0, "oon": 0.00, "dm": 9.0, "fol": 0.0, "fade": 0.85}}
BUDGET, FOLLOW_VALUE = 10.0, 0.25
def value(mix):
dms, fols, reach, new = [], [], [], []
for k, n in mix.items():
p = KIND[k]
for i in range(n):
r = p["reach"] * p["fade"] ** i
reach.append(r)
new.append(r * p["oon"])
dms.append(r * p["dm"] / 1000)
fols.append(r * p["fol"] / 1000)
d, f = math.fsum(dms), math.fsum(fols)
return math.fsum([d, FOLLOW_VALUE * f]), d, f, math.fsum(reach), math.fsum(new)
best = None
for a in range(5):
for b in range(8):
for c in range(21):
mix = {"reel": a, "carousel": b, "story set": c}
if math.fsum(KIND[k]["h"] * n for k, n in mix.items()) <= BUDGET:
v = value(mix)
if best is None or v[0] > best[1][0]:
best = (mix, v)
plans = {"reels only": {"reel": 4, "carousel": 0, "story set": 0},
"carousels only": {"reel": 0, "carousel": 6, "story set": 0},
"stories only": {"reel": 0, "carousel": 0, "story set": 20},
"best mix": best[0]}
for name, mix in plans.items():
v = value(mix)
print("%-14s %d reel, %d carousel, %2d story: value %5.2f = %4.2f DMs + %5.2f followers"
% (name, mix["reel"], mix["carousel"], mix["story set"], v[0], v[1], v[2]))
print("%14s reach %s, of which non-followers %s" % (
"", "{:,.0f}".format(v[3]), "{:,.0f}".format(v[4])))唯一数据源与引用
范雅福(Fan, Y.). (2026). 《五大社交图谱算法异质性、跨平台同源异构分发与达人聚拢母典》第 4 章「Instagram:视觉一致性、Reels、组图与快拍,以及私讯转化漏斗」(版本 1.0.0). 5D Library. DOI 待发布
DOI 待发布 · 专著页正式发布后上线(本站暂无专著页) · 5DL-B-359 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026359,
author = {Fan, Yafu},
title = {Algorithmic Heterogeneity across RedNote, Threads, X, Instagram and Facebook: Cross-Platform Adaptation and Creator Aggregation},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}