Analysis of Market Demand and Skills of Product Manager

Zijing Wu

Computer Science and Technology, Xiamen University Malaysia, Sepang, Malaysia

Keywords: Salary Structure, Human Capital, Market Demand Analysis, Regression Modeling.

Abstract: This paper analyzes the determinants of salary levels in the modern labor market and focuses on the product

manager position. Based on over 10,000 job postings from four major cities in China, this paper investigates

the impact of education, work experience, company size, and city on salary levels. Through data

preprocessing, dummy variable transformation, and multiple linear regression analysis, the study identifies

clear patterns: salary levels increase with higher educational attainment and longer work experience;

first-tier cities offer significantly higher wages than lower-tier ones; and larger enterprises generally provide

better compensation structures. The city and work experience variables explain the most variance of salary

differences, followed by education and company size. In addition, the analysis reveals that there are

structural wage advantages for the bachelor's degree, three to five years of experience, large company

employment, and city locations in Shenzhen and Shanghai, respectively. The study shows that salary

formation is layered and not linear. Both theoretical and practical implications are provided. For employers,

the results can guide employers to improve their hiring and retaining strategies. For job seekers, the results

provide evidence based on their employment plans.

1 INTRODUCTION

In the digital age, the changing demand in the market

creates new opportunities and challenges for the

enterprise every day. Companies face the challenge

of fulfilling customers' needs more effectively and

staying up to date with the technological world. The

role of the product manager (PM) is of vital

importance since the manager connects technology,

business, and the user, making sure that the product

follows the market and the company's needs. As

digital transformation accelerates, product managers

are expected to go beyond functional delivery - they

are now responsible for leading cross-functional

teams, shaping product vision, and making strategic

decisions in an uncertain environment. Luchs, Swan,

and Griffin (2022) describe this shift as a form of

form of 'perceptual shaping' where product managers

interpret weak signals from different stakeholders

and markets to proactively guide product innovation.

This redefinition of the product manager's role

positions them as key coordinators of digital value

creation (Luchs et al., 2022).

The role of PMs has broadened due to the rapid

pace of change in the Internet industry. Historically,

PMs' focus on requirement management and

coordination has widened to encompass data

analysis, business strategy, and constant

improvement. According to McKinsey, the reliance

on data for both decision-making and the need to be

customer-centric have greatly extended the role of

the PM and made it a requisite in all . This is an

important realization for PMs that they must possess

an even broader skillset, one that transcends

technical skills to include strategy. And it is this

realization that also compels organizations to expect

PMs not only to help ensure a product fits the

market, but also to drive innovation and long-term

business success (McKinsey, 2017).

Persistent frictions exist in the supply and

demand dynamics of the product manager (PM)

labour market. Job seekers often face challenges

interpreting job descriptions that lack clarity or

contain conflicting requirements. Conversely,

employers find it difficult to select job candidates

with the right skill sets and experiences. Humburg

and van der Velden have in discrete choice

experiments shown that employers prefer CV

attributes that signal high levels of

occupation-specific human capital, e.g. relevant

work experience and that the job seeker's education

and job tasks. Yet, employers' preferences differ and

Wu, Z.

Analysis of Market Demand and Skills of Product Manager.

DOI: 10.5220/0013860400004719

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 2nd International Conference on E-commerce and Modern Logistics (ICEML 2025), pages 757-765

ISBN: 978-989-758-775-7

757

generic job postings might impair the clear

specification of skills required. The resulting

misalignment between job seekers and employers'

preferences and job seekers' lack of information on

employers' preferences in the PM labour market

constitutes an information friction (Humburg et al.,

2015).

Moreover, the variation in compensation due to

factors such as job type, working environment, and

individual background makes it more challenging for

both job seekers and recruiters to plan effective

career paths and recruitment strategies. Gazi

conducted an empirical study on industrial workers

and found that job satisfaction—which is influenced

by factors including experience, job roles, and

workplace conditions—has a significant impact on

employee behavior and performance. While higher

compensation may correlate with enhanced

satisfaction and productivity, this relationship is also

shaped by the structure and scale of the employing

organization. For job seekers, especially in technical

roles like product managers, understanding these

variables is essential for making informed decisions;

for organizations, they offer insights for building

more effective incentive strategies (Gazi et al.,

2024).

Against this backdrop, data-based recruitment

seems to be a panacea for finding prospective

employees among job seekers. Job posting data and

analytics help companies refine job postings and

identify potential job candidates. Peoplebox said that

mining data to attract employees will help

companies fill job vacancies more efficiently and

hire better talent through job seekers' data and

analytical results to fulfill job roles. For product

managers, this approach helps identify

high-frequency skill requirements and map out

targeted learning paths to remain competitive in the

job market (Peoplebox, 2024).

In response to these practical needs, this study

systematically analyses the recruitment data for

product manager positions on the BOSS Direct

platform. By extracting and processing more than

10,000 job postings from 17 major cities, this study

quantitatively examines the impact of four core

variables: city, education level, work experience and

company size on salary outcomes. Through

structured preprocessing and regression modelling,

this study constructs a data-driven framework that

reveals the valuation logic behind product manager

positions in the digital age.

This study aims to explore the impact of four

dimensions, namely education level, work

experience, city level, and company size, on the

compensation level of product managers, and

identify key patterns and thresholds that affect career

development. Specifically, the study will analyze the

salary differences at different educational levels

(from undergraduate to master's), identify key career

turning points such as 3-5 years of work experience,

and evaluate the relationship between salary

differences between city level and company size.

The research results will provide data support for job

seekers to plan their career development paths, while

helping companies develop evidence-based

recruitment and compensation strategies, thereby

optimizing the matching efficiency of talent supply

and demand in the rapidly developing labor market

of the digital age.

2 LITERATURE REVIEW

2.1 Core Concepts of Market Demand

Salary level is an important indicator in the labor

market that reflects the value of positions and the

intensity of human resource demand. What is

implied behind it is the real market demand for

different types of talent. In the modern recruitment

market, enterprises' demand for talent is no longer a

single-dimensional match but a systematic choice

based on multiple factors such as comprehensive

quality, regional adaptation, job ability, and

experience accumulation. At this level, salary is a

holistic indicator that reflects not only individual

attributes but also certain structural characteristics of

the market. That is, can always find four basic

variables, i.e., city, education, working experience,

and company size, which have obvious impacts on

salary. These four aspects are the basis of analysis of

this study, attempt to empirically verify their impacts

on salary and construct a quantitative model to

evaluate the valuation of talents in a market-oriented

environment.

As a spatial unit with highly concentrated

economic resources and opportunities, cities

naturally play a decisive role in job distribution and

salary setting. First-tier cities usually provide higher

job salaries due to their high industrial density, large

number of positions, and fast development pace.

Still, they also correspond to higher competition

thresholds and ability requirements. In comparison,

second- and third-tier cities offer relatively lower

salary levels and job density, but feature lower entry

barriers and less competition, making them attractive

to junior professionals. These differences reflect the

layered structure of urban employment markets.

ICEML 2025 - International Conference on E-commerce and Modern Logistics

758

Therefore, the salary differences between different

cities reflect both the heterogeneity of regional

economic structures and the dynamic game between

job demand and talent supply. Recent empirical

studies confirm this logic, for instance, urban wage

premium research based on Chinese data shows that

larger cities provide significantly higher wages due

to agglomeration effects and industrial density (Liu

et al., 2024). In addition, income disparity across

cities is also shaped by labor market segmentation,

which creates persistent structural barriers between

different classes of jobs and regions (U.S. Census

Bureau, 2023).

As an important indicator for measuring

individual human capital investment, academic

qualifications have always been one of the core

dimensions for enterprises to screen talent. However,

as the market pays more attention to practical ability

and mastery of technical tools, academic

qualifications are no longer the only variable that

determines salary. Especially in cross-border and

complex positions such as product managers, the

"basic skills" and "basic operational capabilities"

required for the position have gradually become

important bases for companies to evaluate whether

talents are competent. Therefore, companies not only

pay attention to academic background, but also tend

to cultivate basic capabilities with job adaptability to

meet their market-oriented development needs.

Worakitjanukul conducted an empirical analysis

of a large sample of data from different regions in

Thailand and pointed out that education level and

work experience have a significant positive impact

on salary level changes, especially in cities, where

workers with higher education and richer experience

receive higher remuneration (Worakitjanukul, 2018).

This study provides an important empirical basis for

this article to explore the role of education and

experience on salary.

Working years represent the accumulation of

experience and the improvement of job proficiency.

When recruiting, companies usually set clear

experience thresholds, such as "3-5 years as a

turning point", to determine whether candidates have

the potential to move from execution to

management. From the perspective of salary

structure, years of experience are often positively

correlated with salary, but whether its marginal

effect continues to increase still needs to be analyzed

in combination with job type and company structure.

Company size also plays an important role in the

recruitment and compensation system. AmorServ

pointed out in its research report that large

enterprises tend to establish more systematic job

classification systems and salary standards due to

their abundant funds and clear management levels.

In contrast, small and medium-sized enterprises

focus more on immediate performance and

short-term returns (AmorServ, 2024). This research

result verifies the hypothetical logic of "the larger

the company size, the higher the salary" proposed in

this article, and highlights the deep impact of the

company's organizational structure on the

formulation of compensation policies.

In summary, city, education, years of work, and

company size, as the four core factors affecting

salary, jointly construct an evaluation system for

talent value in the labor market. Understanding the

inherent logic and interactive relationship of these

variables is of great significance for revealing the

talent supply and demand structure and

compensation mechanism.

2.2 The Role of Market Demand

Analysis

In a recruitment market with highly asymmetric

information, market demand analysis is not only a

tool for understanding employment trends but also a

bridge to link job supply and talent resources. For

employers, accurately grasping the changing trends

of the market's requirements for job capabilities,

education levels, and experience will help optimize

job descriptions, adjust salary strategies, and

improve recruitment efficiency. Especially in highly

competitive positions such as product managers,

companies can make more targeted talent selection

decisions by analyzing talent portraits through

data-based means.

From the employer's internal operations

perspective, companies also establish targeted

capability training systems and salary incentive

structures through demand analysis. As a typical

capability-oriented position, the salary of product

managers is not only a reflection of labor income but

also an evaluation tool for their performance and

potential. Therefore, companies often set up a more

targeted salary structure based on "actual work

performance" and "position adaptation degree",

thereby effectively stimulating the initiative and

growth of talents. Through the feedback of this

incentive mechanism, companies can promote the

continuous growth of high-performance positions

and optimize the overall human resource allocation.

In recent years, AI has been increasingly used in

human resource management to improve recruitment

accuracy and organisational efficiency.Akhter,

Bhattacharjee and Hasan state that AI systems help

Analysis of Market Demand and Skills of Product Manager

759

to screen candidates, match skills to job

requirements and reduce human bias in recruitment

decisions. They further highlight that the use of AI is

particularly effective in recruitment and performance

assessment, where data-driven systems have begun

to replace intuition-based decision-making, thus

aligning with the broader goal of market demand

analysis in optimising talent allocation (Akhter et al.,

2024).

Meanwhile, for job seekers, market demand

analysis has a clear career guidance function. By

understanding the job density and salary levels in

different cities, individuals can plan their career

paths more reasonably; by mastering the relationship

between education and salary, targeted human

capital investment can be made; and understanding

the size and development stage of different

companies will help job seekers weigh the

relationship between short-term returns and

long-term development and make choices that are

more in line with personal development goals.

At the macro level, market demand analysis

provides real and dynamic data support for

policymakers such as governments and educational

institutions. On the one hand, it can assist in

optimizing the educational structure and alleviating

the mismatch between talent training and market

demand; on the other hand, it can also serve as an

important basis for regional human resource

planning and promote the balance of employment

structure between regions.

It is in this context that this study analyzes actual

recruitment data, quantifies the key variables that

affect salary levels, and proposes targeted

suggestions based on job attributes, hoping to build a

data-driven talent matching path between theory and

practice.

3 METHODOLOGY

3.1 Modelling

To study the impact of different cities, education,

work experience, and company size on salary levels,

this paper constructs the following multiple linear

regression model:

εγ

βββ

++++

Controls

CityScaleYears

EducationSalaryAverage

432

(1)

In this model, Average Salary is the dependent

variable, indicating the median monthly salary (unit:

K/K/month). The explanatory variables are defined

as Table 1.

Table 1: Variable settings.

Variable Definition

Education

Minimum education requirement for

the position (e.g., college, bachelor's,

master's, no requirement). The

reference group is "bachelor's

degree".

Years

Required work experience (e.g., no

experience, 1–3 years, 4–5 years, 6–

10 years). The reference group is "1–3

years".

Scale

Company size (10, 60, 300, 700,

5000, 10000 employees, etc.). The

reference group is "300 people".

City

Company size (10, 60, 300, 700,

5000, 10000 employees, etc.). The

reference group is "300 people".

Controls

Job categories (e.g., technical,

marketing, management), used to

control structural effects from

different industries.

The error term, representing

unexplained variation not captured by

the model.

The model aims to evaluate the marginal impact

of the above core variables on the salary level. The

categorical variables, such as education level and

city level, are converted into 0/1 binary variables

through dummy variables. The values of the

reference group are all 0, and the values of the

remaining variables are 1 for "belonging to this

category" and 0 for "not belonging to this category".

The reference group is selected (for example, Beijing

is the reference group for the city, and undergraduate

degree is the reference group for education level),

and the remaining variables are converted into

dummy variables and added to the model. Since the

job salary is continuously distributed, ordinary least

squares (OLS) is used for regression modeling. In

addition, some control variables (such as job type,

etc.) are also considered in the model setting to

reduce the omitted variable bias.

3.2 Data Source

The data in this study comes from the BOSS direct

recruitment data. From April to May 2024, more

than 10,000 product manager jobs postings in 17

ICEML 2025 - International Conference on E-commerce and Modern Logistics

760

first- or second-line cities in China（such as Beijing,

Shanghai, Guangzhou, Shenzhen, Chengdu,

Hangzhou, etc.）are scraped. The variables include

job title, company name, salary range, education

requirement, experience requirement, skill

keywords, company size, job benefits, and location,

etc.

This paper designed a customized data collection

script in Python （ Selenium + EdgeDriver） to

simulate the browsing behavior of users and scrape

the front-end content through XPath. After data

mining, several cleaning steps were implemented by

us removing the HTML tags and other unstructured

characters； dealing with missing values by mean

imputation or deletion, parsing salary range to

extract the minimum and maximum value by regular

expression, then computing the median salary as one

indicator and applying the jieba word segmentation

package to extract high-frequency keywords as the

skill and benefits indicators. Additionally, text fields

such as education and city were converted into

dummy variables, and company size and experience

fields were standardized to ensure consistency for

modeling.

To prepare the dataset for regression analysis,

must make some transformations since the two

categorical variables are not numeric, so converted

them into dummy variables. In this transformation,

“bachelor's degree” was used as the reference

group for education, and “Beijing” for the city.

This paper converted each remaining category into a

column taking values 0 or 1. By doing this, added

qualitative variables into a quantitative model, which

enhanced the statistical robustness and

interpretability of our regression analysis.

4 R ESULT ANA LYSIS

4.1 Descriptive Statistical Analysis

This study focuses on the four core variables of

education, city, work experience, and company size,

and conducts descriptive statistics and variance

analysis on the differences in job salaries (ANOVA

tests are significant, p < 0.001), revealing the key

influence of various factors in the workplace

structure on salary formation.

In terms of education, salary levels increase with

the improvement of education. Masters (19.89K) and

bachelor's degrees (18.31K) are significantly higher

than those with no requirements (14.63K) and

college degrees (13.64K). Among them, bachelor's

degrees are 4.67K higher than college degrees (p <

0.001), and master's degrees are 6.25K higher than

college degrees (p < 0.001), with significant

differences. However, there is no significant

difference between master's and bachelor's degrees

(p = 0.36).

Additionally, the salary distribution for the

master's group shows the greatest variability, with a

standard deviation of approximately 9.04   K,

indicating a broad range of job levels. In contrast,

the junior college group exhibits the most

concentrated distribution (standard deviation 5.90 

K), reflecting a more homogeneous salary structure.

These patterns are clearly illustrated in Figure 1,

where the wider interquartile range and more

numerous outliers for the master's and undergraduate

groups underscore their broader salary dispersion.

Figure 1: Salary distribution by education level.

At the city level, the overall salary distribution

shows a clear gradient. Shenzhen (21.14K),

Shanghai (20.98K), and Beijing (20.96K) are in the

first echelon, significantly higher than other cities (p

< 0.001), and the difference between the three is not

significant (p = 1.0). Hangzhou (17.37K) and

Nanjing (16.03K) are in the middle layer;

Guangzhou (15.85K), Chengdu (14.76K), and

Chongqing (12.75K) have relatively low salaries,

and the largest gap between Shenzhen and

Chongqing is 8.39K (+65.8%). It is worth noting that

although Guangzhou is a first-tier city, its salary

performance is relatively weak. Regionally, it

presents the characteristics of "tiered distribution

within the Yangtze River Delta, Chengdu in the

Chengdu-Chongqing region is higher than

Chongqing"; in terms of salary fluctuations,

Shanghai has the largest standard deviation (9.05K)

and Nanjing is the most concentrated (4.95K),

suggesting that most jobs in the city offer salaries

Analysis of Market Demand and Skills of Product Manager

761

closer to the media. These disparities in both central

tendency and dispersion are reflected in the box plot

shown in Figure 2, where outliers and interquartile

ranges further illustrate city-specific wage

characteristics.

Figure 2: Salary distribution by city.

In terms of work experience, salary increases

significantly with years of experience, from 15.67K

for no experience to 26.83K for 6-10 years of

experience, with a cumulative increase of more than

71%. Tukey test shows that salary growth is mainly

concentrated in the two stages of "1-3 years" to "4-5

years" (+5.58K, p < 0.001) and "4-5 years" to "6-10

years" (+4.78K, p = 0.043), while the difference

between "1-3 years" and "no experience" is not

significant (p = 0.604), indicating that the marginal

effect of initial experience accumulation is limited.

These patterns are clearly illustrated in Figure 3,

where the 6–10 years group shows both the highest

median salary and the widest distribution, with

numerous outliers indicating high-paying senior

roles. In contrast, the 1–3 years group presents the

most concentrated salary structure (standard

deviation: 6.64 K), while the 6–10 years group

exhibits the greatest variation (SD: 17.87   K).

Overall, 3-5 years is the key turning point for a

salary jump, and senior positions have greater salary

elasticity.

Figure 3: Salary distribution by work experience.

In terms of company size, salaries vary

significantly between companies of different sizes,

with an average increase of nearly 10K from 13.44K

for companies with 10 employees to 23.59K for

companies with 10,000 employees (F = 34.57, p <

0.0001). The scale of 300 employees is a key

dividing point, with salaries significantly higher than

those of companies with 60 or fewer employees (p <

0.05); salaries of super-large companies with more

than 10,000 employees are significantly higher than

those of all other groups (p < 0.001), reflecting their

strong capacity to attract talent and offer higher-level

positions.

Figure 4: Tukey HSD multiple comparisons of salary by

company size.

ICEML 2025 - International Conference on E-commerce and Modern Logistics

762

In contrast, the salaries of micro-enterprises, with

10 employees are significantly lower. Interestingly,

the analysis also reveals nonlinear patterns — for

example, companies with 0 employees (e.g.,

self-employed entrepreneurs) report higher average

salaries than some small firms, possibly due to

equity incentives or profit-sharing mechanisms.

Meanwhile, there are no significant differences

between 700 and 5,000 people, and between 300 and

700 people, indicating that the salary structure of

some medium and large companies tends to be stable,

as shown in Figure 4.

In summary, education, city, work experience,

and company size all significantly affect salary

levels, showing hierarchical distribution and phased

premium characteristics. Among those with a

bachelor's degree, 3-5 years of experience,

companies with 300 or more employees, and

first-tier cities, job salaries are significantly better

than other groups, highlighting their structural

advantages in the current workplace.

4.2 Regression Results Analysis

To further explore the impact of variables such as

education, city, years of work, and company size on

salary, this study constructed a multivariate linear

regression model and employed ordinary least

squares (OLS) estimation. The results are presented

in Table 2, Table 3, and Table 4. The model fits

well, and the explanatory power and statistical

significance of key variables are high. Under the

condition of controlling other variables, years of

work have a significant positive impact on salary

(coefficient is 1.61, p < 0.001), indicating that for

every two years of work experience, the average

salary will increase by about 1.6K, verifying the

market premium effect of experience accumulation.

The company size variable also shows statistical

significance (coefficient is 0.00065, p < 0.001), but

the impact is low, indicating that the larger the scale,

the slightly higher the salary, but the actual increase

is smaller, indicating that although the scale is

significant, the actual explanatory power is limited.

In terms of educational background, with

bachelor's degree as the reference group, the

regression results show that the salary of those with

a master's degree is slightly higher than that of those

with a bachelor's degree (coefficient is 0.2991, p =

0.0078), while the salary of those with a junior high

school degree and a college degree is significantly

lower than that of those with a bachelor's degree

(respectively –1.3826, p < 0.001; –0.2391, p ≈

0.93), reflecting that higher education has certain

salary advantages in the workplace, but the

difference between a master's degree and a

bachelor's degree is not significant after controlling

other factors.

Table 2: Impact of work experience, company size, and city on salary.

Coeffi

cient

Stan

dard

Erro

Stat

P-va

lue

Lowe

r 95%

Upp

er 95%

Low

er Limit

95%

to 95%

Interce

12.405

0.40

30.9

7604

6.2E

-174

11..6

2048

13.1

9132

11.6

2048

13.1

9132

Years

1.5623

0.08

9803

17.3

9804

1.88

E-63

1.386

279

1.73

8502

1.38

6279

1.73

8502

Scale

0.0005

5.45

E-05

10.4

0437

9.31

E-25

0.000

0.00

0673

0.00

046

0.00

0673

Shang

hai

3.0187

0.46

4786

6.49

4848

1.03

E-10

2.107

226

3.93

2.10

7226

3.93

Nanjin

-1.590

0.69

021

-2.3

0435

0.02

-2.94

406

-0.2

3692

-2.9

4406

-0.2

3692

Chong

qing

-4.366

0.49

7151

-8.7

8362

3.21

E-18

-5.34

174

-3.3

9182

-5.3

4174

-3.3

9182

Hangz

hou

-0.335

0.48

5295

-0.6

9045

0.48

999

-1.28

678

0.61

6638

-1.2

8678

0.61

6638

Shenz

hen

3.2038

0.49

5118

6.47

0842

1.21

E-10

2.232

861

4.17

4806

2.23

2861

4.17

4806

Guang

zhou

-1.793

0.55

9216

-3.2

0709

0.00

1361

-2.89

013

-0.6

9678

-2.8

9013

-0.6

9678

The city effect is also quite significant. Taking

Beijing as the benchmark, Shenzhen (coefficient =

3.0833, p < 0.001) and Shanghai (coefficient =

3.0181, p < 0.001) are significantly higher than

Analysis of Market Demand and Skills of Product Manager

763

Beijing, indicating that the salary premium of

first-tier cities is still significant after controlling

other variables; while the coefficients of Nanjing,

Chongqing, Chengdu and other places are all

negative and p < 0.05, showing their salary

disadvantages compared with Beijing, further

highlighting the impact of regional development

imbalance on salary levels.

From the perspective of the overall regression

model, the p-values of each variable are generally

significant, indicating that education, experience,

scale, and city all play an important role in

explaining salary changes. Among them, work

experience and city effects contribute the most,

while education and company scale have relatively

weak effects after controlling other factors, showing

the salary structure characteristics of "strong

experience, strong region, weak education, and small

scale".

Table 3. Impact of work experience, company size, and education on salary

Coefficient

Standard

Error

T Stat P-value

Lower

95%

Upper

95%

Lower

Limit 95%

Up to 95%

Intercept 12.67714 0.340618 37.21805 1.6E-233 12.00916 13.34512 12.00916 13.34512

Years 1.565106 0.095349 16.41446 4.46E-57 1.378118 1.752094 1.378118 1.752094

Scale 0.000607 5.71E-05 10.62638 9.92E-26 0.000495 0.000719 0.000495 0.000719

Junior -3.18256 0.489342 -6.50375 9.75E-11 -4.1422 -2.22291 -4.4122 -2.22291

Master 2.93691 0.872881 3.364618 0.00078 1.225113 4.648706 1.225113 4.648706

uirement

-1.23915 1.180281 -1.04988 0.293895 -3.55378 1.075487 -3.55378 1.075487

Table 4: Baseline model with work experience and company size

Coeffi

cient

Stan

dard

Erro

Stat

P-va

lue

Low

er 95%

Upp

er 95%

Low

er Limit

95%

to 95%

Inter

12.192

0.32

2377

37.8

2058

1.7E

-239

11.5

6028

12.8

247

11.5

6028

12.8

247

Year

1.6058

0.09

5239

16.8

6131

6.02

E-60

1.41

9078

1.79

2621

1.41

9078

1.79

2621

Scale

0.0006

5.74

E-05

11.3

798

3.66

E-29

0.00

0541

0.00

0766

0.00

0541

0.00

0766

5 CONCLUSIONS

In the context of the ever-changing development of

the digital economy, the position of product manager

plays the role of a bridge connecting user needs,

technology implementation, and business strategy,

and becomes an important engine for enterprises to

promote product innovation and respond to the

market quickly. This paper takes the product

manager position as the research object, based on the

large-scale recruitment data of BOSS Direct

Recruitment Platform, builds a regression framework

that centers on four critical variables: city, education,

work experience, and company size, and

systematically explores the influence path and

interaction mechanism of these variables on the

salary level.

The results of the study show that city and work

experience are the most significant factors affecting

salary differences, with 3–5 years of experience

often marking a distinct inflection point in career

progression; academic qualifications still have

structural value, and those holding at least a

bachelor's degree tend to enjoy a notable wage

premium in the market; and company size is

reflected as a systematic advantage of the salary

system and career stability. In large companies in

Tier 1 cities, job seekers with a mid-to-high level of

education and key experience thresholds are more

likely to have significant salary returns and career

advancement opportunities. These findings not only

highlight the layered nature of the digital labor

market, but also reflect the deep game of human

ICEML 2025 - International Conference on E-commerce and Modern Logistics

764

capital, regional resources and organisational

systems behind the pay structure.

Further, the study not only provides quantitative

evidence, but also responds to the structural

problems of 'mismatch' and 'information asymmetry'

in the real market. When job seekers face the

dilemmas of skill prioritisation and ambiguous job

descriptions, it is often difficult for them to precisely

interpret what employers are truly seeking; while

enterprises may also miss out on highly matching

talents due to unclear job definitions and irrationally

set experience thresholds. Through quantitative

analysis of recruitment data, this paper constructs a

coherent reasoning path linking market expectations

to salary outcomes, which provides practical support

for improving the accuracy of job seekers' career

planning and the efficiency of corporate recruitment.

In terms of methodology, this paper integrates

web crawlers, natural language processing, and

statistical modelling techniques, effectively

translates a large corpus of unstructured recruitment

text into measurable analytical features. This

technical path not only breaks through the

dependence of traditional job analysis on

questionnaires, interviews, and other methods, but

also holds strong potential for adaptation across roles

and platforms. In the future, if combined with

in-depth semantic mining, industry classification

models or time series modelling, it is expected to

further portray the trend of skills evolution and

dynamic changes in job demand, thus realizing

data-driven career ecology research in the true sense.

From a broader perspective, the structural laws

revealed in this paper are not only applicable to

product manager positions but also provide

theoretical references for understanding the value

composition of high-knowledge and

high-technology-intensive occupations in the context

of the new era. In today's world, where companies

are constantly pursuing agile innovation and

high-quality growth, job pricing is no longer just a

linear function of experience and education, but a

synergistic game between corporate strategic goals,

organisational structure, and talent ecosystem. Salary

is often a reflection of the organisational

expectations of a position and its criticality in the

value chain.

Therefore, the significance of this paper is that it

not only responds to the individual's confusion about

the reality of career development paths but also

provides a basis for organisations to optimise the

allocation of human resources and build a scientific

and reasonable job system. More importantly, it

shows how human resource management and career

planning can achieve more efficient docking and

synergy under the data-driven logic, and promote the

employment market from 'empirical judgement' to

'intelligent matching', to better serve the dynamic

adaptation of people and jobs in the digital society.

The dynamic matching between people and jobs in

the digital society will be better served.

REFERENCES

Akhter, F., Bhattacharjee, A., & Hasan, A., 2024.

Application of artificial intelligence in human resource

management: A Bangladesh perspective. Munich

Personal RePEc Archive (MPRA).

AmorServ, 2024. 7 factors to consider when determining

the salary of developers in 2024.

Gazi, M. A. I., Yusof, M. F., Islam, M. A., Amin, M. B., &

Senathirajah, A. R. S., 2024. Analyzing the impact of

employee job satisfaction on their job behavior in the

industrial setting: An analysis from the perspective of

job performance. Journal of Open Innovation:

Technology, Market, and Complexity, 10(4), 100427.

Humburg, M., & van der Velden, R., 2015. Skills and the

graduate recruitment process: Evidence from two

discrete choice experiments. Economics of Education

Review, 49, 24-41.

Liu, Y., Zhang, M., & Fang, H., 2024. Trade liberalization,

city size, and urban wage premium: Evidence from

China. Nature Humanities and Social Sciences

Communications, 11(1), 26.

Luchs, M. G., Swan, S., & Griffin, A., 2022. How product

managers use senseshaping to drive the front-end of

innovation. Research-Technology Management, 65(1),

30-39.

McKinsey & Company, 2017. Product managers for the

digital world.

https://www.mckinsey.com/capabilities/mckinsey-digit

al/our-insights/product-managers-for-the-digital-world

Peoplebox, 2024. The ultimate guide to data driven

recruitment strategies.

U.S. Census Bureau, 2023. Labor market segmentation

and the distribution of income (CES-WP-23-41).

Worakitjanukul, P., 2018. The effect of education and

experience on wages: The case study of Thailand.

American Journal of Industrial and Business

Management, 8, 159-172.

Analysis of Market Demand and Skills of Product Manager

765