Methodology
Transparency about our data collection and analysis.
Data Collection
Data is collected via an anonymous online questionnaire. Participants are employed SEOs in agencies and service companies.
Activities over Job Titles
Unlike other salary studies, we don't ask for job titles. Titles like "Manager" or "Head of Department" are not comparable across companies — a "Manager" at Company A may have completely different responsibilities than at Company B.
Instead, we capture 16 specific activities and how much time participants spend on them (scale 0–5). This makes the data comparable.
Collected Dimensions
Beyond the 16 activities, we capture additional factors that influence salary.
Salary & Compensation
Annual salary (gross), bonus, benefits, salary increase vs. prior year, desired salary.
16 Activities
How much time is spent on each activity (scale 0–5). From keyword research and content to technical development.
Experience & Profile
Professional experience, total work experience, age, gender, education.
Company
Type (startup, mid-size, enterprise), industry, internationalisation, tenure.
Leadership & Responsibility
Technical and disciplinary leadership, team size, budget responsibility.
AI Usage
How agency SEOs use artificial intelligence — from content creation to data analysis.
Statistical Analysis
Median instead of Average
We use the median instead of the arithmetic mean. The median is robust against outliers — a few very high or low salaries don't skew the picture. At the median, exactly 50% earn more and 50% earn less.
Quartiles (Q1, Q3)
In addition to the median, we show the quartiles:
- Q1 (25%): 25% earn less, 75% earn more
- Median (50%): The middle
- Q3 (75%): 75% earn less, 25% earn more
Minimum Data Points
Statistics are only calculated when at least 5 data points are available. Categories with fewer than 30 data points are marked as "low confidence" and visually distinguished — their statistical significance is limited.
Data Cleaning
Not all submitted data is usable. The following filters ensure data quality:
Limitations
Self-Selection
Participants decide whether to take part. People with extreme salaries may be under- or over-represented.
Self-Reporting
Salaries are not verified. We rely on honest responses.
Time Period
Older data does not necessarily reflect current salaries. The cohort filter helps to separate time periods.
Open Data
The aggregated statistics are publicly available. Raw data is not published to protect the anonymity of participants.
Participate now