App-Based Worker Research

During the first 18 months of implementation, the ordinance covered 92,801 workers working for the five largest network companies. 

• From Quarter 2 2024 to Quarter 2 2025, the number of unique engaged workers grew slightly by 2.8%.
• More than 85% of the workers only worked on one app. On average, app-based workers worked 12 hours per week per app. 

The number of offers completed by the largest network companies grew by 3.2%.

• When comparing the first half of 2024 to first half of 2025, the number of weekly completed offers increased from 197,148 offers to 203,469 offers per week. 

Average worker pay increased significantly compared to pre-ordinance reports and continued increasing from 2024 to 2025.

• The average network company payment, or “base payment,” per engaged hour was $30.12, after subtracting mileage expenses incurred during actively engaged time.
• The average “pay for time online” – a metric that accounts for additional work time and expenses, including time and expenses while logged on but not actively completing tasks – was $15.98 per hour.
• This represents a significant increase from pre-ordinance levels, which, according to outside analysis, were as low as $3.17 per hour.
• Workers’ pay for time online increased from 2024 to 2025, from $15.29/hour to $16.29/hour during the first half of each year.
• Pay for time online varied substantially based on network company, ranging from $12.09 to $25.37 per hour.

Tips and bonuses made up a much smaller share of earnings than pre-ordinance estimates, indicating that earnings became more predictable and transparent.

• Worker pay was comprised primarily of base payment, rather than tips and bonuses. Greater than 85% of workers’ compensation was base payment, around 14% was from tips, and only about 0.2% was bonuses/incentives from network companies.
• This is a substantial shift from the composition of earnings prior to implementation, when tips were reported to make up as much as 48.8% of earnings.
• When including tips and bonuses, worker earnings averaged $36.36 per engaged hour, and pay for time online was $19.98/hour.

A substantial portion of customer payments were network company fees, and fees increased from 2024 to 2025.

• When paying for orders, nearly 20% of the customer payments were fees paid to network companies. (Network companies have multiple other revenue streams in addition to customer fees, including but not limited to commissions, service fees, and advertising fees paid by third parties such as restaurants, brands, and retailers.)
• When comparing the first half of 2024 to the first half of 2025, there was an increase in the average platform fee percentage from 18.5% to 19.6%.

OLS analyzed the number of completed offers. For the period shown, there was an average of 196,871 offers per week completed by the five largest companies. From Half 1 2024 to Half 1 2025 (excluding the first two weeks of each half), there was a small increase in average weekly offers from 197,148 offers to 203,469 offers (+6,321 offers per week or +3.2%).

OLS assessed two pay metrics: “pay per engaged hour” and “pay for online time per hour.” (Please refer to the report for the calculation methods of both metrics.) Both pay metrics increased significantly compared to pre-ordinance reports and continued increasing from 2024 to 2025.

The average “pay for time online” – a metric that accounts for additional work time and expenses, including time and expenses while logged on but not actively completing tasks –  during the first eighteen months after SMC 8.37 became effective was $15.98 per hour. This represents a significant increase from pre-ordinance levels, which, according to outside analysis, were as low as $3.17 per hour. 

Between Half 1 2024 and Half 1 2025, there was an increase in average “pay for time online” from $15.29/hr to $16.29/hr (+$1.00/hr or +6.5%). When adjusting for inflation for wage-earners in the Seattle Metropolitan area, the average pay for time online hourly pay increased in real terms (in terms of June 2025 $) by +$0.66 (+4.2%) between Half 1 2024 and Half 1 2025 (excluding the first four weeks of each half). 

In the first eighteen months after SMC 8.37 went into effect, the average “pay per engaged hour” was $30.12, which is workers’ hourly base pay subtracting mileage expenses incurred during actively engaged time.

The average pay per engaged hour was comparatively consistent across network companies, however pay for time online per hour varies substantially between network companies, ranging from $12.09 to $25.37 per hour.

OLS analyzed the average network company fee as a percentage of the total customer payment for four network companies. The network company fee is defined as “any amount charged to paying customers in excess of the amount charged for the underlying goods or services provided.” SHRR 240-100.6. When fees are waived or reduced due to a customer’s active subscription, the cost of subscription is not included in this calculation, which only includes fees specifically attached to individual offers.

For the period shown, there was an average platform fee of 19.3% per week. From Half 1 2024 to Half 1 2025 (excluding the first two weeks of each half), there was an increase in average platform fee from 18.5% to 19.6% (+1.1 percentage points).

Question 1: How does your analysis compare to other reports?

OLS’ report is the first to use comprehensive data sets submitted by the large delivery network companies, which include all associated workers and offers covered by the ordinance. Other publications have relied upon more limited, and therefore less reliable, samples from self-reporting workers. For example, data aggregated from applications used by workers to monitor and/or maximize earnings may overrepresent workers who are highly active and experienced users of network company platforms.

In contrast with other studies that ended shortly after SMC 8.37 was implemented, this report is the first to examine an 18-month timeframe post-ordinance. This means that the report can capture a picture over the longer term that other analyses missed. The longer time frame also allowed OLS to compare the first half of 2024 with the first half of 2025, making possible the calculation of broader trends and reducing the concern of small window size and quarter-over-quarter fluctuations.

OLS’ report provides a range of pay statistics that account for workers’ time and expenses, and the report clearly explains how each was calculated. However, some other analyses only consider the time the worker spends actively completing an offer and do not include additional time workers spend on the app, or do not consider mileage and other expenses that app-based workers incur. Readers should carefully compare OLS’ numbers to figures cited in other reports to understand whether they are an apples-to-apples comparison and whether they accurately describe workers’ take-home pay. 

Question 2: I heard the law caused the tips to drop, and isn’t it bad for the workers? What does your report say about workers’ tips?

Our report shows workers’ total pay increased even though some network companies changed their app design to discourage tipping in response to the Ordinance. Our data also shows the ratio of tips decreased post implementation. That’s because base payment increased substantially post implementation and is now greater than 85% of workers’ total compensation. Reliance on base payment makes income more transparent and predictable, rather than chasing tips and promotions which can be unreliable, stressful, subject to customer bias, and vulnerable to bad customer behaviors.

Question 3: How is business faring for large network companies? Is Seattle’s minimum pay standard impacting network company profits?

When the App-Based Worker Minimum Payment Ordinance went into effect, some network companies implemented “Seattle regulatory fees” in response to the law that could have suppressed consumer demand for services. However, our report shows after the initial decline, demand for services stabilized and grew. The network companies’ business volume grew by 3.2% when comparing the first half of 2024 to first half of 2025: The number of completed offers increased from 197,148 to 203,469 per week.

OLS does not collect data on network companies’ profits, but we do have data on customer fees. The on-demand network companies keep almost 20% of customers’ payments in the form of various fees, including “Seattle regulatory fees” that some companies implemented post-ordinance. When comparing the first half of 2024 to the first half of 2025, there was an increase in the average platform fee percentage from 18.5% to 19.6%.

Additionally, network companies may charge commissions or service/advertising fees from third parties such as restaurants or grocery stores. Moreover, a network company may generate revenue outside of individual orders, from, for example: customer subscriptions, advertising products available at retailers, fees charged to retailers for additional services, and/or use of user data.

Question 4: Why is there such a wide range of “pay for time online”?

Variation in time spent online but not actively completing offers (referred to as “available time” in the data reporting framework) appears to be the key factor impacting average hourly pay for time online.

The network company that facilitates pre-scheduled offers had the highest average hourly pay for time online. At the same time, the quantity of uncompensated time spent online differed significantly across the on-demand worker platforms. In other words, how long workers spent logged on and not actively completing an offer depended on the company. Companies that primarily facilitate restaurant deliveries reported more uncompensated online time than the company that primarily facilitates grocery deliveries. Nevertheless, variation was still considerable among the three companies that deliver meals.

Qualitative research is needed to understand workers’ waiting-time behaviors and how platforms’ models shape those behaviors.

Question 5: What are some key questions and options for future research?

Direct worker research (e.g., surveys and focus groups) could help to provide an accurate explanation of the variation in worker’s “pay for time online” across companies. We could examine how different companies or business models impact the length of uncompensated time on apps, the practice of “multi-apping,” and how the minimum pay requirements interact with other provisions of the ordinance such as the “flexibility” rights to accept and reject offers.

Additionally, the report shows a population of workers who enter, exit, and sometimes re-enter the industry. Further qualitative research could help us understand the impacts of the ordinance on these short-term workers vs. longer-term and/or more full-time workers.

Finally, OLS is collecting data from multiple smaller network companies and will conduct a broader analysis. Future reports will also benefit from a longer timeframe that will allow us to understand longer-term impacts of the law.