Upwork reported Monday that active clients fell 10% to 717,000 while its marketplace take rate rose 110 basis points to 20.0%.
If you run a store on Ecommerce Paradise principles, you probably pay freelancers or virtual assistants for listings, ad creative, support or bookkeeping. A platform that is losing buyers and has told investors it will keep adding ways to charge them is a platform whose prices you should expect to drift up, so this is a hiring-cost story, not a stock story.
Below: the numbers with baselines, why the clients are leaving, what higher fees could do to a typical store’s contractor budget, and five moves for this week. If you are newer to the model, my guide on what high-ticket dropshipping is explains why staffing costs decide how much margin you keep.
Need a long-term assistant instead of renting one by the hour? OnlineJobs.ph describes itself as the job board for virtual workers in the Philippines, which makes it a natural place to hire the assistant who handles support, orders and listings every day. Hire a virtual assistant on OnlineJobs.ph →
Upwork Q3 Results: 717,000 Clients, $914M GSV, 20% Take Rate
According to Upwork’s third-quarter release, issued November 2, revenue was $185.6 million, down 8% from $201.7 million a year earlier. Marketplace revenue fell 9% to $159.3 million from $174.6 million. Enterprise revenue slipped 3% to $26.3 million from $27.2 million.
Gross services volume (GSV), the total dollars clients spend through the platform, was $914.1 million against $1,016.7 million, a 10.1% drop. Active clients were 717,000 against 794,000. Each remaining client spent more: GSV per active client was $5,418, up 8%.
Profit moved the other way. GAAP net income was $31.8 million against $29.3 million, and adjusted EBITDA was $63.8 million, a 34% margin versus 30% a year ago. Free cash flow fell to $53.9 million from $69.4 million.
On the earnings call, per the Motley Fool transcript, the marketplace take rate (Upwork’s cut of GSV) reached 20.0%, up 110 basis points from a year earlier. CFO Erica Gessert said “the bias for take rate in the marketplace is up,” while adding that the growing share of its enterprise business, Lifted, would moderate the rate sequentially in the fourth quarter.
Management listed the ways it plans to earn more from the marketplace: Connects (the credits freelancers spend to bid on jobs), variable talent-side fees, a wider range of client-side fees, and dynamic pricing. It did not say which fees, how large, or when. The release itself mentions no marketplace fee change.
Fee rules today sit on Upwork’s help pages. Upwork’s freelancer fee article says the service fee runs from 0% to 15% per contract, is shown before a freelancer accepts the work, and typically sits at 10% on enterprise contracts. That page covers the freelancer side only, so I cannot give you today’s client-side rates from it.
Guidance for the fourth quarter is revenue of $169 million to $177 million, and $742 million to $750 million for the full year. StockStory’s analysis, published on Yahoo Finance, put the Q4 midpoint at $173 million, a 12.8% year-over-year decline and slightly under the $174.7 million consensus. Shares rose about 14.6% to $11.16 after the report, per the same article, because third-quarter revenue and profit beat expectations of $180.2 million and $0.31 adjusted EPS.
Put the client number in plain terms: 77,000 fewer active clients than a year ago, roughly one in ten. Revenue fell less than clients did because the survivors spend more and pay more of each dollar to the platform, which is exactly what a rising take rate on a smaller base looks like.
Two growth lines stand out. Business Plus, the small-business product, grew GSV 132% and now makes up 8% of the total, per the transcript. GSV from freelancers with AI skills rose 55% to $112 million, and Upwork said work started through its MCP server, an interface for AI tools, grew more than 150% month over month in October.
Why Upwork Is Shedding Clients and Raising Its Take
The report did not arrive in a calm year. Before the results, 24/7 Wall St. reported the stock was down 52% in 2026, after management cut full-year revenue guidance twice, from $835 million to $850 million in February to $730 million to $750 million. A May restructuring cut about 24% of the workforce and targeted $70 million in annualized expense savings, per the same article.
On the call, management gave its own reasons for the client decline. Low-value work is rotating off the platform and will keep doing so through the first half of 2027. Google search changes are pressuring how Upwork acquires digital clients. Gessert said the company expects total growth rates “to continue to be headwinded” in the first half of 2027, with some improvement later.
That low-value work is the kind store owners post most often: rewriting product descriptions, resizing images, uploading listings, pulling lists of suppliers. AI tools now do much of it in minutes, which is the pattern 24/7 Wall St. described when it said automation of low-complexity work accelerated in the second quarter.
The counterpoint is real. Upwork is profitable, spend per client is rising, and the SMB product is growing fast. The same 24/7 Wall St. piece noted the company had beaten EPS estimates four straight quarters while the stock fell after three of them, so a rally on one print proves little. I read the quarter as a company shrinking its customer count on purpose while squeezing more from each one, and management’s phrase “quality over quantity” fits that reading.
There is precedent for the fee side. When I covered Etsy’s Q3, where its take rate hit 26.2%, the pattern was the same: a marketplace with slowing buyer growth leans on fees.
It is not universal. In my write-up of eBay’s Q3, the take rate fell 20 basis points. My note on Amazon’s seller services and ads results shows a third platform monetizing through ads instead of headline fees.
What Rising Upwork Fees Do to a Store’s Freelancer Budget
My read: expect fee creep, not fee cuts. A platform with fewer clients, a profit target and a stated plan to add client-side fees and dynamic pricing has one direction to go. Freelancer-side fees get priced into the rates freelancers quote you, and client-side fees land on your invoice, so you pay on both ends even if you only ever see one of them.
Here is hypothetical math, not a reported figure. Say your store spends $4,000 a month across four Upwork contractors. Every extra point of client-side fee costs you $40 a month, or $480 a year. A three-point change costs $1,440 a year. Real money, but not the part that worries me.
The bigger risk is the talent pool. If good freelancers see more of each contract go to the platform, the strongest ones move to direct contracts, agencies or other boards, and the bench you hire from gets thinner without any fee ever appearing on your invoice. I cannot prove that from one earnings call. It is the scenario I would plan around, and the signs to watch are quote prices creeping up and fewer qualified applicants per post.
Scenarios, with thresholds I would use. If your quotes and platform fees rise less than 5% over the next two quarters, change nothing. If they rise 5% to 10%, or you see qualified applicants per job post drop by about a third, move your recurring roles to direct hires and keep the marketplace for projects. Above 10%, treat the platform as a project tool only. These are my thresholds, not industry standards, and you should adjust them to your own margins.
Dynamic pricing is the unknown. Management did not describe it, so nobody outside the company can say whether it means surge-style fees for urgent hires or something milder. I would not budget for it yet. I would make sure my staffing does not depend on one marketplace when it arrives.
Second, the work worth paying a human for is changing. When Upwork says low-value tasks are leaving the platform, it is describing the jobs store owners should already be handing to AI. Draft your descriptions, alt text and FAQ answers with Claude, then pay a person for judgment: supplier calls, phone sales, customer service, ad management. I laid out the process in my post on AI-ready SOPs for ecommerce workflows.
Third, here is a decision rule. If a role runs more than about 20 hours a week for more than three months, price a dedicated assistant against the marketplace total before you renew. My walkthrough on hiring through OnlineJobs.ph covers how to post and screen. Project work can go elsewhere, and my breakdown of Fiverr alternatives shows where. Marketplace hourly still wins for a one-off task you will not repeat.
Fourth, vetting is your job on any platform. A platform that is fighting for clients will not screen harder for you. I covered the extreme case in North Korean IT workers hitting 70-plus US firms, and the lesson applies to a VA with access to your Shopify admin: verify identity, limit permissions, and use a separate login for every person.
Finally, staffing is where owner dependence builds. If your store only works when you personally coordinate five freelancers across two platforms, you do not own a system. I wrote about the exit-value cost of that in my piece on owner dependence. When the hiring, vetting and tooling are the part you do not want to run, that is what my turnkey done-for-you build exists to take off your plate, and my full-service management runs $2,000 a month.
Want to compare your staffing stack against other store owners who are cutting freelancer costs right now? Join me inside the Ecommerce Paradise Academy and start the 7-day free trial (paid after the trial). Join the Skool community →
Five Moves to Make Before Your Next Freelancer Hire
Here are five moves, in the order I would do them this week:
- Pull 90 days of contractor spend by platform. Add up what each person cost, including the platform fees on your invoices, and flag every role above 20 hours a week. Those are your candidates for a direct hire. The rest of your cost-cutting list is in my post on reducing operating costs as your store grows.
- Move recurring roles off hourly marketplace contracts. Post customer service, order processing and listing maintenance as long-term roles. Compare the all-in cost of a marketplace freelancer on Upwork with the cost of a long-term assistant, and confirm current pricing on each platform’s own pages before you decide.
- Keep marketplaces for fixed-scope projects. A theme tweak, a product video or a one-off audit fits Fiverr or Upwork well. Set a fixed price, define the deliverable in writing, and release payment on delivery.
- Strip the low-value tasks out of every contractor’s list. Anything repetitive and text-based goes to an AI draft first, with a human editing. Track what the saved hours were worth so you can cut the invoice, not just feel faster.
- Track hours and pay efficiently. Use Hubstaff for time tracking on hourly roles. Pay overseas contractors through Wise or Payoneer after checking fees.
If you would like a second set of eyes on your staffing costs and which roles to hire directly, that is exactly what my 1-on-1 coaching covers. For a wider comparison of sites, my list of the best sites for remote jobs compares costs and fit across eight boards and marketplaces.
Frequently Asked Questions
Did Upwork announce a fee increase?
No. Its release mentions no marketplace fee change. Management said on the call that it is monetizing through Connects, variable talent-side fees, a wider range of client-side fees and dynamic pricing, without giving rates or dates.
What does Upwork charge freelancers today?
Per Upwork’s help page, the freelancer service fee ranges from 0% to 15% per contract and is fixed once the contract starts. Enterprise contracts typically carry 10%.
Is Upwork in financial trouble?
Nothing in the report says so. It posted GAAP net income of $31.8 million and beat revenue and profit expectations, though revenue and clients are shrinking and 24/7 Wall St. noted the stock fell 52% this year before the report.
Upwork or OnlineJobs.ph for a virtual assistant?
For a long-term daily role, I would start with a dedicated assistant and read my guide to hiring on OnlineJobs.ph. For a defined project with an end date, a marketplace contract with a fixed price is usually the cleaner fit.
Which tasks should I stop paying people to do?
Anything repetitive and text-based, such as description rewrites, alt text and first-draft FAQs. Pay for work that needs judgment or a phone call. If support is the role you are weighing, my post on Gorgias pricing shows what a help desk costs per plan, so you can compare it with a person’s monthly cost.
How do I keep contractor costs from eating margin on a high-ticket store?
Track cost per order for every role, and keep your product mix in niches with healthy margins. If you are still choosing a niche, start with my free high-ticket niches list. The longer list of 150-plus niches goes deeper.
Would you rather not run hiring, vetting and tooling yourself? My team builds and runs the high-ticket store so your time goes to supplier relationships and growth. See the turnkey done-for-you service →
Watch your invoices for the next two quarters, keep a second source for every critical role, and move recurring work to people you hire directly. Subscribe to the YouTube channel for daily breakdowns. More breaking news coming through the day.
Related Articles
If this was useful, these go deeper:
- How to Hire on OnlineJobs.ph in 2026: Post Free First, Pay Second
- Fiverr Alternatives: 5 Places to Hire Freelancers Compared by Fees and Fit (2026)
- Deel Alternatives: 5 Global Hiring and Payment Tools Compared by Price (2026)
- Hubstaff Review 2026: Worth It for Managing Remote VAs?
- North Korean IT Workers Hit 70+ US Firms: Vet Your VAs
Trevor Fenner is an ecommerce entrepreneur and world traveler from Seattle, Washington, and the founder of Ecommerce Paradise, a platform focused on helping entrepreneurs build and scale profitable high-ticket ecommerce and dropshipping businesses. With over a decade of hands-on experience, Trevor specializes in high-ticket dropshipping strategy, niche and product selection, supplier recruiting and onboarding, Google & Bing Shopping ads, ecommerce SEO, and systems-driven automation and scaling. Through Ecommerce Paradise, he provides free education via in-depth guides like How to Start High-Ticket Dropshipping, advanced training through the High-Ticket Dropshipping Masterclass, and fully done-for-you turnkey ecommerce services for entrepreneurs who want a faster, more hands-off path to growth. Trevor is known for emphasizing sustainable, real-world ecommerce models over hype-driven tactics, helping store owners build scalable, sellable, and location-independent brands.
He maintains homes in Seattle and Bali, spends most of the year overseas, and usually returns to Seattle for about a month each summer.
