How CometFuel Cut Proposal Labor 80–90% and Scaled Leads 4X

How CometFuel Cut Upwork Labor Costs by 80–90% and Increased Leads 250–430% with GigRadar

‍

Client Overview

Jared Spiewak is the founder of CometFuel and has been active on Upwork since 2015. Over time, he built a serious internal engine for Upwork acquisition — one that produced reliable lead volume but demanded constant operational input to keep it running smoothly.

The Starting Point

Before adopting GigRadar, CometFuel’s Upwork motion was already “working” — but it was expensive in time and maintenance.

Jared described spending about $3,700/month “between our on-platform costs and somebody who was working about 20 hours a week to vet and submit proposals,” which resulted in “about 15 leads a month.”
He also made an important distinction: the issue wasn’t that the economics were broken.
‍

“We were happy with our cost per lead — it was just a lot of manual work to maintain.”

That line matters, because it explains the real bottleneck: not lead quality or strategy — but the operational burden required to keep the system alive week after week.

The Challenge

Jared’s description gives a clear picture of what it takes to be consistently competitive on Upwork at an agency level — and why it becomes operationally heavy:

1) Speed and responsiveness

To avoid missing opportunities, the team had to be ready the moment new jobs appeared. Jared said they had to build systems “to be notified of when new job postings went live.”

This implies a constant “monitoring” requirement — because without fast response, even a good proposal strategy can underperform simply due to timing.

2) Visibility beyond Upwork’s native analytics

Jared shared that they “built out custom tracking to have analytics beyond what Upwork provides.”

That’s an insight into how mature teams operate on Upwork: they don’t rely only on platform-level stats — they build extra tracking to understand what’s actually driving lead flow and outcomes.

3) Proposal and Q&A infrastructure… in documents

Jared said they had “extensive Word docs for different types of cover letters and Q&A responses.”

In practice, that’s a sign the team already had repeatable messaging patterns — but managing and deploying them still required manual effort.

So the “real problem” was the workload: multiple supporting systems, running in parallel, held together by people-hours.
‍

Why GigRadar

GigRadar appealed to CometFuel specifically because it could automate what was already proven.
‍

“What originally attracted us to GigRadar was automating our existing processes.” 

It was a way to preserve the parts that worked — and remove the grind needed to keep them running.
‍

Implementation: Getting to a Working Setup Fast

Jared emphasized speed-to-value once the team understood the tool:

“Once we figured out how to, like, use the platform, it only took about a day for us to get the scanners to work how we wanted them.”

Two important takeaways from that sentence:

• There’s a short learning curve (“once we figured out how to… use the platform”).
  After that, configuration to match their needs was fast (“a day… to work how we wanted”).
  
  

Then, once scanners were activated, the process became continuous:
‍

“Once we got our scanners activated, we were active from there.”

That suggests a shift from “manually maintained” operations to a more ongoing, automated workflow — where the default state is “always running,” not “only running when humans are online.”
‍

Results

After adopting GigRadar, Jared reported improvements across the three most sensitive levers on Upwork: labor, platform spend, and lead volume.

1) Labor cost reduction

The automations of the platform dropped our cost of labor by about 80–90% a month. This is especially meaningful given the baseline: they were paying for “somebody… working about 20 hours a week to vet and submit proposals.” So the gain here isn’t abstract — it’s the operational time that used to be required.

2) On-platform cost reduction

Our on-platform costs dropped by about 25% a month. This indicates savings not just in labor, but also in the direct costs tied to operating on Upwork (as he framed it, part of the original $3,700/month).

3) Lead volume growth

Our leads per month also increased by about 250–430% depending on the month. That range is an important nuance — it shows variability by month, but consistently far above baseline once their setup was running.
‍

ROI: Fast Payback

Jared’s timeline for payback was explicit:
‍

“It was less than 30 days before we had a return on our annual subscription cost.”

Given he also said it took “about a day” to configure scanners the way they wanted, this paints a very clear story arc:

• short setup time,
• scanners activated,
• meaningful lead lift + cost reductions,
• ROI inside a month.
  ‍

Beyond the Tool: Strategy Expansion via Community

Jared also attributed part of the upside to something that’s not purely software — the community.
‍‍

“In addition to the software, the community has also helped me understand just how much more we could actually get out of our profile and our Upwork strategies.”

That line adds an “insight layer” that’s very relevant to Upwork realities: automation can scale activity, but profile and strategy improvements can increase the yield of that activity (how much value you get from the same market).

And he’s not describing it as “nice to have.” He ties it to future growth expectations:
‍‍

“Despite all the growth that we’ve seen so far, we can probably still at least double what we’re currently getting out of Upwork.”

Key Takeaway

CometFuel didn’t come to GigRadar because they couldn’t make Upwork work. They came because a working Upwork system required too much human effort to maintain at speed.

Jared’s story captures the core advantage clearly:

• they were already spending heavily and generating leads (“$3,700/month… netting us about 15 leads a month”),
• but the engine demanded constant manual upkeep (“a lot of manual work to maintain”),
• GigRadar automated their existing workflow (“automating our existing processes”),
• reduced labor and platform costs (“80–90%” labor, “25%” on-platform),
• and scaled lead volume dramatically (“250–430%”),
• with payback “less than 30 days,”
• plus ongoing upside from improving profile and strategy via community learning.

‍