Campaigns are the core of how Lancer automates your job applications. A well-structured campaign ensures:
You target the right type of jobs.
You avoid wasting connects on unsuitable opportunities.
You maximize your ROI (Return on Connect Spent).
Although Lancer lets you create unlimited campaigns, we recommend starting with one campaign, refining it, and then expanding.
Campaign Layout
When creating or managing a campaign:
Left side: Job feed – all jobs that match your current filters.
Right side: Filter section – where you fine-tune targeting.
Top bar: Forecast – shows how many jobs your campaign can monitor monthly.
Example: Adding “Shopify” as a keyword might return 10,000 jobs per month. This does not mean Lancer will apply to all of them. Instead, it means Lancer will monitor these jobs, and only after applying contextual filtering (via your Knowledge Base), proposals will be sent to suitable ones.
The monthly number is an average over the trailing 12 months, not a raw count of last month alone — a single month is too noisy a sample on its own. For low-volume niches it carries one decimal (e.g. 0.3/mo), so a real-but-rare match — roughly one job a quarter — reads as rare rather than rounding down to a 0 that looks dead. Hover the forecast to see the raw 12-month job total the average is built from, shown alongside it both here and in the campaign list.
The numbers beside each filter option (experience level, payment type, verified payment/phone, region) use the same unit as the forecast: jobs per month, averaged over the trailing 12 months — not a last-30-days count. So an option's number is directly comparable to the forecast at the top, and applies your other active filters (but not the filter group it sits in, so you can see what each option would add).
Real-Time Job Monitoring
Campaigns monitor newly posted jobs in real time. Proposals are only sent after a campaign is activated.
Lancer does not send proposals retroactively to older posts.
Think of campaigns as a live job monitor with rules that decide what opportunities to apply for.
Building Campaign Filters
Keyword & Query Builder
All of these words: All listed keywords must be included (acts as an “AND” filter).
Any of these words: At least one keyword must appear (“OR” filter).
None of these words: Excludes unwanted keywords.
Exact phrase: Use quotation marks for phrase-matching (“conversion rate optimization”).
Title-only search: Limit filters to only appear in job titles (e.g., “Shopify expert”).
Skill tags (All / Any / don't contain any): Match against the job's Upwork skill tags instead of its title or description — a job tagged Shopify that never uses the word “Shopify” in its text is only reached this way. A term matches a whole word inside a tag (
aimatches “AI Agent Development” but not “Email Communication”); use quotes for multi-word tags, e.g."machine learning".
In Custom query mode, skills: is its own field operator alongside title: — for example skills:shopify, skills:"machine learning", NOT skills:wordpress, or skills:(react OR "next.js"). Quote a multi-word phrase after any field prefix to keep it whole and scoped to that field — title:"google ads" requires the phrase "google ads" in the title, and NOT title:"web design" excludes only that phrase.
When a skill tag matches, Lancer highlights that tag in the job feed the same yellow used for a matched title word, so you can see at a glance why a job showed up.
Searching for C#, C++, F# or .NET: a keyword that contains # or + (like C#, C++, F#) or starts with a dot (like .NET, .js) is matched exactly, so C# no longer matches every job that merely contains a letter C (for example “USB-C”). This applies in the keyword builder, in Custom query mode and in title-only searches. Very old jobs (roughly older than 90 days) may still match the looser way.
Pro Tip:
Keep campaigns under 1,000 jobs/month (12-month average) for better precision. Too many jobs usually means filters are too broad.
Categories
Choose or exclude entire categories (e.g., Sales & Marketing, E‑commerce Development).
Client Requirements
This is one of the most powerful sections for improving campaign quality.
Hire Rate:
Recommended: Filter out clients with <50% hire rate.
Exception: Allow new clients (e.g., fewer than 4 posted jobs).
Payment Verified: Always enable this to avoid wasted proposals.
Location:
Option 1: Target high-value geographies (e.g., US, UK, EU, Canada, Australia).
Option 2: Exclude specific low-value regions.
Picking a country in the Include/Exclude countries dropdown matches every job stored under that country, even if Upwork's data used a different spelling or code for it (e.g. selecting Germany also catches jobs saved as “DEU”) — you don't need to add every variant yourself.
In the Include countries dropdown, choose Select Highest Paying Countries to auto-fill the countries with the strongest client pay data (highest average hourly rate and total spend) instead of picking them one by one.
Spend & Budget Quality:
Target clients who have spent significantly (e.g., $10,000+).
Look at total spend per posted job to find fair-paying clients.
Example: 300 jobs, $3M spent vs 10 jobs, $200 spent.
Average Hourly Rate Paid: Good to identify higher-value clients, but note it only applies for hourly jobs, not fixed-price.
Minimum Reviews: Optional. Leaving it open is fine unless you want only experienced clients.
Client Rating (feedback from freelancers):
Best clients: 4.5+ rating.
Avoid <4.0 — usually difficult clients.
Compensation & Budget
Hourly: Define min/max hourly ranges (e.g., $30+ recommended for skilled freelancers).
Fixed Price: Be flexible with placeholder budgets (e.g., $50 for a $5,000 project). Many clients just insert rough numbers.
Unspecified budgets: Often indicate high-quality clients who are willing to negotiate. Always include them.
Best Practices for Filters
Layer multiple filters together (hire rate + spend + keywords).
Leave some flexibility (don't exclude new clients entirely, but set limits).
Test & iterate — small adjustments can drastically improve campaign quality.
Track your Return on Connect Spent (ROCS) to ensure proposals are targeted at clients who are most likely to hire.
Key Takeaways
Filters give you control over campaign focus.
Don't aim for quantity. Aim for specificity and quality.
Campaign monitoring works only forward (no retroactive proposals).
The Knowledge Base + Filters work together:
Filters = broad targeting.
Knowledge Base = final contextual selection.