Here is a scenario that plays out in agencies every day: an account manager sets up a campaign with well-researched keywords, writes solid ad copy, configures bidding, and moves on to the next client. Six weeks later, a glance at the search terms report reveals the campaign has been spending on queries like "google ads certification free," "how to run ads myself," and "PPC agency jobs hiring." None of these will ever convert.
This isn't an edge case. Across the industry, agencies routinely waste 15-30% of client ad spend on irrelevant search terms. Not because they don't know better, but because search term management falls through the cracks in the daily rush of account management.
Why Search Terms Matter More Than Keywords
Your keyword list represents your intent. Your search terms report represents reality.
With broad match (and even phrase match in its current expanded form), Google shows your ads for queries that it considers semantically related to your keywords. The gap between what you think you're bidding on and what you're actually paying for can be enormous.
A keyword like "enterprise project management software" might trigger ads for:
- "free project management tools" -- wrong intent, will never convert at enterprise pricing
- "project management certification" -- educational intent, not buying intent
- "asana vs monday" -- competitor comparison, might work but might not be your client's positioning
- "project management jobs" -- completely irrelevant
Each of these clicks costs money. And in competitive B2B verticals, each click might cost $15-40 (check your cost per click). Ten irrelevant clicks a day across a campaign is $150-400 in daily waste -- $4,500-12,000 per month from a single campaign.
The "Set It and Forget It" Trap with Broad Match
Google has been aggressively pushing broad match + Smart Bidding as the default approach. And in fairness, it works better than it used to. Smart Bidding can learn to avoid some low-quality queries over time.
But "better than it used to" is not the same as "handles it for you." Smart Bidding optimizes toward your conversion goal, but it has no way to know that a query is fundamentally irrelevant to your client's business. It will happily bid on "free project management templates" if someone who clicked that query happened to convert once.
The agencies that rely entirely on Smart Bidding to handle query relevance are the ones bleeding the most spend. Automated bidding adjusts bids -- it doesn't fix targeting. That's your job.
The Negative Keyword Workflow
Search term management isn't complicated, but it needs to be systematic. Here's a workflow that works at scale:
Step 1: Review
Pull search terms reports at a regular cadence. For high-spend accounts (over $10K/month), review weekly. For smaller accounts, biweekly is sufficient.
Filter the report to show:
- High spend, zero conversions -- These are your biggest waste candidates
- High impressions, low CTR -- Signals that the query doesn't match the ad, dragging down Quality Score
- Queries with clicks but no engagement (high bounce rate if linked to Analytics)
Sort by cost descending. The most expensive irrelevant terms should be addressed first.
Step 2: Categorize
Not every unfamiliar search term is a negative. Categorize each term into one of four buckets:
- Clearly irrelevant -- Add as a negative immediately. "Jobs," "free," "DIY," "how to" (in most commercial contexts), competitor names you don't want to bid on.
- Possibly relevant but not converting -- Give it more data before deciding. Flag it for review next cycle.
- Relevant but wrong match -- The query is good, but it's triggering the wrong ad group or campaign. Consider adding it as an exact match keyword in the right location and as a negative in the wrong one.
- New opportunity -- A query you didn't think of that's converting well. Add it as a keyword to capture it intentionally. Use our keyword mixer to quickly build out variations with different match types.
Step 3: Exclude
When adding negatives, be deliberate about match type and level:
- Exact match negatives for specific queries you want to block without risk of over-excluding
- Phrase match negatives for patterns (e.g., adding "free" as a phrase match negative blocks "free project management," "project management free trial," etc.)
- Broad match negatives sparingly and carefully -- they can block more than you intend
Add negatives at the appropriate level:
- Ad group level for negatives specific to one ad group's targeting
- Campaign level for negatives that apply across the campaign
- Account level (via negative keyword lists) for terms that should never trigger any ad
Building Shared Negative Lists Across Clients
One of the highest-leverage activities in agency search term management is maintaining shared negative keyword lists. Instead of discovering that "salary" is irrelevant in each client's account separately, build it once and apply everywhere.
Recommended shared lists:
Universal negatives -- terms almost never relevant for commercial intent:
- Job-related: jobs, hiring, salary, careers, resume, interview
- Educational: how to, tutorial, course, certification, training, PDF, template
- Free-seeking: free, cheap, open source, DIY
- Informational: what is, definition, meaning, wiki, reddit
Vertical-specific negatives:
- E-commerce: wholesale, bulk, used, refurbished, repair, manual (varies by client)
- B2B SaaS: free, open source, comparison chart, alternatives to [competitor], internship
- Local services: DIY, how to fix, parts, [distant city names]
- Legal: pro bono, free consultation (if client doesn't offer it), law school
Important caveats:
- Review shared lists quarterly. A term that was irrelevant six months ago might become relevant as a client's business changes.
- Don't add terms to shared lists without checking that they're genuinely universal. "Free" is a negative for most clients but not for a freemium SaaS product running a free-trial campaign.
- Document what's in each list and why. When a new account manager inherits a client, they shouldn't have to reverse-engineer why certain terms are blocked.
Real Examples of Wasteful Search Terms
These are drawn from actual account audits (details anonymized):
B2B software client ($25K/month spend):
- "excel project tracker template" -- $340/month, 0 conversions. User wants a free spreadsheet, not enterprise software.
- "project management internship" -- $180/month. Job seekers, not software buyers.
- "trello free plan" -- $220/month. Competitor's free tier seekers are not enterprise prospects.
E-commerce client ($15K/month spend):
- "[product name] recall" -- $95/month. People checking if a product was recalled are not buying it.
- "[product] repair parts" -- $260/month. Repair intent, not purchase intent.
- "cheapest [product category]" -- $410/month, 0 conversions. Price-sensitive bargain hunters didn't convert on premium products.
Local service client ($8K/month spend):
- "[service] jobs near me" -- $175/month. Job seekers, not customers.
- "how to [do the service] yourself" -- $310/month. DIY intent, will never hire a professional.
- "[service] license requirements" -- $85/month. People researching how to start a competing business.
In these three accounts alone, over $2,000/month was going to obviously irrelevant queries. That's $24,000/year in wasted client spend -- and these were actively managed accounts at a competent agency. The terms just hadn't been caught because search term review wasn't happening at a consistent cadence.
Automating the Review Cadence
The biggest failure mode in search term management isn't technique -- it's consistency. Here's how to make sure reviews actually happen:
Set calendar blocks. Search term review should be a recurring calendar event, not something you do "when you have time." For a portfolio of 15-20 accounts, block 2-3 hours weekly.
Use scripts or automation to surface the worst offenders. A simple Google Ads script can flag search terms that have spent over a threshold with zero conversions and send a daily email. This turns a manual review process into a triage process.
Track the impact. Maintain a simple log of negatives added and the monthly spend they were consuming. This does two things: it quantifies the value of the work (useful for client reporting), and it motivates the team to keep doing it (seeing "$3,200/month saved" is more satisfying than "added 47 negatives").
Build it into your SOP. Search term review should be step 3 or 4 in your weekly account optimization checklist, right after checking pacing and performance. If it's an afterthought, it won't happen.
The Bottom Line
Search term management is unglamorous work. Nobody gets excited about adding negative keywords. But across an agency portfolio, it's consistently one of the highest-ROI activities available. Reducing waste by 15-20% effectively increases your client's budget by the same amount -- without asking them for more money.
The agencies that do this well don't have a secret technique. They just have a process they actually follow, every week, for every account. Build the workflow, set the cadence, and protect the time. Your clients' budgets depend on it.