Bad leads are mostly a measurement problem dressed up as a marketing one. Here is the simple system we install for every retainer client.
Step 1: Install call tracking
CallRail, CallSource, Marchex, whatever fits your budget. Get a separate tracking number for every paid channel. One for Google Ads, one for organic, one for Yelp, one for billboards if you run them.
The cost is around 60 to 200 dollars a month depending on volume. The information you get back pays for itself the first month.
Step 2: Record your calls
This is the step everyone skips. Listen to a sample of inbound calls every week. You will discover three things in your first round of listening that will change how you spend.
- Some of your "leads" are not leads. They are people calling to ask about jobs, sell you something, or wrong-number you. If 30 percent of your inbound from Google Ads is non-customer, your real cost per lead is 43 percent higher than your dashboard says.
- Some of your leads are being lost on the call. You will hear your office person mishandle a price question, say "I'll have someone call you back" and then nobody calls back, or just sound annoyed. We have caught dozens of accounts where the marketing was working fine and the close rate was dragged into the dirt by the answering process.
- You will hear what people actually call about. Those become FAQ entries on your site, talking points your sales people memorize, and angles for your next round of ads.
Step 3: Tag every lead by quality
Do not get fancy. Use four buckets.
- Qualified (real customer with real budget in your service area)
- Low-intent (price shopper or researcher)
- Wrong fit (out of service area or out of scope)
- Waste (telemarketer, wrong number, spam)
Tag them in your CRM or even in a Google Sheet for the first month.
Step 4: Reverse-engineer the bad ones
Pull your last 30 days of "wrong fit" leads. Where did they come from? Usually you find one specific keyword, one ad placement, or one form on your site that is generating most of them.
Last month I found a client whose top 10 wasted leads all came from a single Google Ads keyword that had been on broad match for two years. Killing that keyword saved 1,100 dollars in monthly spend with no impact on real leads.
Step 5: Optimize the form, not the ads
People spend hours tuning ad copy and ignore the form on the landing page. The form is where lead quality is actually decided. Three changes that consistently lift quality without killing volume:
- Add a budget-range question with options that start at your minimum job size. Even just including a "1,000 to 5,000 dollars" floor option screens out tire-kickers.
- Add a project-timeline question. "When do you need this done?" with options including "just researching" tells you who is actually ready.
- Require a phone number. People who give you a real phone number are 3 to 5 times more likely to book than form-only fills.
Step 6: Feed qualified-lead data back to your ad platforms
This is the one that matters most.
Google and Meta will both let you mark conversions as "qualified" via the API. If you tell Google which leads were actually good, it will optimize for more of those.
Most service businesses are still optimizing for "form fill" and getting the cheapest, lowest-intent leads they can possibly attract. Quality-as-a-conversion is the single biggest lever in modern paid ads, and almost nobody small uses it.
What to expect
Run this system for 60 days and your cost per qualified lead will drop somewhere between 25 and 50 percent.
Not because you spent less. Because you stopped paying for the bad ones.