Are “SEO Leads” Real Leads? How to Judge Lead Quality

TL;DR: Most arguments between a business and its marketing provider come down to nobody having defined “lead.” Four definitions are in common use and they produce wildly different counts from the same month. Pick one, score every lead against it, and track lead-to-job rate by channel. That single discipline resolves more disputes than any reporting upgrade.

Table of Contents

• Four definitions of “lead” and why they conflict
• A qualification score you can actually maintain
• Tracking lead-to-job rate by channel
• Why organic leads close differently
• What to renegotiate when quality drops
• The one report you should ask for

Four Definitions of “Lead” and Why They Conflict

Same month, same business, four honest counts.

Definition 1 — Any contact event. Every form submission, every tracked call, every chat. Count: 147. This is what most dashboards report, and it includes the vendor calling about your business insurance.

Definition 2 — Any contact from a potential customer. Filters out spam, wrong numbers, applicants, and vendors. Count: 88.

Definition 3 — Any qualified potential customer. In your service area, needing a service you offer, with a real timeline. Count: 51.

Definition 4 — Any qualified customer you actually quoted. Contact made, need confirmed, estimate delivered. Count: 34.

Now watch the argument. The agency reports 147 leads. The owner counted 34 estimates. Both are telling the truth. Both think the other is either lying or incompetent.

Definition 3 is the right one to standardize on for judging marketing performance. Definition 4 mixes in your sales response speed, which is your responsibility, not the channel’s. Definition 1 measures nothing. Definition 2 is a reasonable second-best if you can’t score qualification consistently.

A Qualification Score You Can Actually Maintain

Elaborate scoring models die within a month. This one survives because it takes about eight seconds per lead and can be done by whoever answers the phone.

Score each lead 0 or 1 on five items:

4–5 = qualified. Counts as a lead for performance purposes.
2–3 = marginal. Track separately; often a nurture problem, not a lead problem.
0–1 = not a lead. Excluded from the count, and disputed if you’re paying per lead.

Log it in whatever you already use — CRM field, spreadsheet column, a tag in your call tracking. What matters is that the same standard is applied every day. A slightly wrong definition applied consistently produces useful data. A perfect definition applied intermittently produces noise.

Add one required field: source. Organic, GBP, paid search, LSA, referral, repeat customer, lead vendor. Without source, none of the rest of this works.

Tracking Lead-to-Job Rate by Channel

Once you have qualified counts by source, build one table monthly. This is the most valuable report in a service business and almost nobody maintains it.

Three things become obvious the first time you build it, and they’re usually the same three:

1. Referrals close at three to four times the rate of everything else. Almost nobody spends proportionally on generating them.
2. Lead vendors deliver volume at poor conversion. Which can still be worth it — but at 17%, that channel is buying you busywork more than jobs.
3. Paid search cost per job is often the highest while looking the most controllable, because the dashboard shows cheap clicks and hides the conversion gap.

Give it three months of data before drawing conclusions. One month of a service business is mostly weather.

Why Organic Leads Close Differently

Organic and GBP leads consistently close better than paid or vendor leads for the same business. Three reasons, all structural.

They found you, not an ad. Someone who read your page and then called has self-selected. Someone who tapped the first thing on the screen has not.

They’ve usually seen your reviews. Anyone arriving through the map pack has passed your rating and read a couple of reviews. A meaningful amount of selling has already happened.

They arrive later in the process. Organic research queries skew toward people comparing, reading, and deciding. Paid captures earlier, broader intent — including a lot of price checking.

The practical implication: don’t hold channels to the same close rate. A 17% close on vendor leads and a 40% close on organic may both be normal. Judge each channel against its own trend and its own cost per job, not against each other’s conversion rate.

What to Renegotiate When Quality Drops

If qualified-lead rate falls while total contacts hold steady, something specific changed. Diagnose before you renegotiate.

If it’s a lead vendor: ask what changed about their sourcing or territory. Bring your dispute log. Renegotiate on the definition first, price second. A tighter definition is worth more than a discount.

If it’s paid search: the usual culprits are broad match expansion, a new location target, or automated bidding chasing conversions you configured badly. Check search terms first — you’ll often find you’re paying for queries you’d never choose.

If it’s organic: unusual, and worth investigating rather than accepting. Typical causes are ranking for informational queries instead of commercial ones, or a page starting to rank in a city you don’t actually serve.

If it’s every channel at once: it’s probably not the channels. Check whether your intake changed — new person answering, new phone system, a form that quietly stopped emailing. We find broken forms more often than we find broken campaigns.

The One Report You Should Ask For

If you take one thing from this: ask your marketing provider for a monthly report with qualified leads and lead-to-job rate by channel, using your definition of qualified.

Some will say they can’t produce it without CRM access. Give them access. Some will resist because it makes their weaker channels visible. That resistance is itself the answer to a question you were probably already asking.

Quick Recap

• Four definitions of “lead” are in use; standardize on “qualified potential customer” for judging marketing.
• Score five binary items per lead — service area, service match, timeline, authority, reachability. Four or five means qualified.
• Always record source. Without it, nothing else works.
• Track lead-to-job rate and cost per job by channel monthly, and give it three months before deciding anything.
• Judge each channel against its own trend, not against another channel’s close rate.

Frequently Asked Questions

What’s a good lead-to-job rate?
For home services, 30–45% on organic and GBP leads, 20–30% on paid, 15–25% on shared vendor leads, and 60%+ on referrals are reasonable benchmarks. Your own trend matters more than any benchmark.

Should I pay for leads that don’t convert?
For advertising, yes — you’re buying access, not outcomes. For pay-per-lead vendors, only if they meet the agreed definition. That’s exactly why the definition needs to be written down.

How do I count a repeat customer who found me on Google?
Separately. Repeat and referral business shouldn’t be credited to marketing channels, or you’ll overstate performance and make bad budget decisions.

My provider reports far more leads than I remember. Who’s wrong?
Probably neither. They’re almost certainly counting contact events and you’re counting estimates. Align on Definition 3 and recount one month together.

Can I automate lead scoring?
Partially. Service area and service match can be automated from form data or call routing. Timeline and authority need a human. Eight seconds of human judgment per lead is cheap for the accuracy it buys.