Lead Quality Score by Traffic Source: How Local Businesses Should Weight Leads Before Calculating True ROAS

The CPL Trap That Kills Local Ad Budgets

Here is a scenario that plays out constantly in local business ad accounts: paid social delivers leads at $18 each, paid search delivers them at $54 each. The obvious call seems to be to shift budget toward social.

That call is usually wrong — and it is wrong because cost-per-lead is not a performance metric; it is a volume metric dressed up as one.

If the $18 social lead closes at 5% and the $54 search lead closes at 30%, the math flips completely. You are paying three times as much per lead on search and getting six times the close rate — meaning search is actually producing revenue far more efficiently. Unweighted CPL comparisons hide this reality every single time.

The fix is a Lead Quality Score: a simple weighting layer you apply to each traffic source before you calculate true ROAS.

Why Traffic Source Predicts Close Rate

Not all leads arrive with the same intent. The channel a lead comes through is one of the strongest signals you have about where they are in the buying decision.

• Paid search (Google/Bing): The user typed a specific query. Intent is explicit and often transactional — they are already problem-aware and actively looking for a solution. This tends to produce the highest close rates of any paid channel for local service businesses.
• Google Business Profile (GBP): Calls and direction requests from GBP represent people who have already narrowed their search to your category, often in your neighborhood. Close rates here are typically strong, comparable to or sometimes exceeding branded paid search.
• Organic search: Varies widely depending on the keyword. Informational content drives low-intent traffic; transactional or local-intent pages can drive close rates that rival paid search — with no direct media cost.
• Paid social (Meta/Instagram): Ads interrupt people who were not searching for your service. Even well-targeted campaigns are reaching audiences earlier in the funnel. Higher volume, lower average intent, lower close rates — not always, but as a structural baseline.

None of this means paid social is bad. It means it plays a different role, and measuring it on the same unweighted CPL scale as paid search is an apples-to-hammers comparison. (For more on how the wrong campaign objective distorts social performance, see our article Wrong Meta Objective? Here's What It Costs You.)

Building a Lead Quality Weighting Model

You do not need expensive attribution software to apply this framework. You need three numbers per channel:

1. Volume: How many leads did this source produce this month? 2. Close rate: Of those leads, what percentage became paying customers? (Pull this from your CRM or, if you lack one, a simple spreadsheet tracking lead source → job status.) 3. Average job value: What does a closed job from this source typically pay?

From those three numbers, you build a Quality-Adjusted Revenue per Lead figure for each source:

> Quality-Adjusted Revenue per Lead = Close Rate × Average Job Value

Then your true cost efficiency metric becomes:

> True Cost per Acquired Customer = Ad Spend ÷ (Leads × Close Rate)

And true ROAS:

> True ROAS = (Leads × Close Rate × Average Job Value) ÷ Ad Spend

This is the number that should govern budget decisions — not raw CPL.

A Worked Model: HVAC Company, Four Sources

The numbers below are an illustrative model — not a cited study — built on close-rate ranges that are broadly consistent with what local service businesses report across industries. Use this as a template, not a benchmark.

| Source | Monthly Leads | Illustrative Close Rate | Closed Jobs | Avg Job Value | Revenue Generated | Ad Spend | True ROAS | |---|---|---|---|---|---|---|---| | Paid Search | 40 | 28% | 11.2 | $1,800 | $20,160 | $2,160 | 9.3x | | GBP | 25 | 35% | 8.75 | $1,800 | $15,750 | $0 (GMB mgmt only) | — | | Paid Social | 110 | 6% | 6.6 | $1,800 | $11,880 | $1,980 | 6.0x | | Organic | 30 | 22% | 6.6 | $1,800 | $11,880 | $0 | — |

The naive CPL read: Paid social at ~$18/lead beats paid search at ~$54/lead. Cut search, scale social.

The quality-weighted read: Paid search produces true ROAS of ~9.3x vs. paid social at ~6.0x. The right move is the opposite — protect search budget, optimize social targeting or shift social to a top-of-funnel awareness role rather than a direct-conversion role.

This is also why smart bidding algorithms need enough closed-job signals — not just lead signals — to optimize correctly. If you are feeding Google only raw lead volume, it will optimize toward the same cheap-but-soft leads your CPL dashboard celebrates. See our article Smart Bidding: How Many Conversions Do You Actually Need? for the conversion volume thresholds that make automated bidding reliable.

How to Assign Close Rates When You Don't Have Clean Data

Most local businesses do not have a CRM with perfectly tagged lead sources. Here is a practical starting point:

Step 1 — Tag everything at intake. When someone calls or fills out a form, ask: 'How did you find us?' Log it. Even imperfect self-reported data is better than nothing.

Step 2 — Use call tracking numbers per source. Assign a unique tracking number to your website (for organic/paid), your GBP listing, and any social landing pages. Services like CallRail make this straightforward and will attribute closed calls by source.

Step 3 — Apply provisional weights while you collect data. As a rough starting assumption — not a citation — paid search and GBP leads for local services tend to close materially better than paid social leads. If you have no data yet, you can use a conservative placeholder like 3:1 (search vs. social close rate) and refine it monthly as real data accumulates.

Step 4 — Review quality weights monthly, not quarterly. Seasonality, offer changes, and competitor moves all shift close rates. A weight you set in January can be stale by March.

Where This Connects to Impression Share and Conversion Share

One reason businesses under-invest in paid search specifically is that they look at impression share metrics and see gaps — and interpret those gaps as waste. In reality, impression share gaps on high-intent queries often represent missed revenue, not saved spend.

Our article Impression Share vs. Conversion Share in Google Ads walks through why conversion share — not impression share — is the metric that connects back to the true ROAS model above. The two frameworks are complementary: quality-weight your leads to find which channel deserves investment, then use conversion share analysis to find where you are leaving revenue on the table within that channel.

The broader principle: every efficiency metric in local advertising has to trace back to revenue, not volume. Leads are not revenue. Impressions are not revenue. Even clicks are not revenue. The chain is: spend → leads → qualified leads → closed jobs → revenue — and quality-weighting is how you keep that chain visible.

Start With One Month of Real Data

You do not need to overhaul your entire reporting stack to run this model. Pull last month's numbers:

• Leads by source (ads dashboard + call log)
• Jobs closed by source (your invoices or CRM)
• Revenue by source
• Ad spend by source

Drop them into the formula above. In most cases, the quality-weighted true ROAS picture looks meaningfully different from the CPL picture — and the budget reallocations it suggests pay for the analysis many times over.

If you want a second set of eyes on your channel mix before you move budget around, we can run this model against your actual numbers. [Book a no-pressure call with the Nika Spark team](https://nikaspark.com/contact) — we will tell you where your weighted ROAS story diverges from your CPL story, and what to do about it.