> ## Content Index
> Fetch the complete content index at: https://cascader.io/blog/llms.txt
> Use this file to discover other available public pages before exploring further.

# Can Google Search deliver medical-opinion leads at Meta’s CPL and the volume I need?
- URL: https://cascader.io/blog/google-search-medical-opinion-leads-meta-cpl-volume/
- Published: 2026-09-21T20:00:44.000Z
- Updated: 2026-09-21T20:00:44.000Z
- Description: Test a proposed Search budget against affordable clicks, the conversion rate needed to match Meta’s CPL, and mature lead quality—without assuming medical purchase data can be sent back to Google.
- Author: Rick Hecker
- Tags: Cascader Answers, #Import 2026-09-20 14:50

Google Search may be able to match Meta’s cost per lead for a medical-opinion service, but that does not mean it can absorb the entire budget you want to move. **The useful question is how many comparable leads Search can supply at an affordable price—and whether enough of them become the outcomes your business needs.**

Before transferring $100,000 a month, work backward from the lead target to the clicks and conversion rate it requires. Then follow those leads far enough to judge quality. If the plan depends on sending medical purchases back to Google to improve bidding later, establish whether that exact event and data route are permitted before making the transfer depend on them.

Here is how to make the decision with a forecast, an uncertain conversion rate, or neither.

## Turn the budget into a requirement

Suppose your comparable Meta CPL is **$50 per initial lead**, and you want to move **$100,000 per month** into Search. These are illustrative numbers, not medical-sector benchmarks.

Spending the full amount at $50 CPL requires **2,000 initial leads a month**:

> Required leads = proposed spend ÷ target CPL  
> $100,000 ÷ $50 = 2,000 leads

First make “lead” the same event on both sides. For example, count one genuine new inquiry per person, apply the same duplicate and spam exclusions, and keep qualified inquiries and purchases as separate later outcomes. Reconcile the channel reports to that definition in your business records; two dashboard columns named “leads” are insufficient. Use a recent, representative comparison period rather than mixing an old Meta low with a current Search estimate.

Next ask what it would cost to obtain enough relevant clicks. Suppose a forecast for the same service, geography and month produces these three alternatives:

| Illustrative monthly forecast | Clicks | Average cost per click (CPC) | Leads needed at $50 CPL | Required click-to-lead rate |
| ----------------------------- | ------ | ---------------------------- | ----------------------- | --------------------------- |
| $25,000 spend                 | 5,000  | $5                           | 500                     | 10%                         |
| $60,000 spend                 | 10,000 | $6                           | 1,200                   | 12%                         |
| $100,000 spend                | 12,500 | $8                           | 2,000                   | 16%                         |

These are alternative total spending levels. Do not add the rows together.

The calculation is:

> Required click-to-lead rate = spend ÷ (forecast clicks × target CPL)

At the $100,000 point, that is $100,000 ÷ (12,500 × $50), or 16%. You would need roughly one initial lead from every six clicks.

Now suppose **8–12%** is the range you consider plausible, pending a Search test. At 12,500 clicks, that range produces **1,000–1,500 leads**, costing **$100–$66.67 each**. Even its optimistic end misses both the $50 CPL and the 2,000-lead requirement.

That gives you a decision: this forecast and that conversion assumption do **not** support the full transfer. The $25,000 point needs 10%, which sits inside the assumed range and makes it a candidate for investigation. It is not yet proven capacity. The $60,000 point only meets the target at the range’s best end.

More available money does not repair this mismatch. In this example, buying the additional clicks raises their average price. You need a better conversion rate, cheaper relevant traffic, a different target, or a smaller transfer. The equation tells you which assumption has to change.

## Get the clicks that belong in that calculation

Build the plan around the medical-opinion service you actually provide and the places you can serve. Separate searches for that service from broader health information, services you do not offer, and searches for your own brand. Start with those service terms in **Keyword Planner → Get search volume and forecasts**. Set the intended locations, network scope, match types and dates, then inspect clicks and cost at different spending levels. Save the inputs and dated output so the later test has something concrete to compare with. [Google’s Keyword Planner instructions](https://support.google.com/google-ads/answer/7337243?hl=en&ref=cascader.io)

Use the forecast’s **clicks and cost**, with matching time periods. Average monthly searches are not clicks available to your ads. Google’s estimates depend on bids, budget, targeting, seasonality and historical performance. Overlapping keyword or campaign forecasts can also overstate the combined opportunity if you add them as independent audiences. [How Keyword Planner forecasts work](https://support.google.com/google-ads/answer/3022575?hl=en&ref=cascader.io)

Keep brand traffic visible separately throughout the exercise. If someone discovers you through Meta and later searches your name, a cheap branded Search lead does not establish an additional prospect created by the budget move. Reducing Meta could also change the supply of those later searches. That is a possible channel interaction to investigate, not an assumed effect of every transfer. Compare total business outcomes as well as channel-attributed ones.

Check the proposed service and target locations against Google’s healthcare advertising rules before treating any forecast as spendable. “Medical opinions” or “telehealth” alone does not establish the policy category. Google restricts some healthcare advertising by service and location, with certification required in certain cases. [Healthcare and medicines policy](https://support.google.com/adspolicy/answer/176031?hl=en&ref=cascader.io)

**If relevant keywords return no forecast, leave that part of the calculation unknown.** Google says very low-volume or sensitive keywords may be unavailable in Keyword Planner; health terms can fall into the latter group. Missing data therefore does not establish zero demand or permission to advertise. Research other accurate descriptions of the same service, but do not count unrelated traffic to fill the gap. If relevant forecasts and usable campaign history are both absent, an appropriately permitted, capped launch can establish new observations. Those observations initially describe its tested scope, not $100,000 of capacity. [Keyword Planner’s availability limits](https://support.google.com/google-ads/answer/7337243?hl=en&ref=cascader.io)

If clicks can be forecast but you have no Search conversion history, keep the **required rate** and label any comparison range as an assumption. A sitewide rate that mixes social, organic and returning visitors cannot establish what new Search traffic will do. Relevant prior Search results can inform a range; a new offer or materially different landing page weakens that comparison. Without such evidence, your first task is to measure the rate, not to make the forecast look plausible by typing in the rate you need.

## Carry the same leads through to quality

Suppose Search eventually does reach the illustrative 2,000 initial leads at $50 each. The first requirement has passed. The business requirement may still fail.

If 25% of those leads qualify under a definition you set beforehand, you have **500 qualified leads at $200 each**. If only 10% qualify, you have **200 at $500 each**. Initial CPL is $50 in both cases.

That is the cost of postponing quality: you can spend the entire $100,000, hit the early metric, and discover later that it bought fewer useful opportunities than the spend it replaced. If the business needs 500 qualified leads from this allocation, the qualification threshold is 500 ÷ 2,000, or **25%**. An unknown qualification rate is a remaining condition, not a promise that optimization will eventually supply it.

Choose the outcome that actually governs the transfer—qualified inquiry, completed consultation or sale—and define it consistently across channels. In your CRM or another appropriate business record, keep each lead’s source, creation date, unique record identifier, eventual outcome and outcome date. Count the same person once under the chosen rule. These records let you evaluate the result without requiring every event to be uploaded to an advertising platform.

Before launch, trace a few source-tagged test inquiries through the form-to-CRM handoff and verify that their later outcomes stay attached to the same records. Check what “source” actually means: Pipedrive’s `source_origin`, for example, describes how a record entered the system. “API” does not identify which advertising channel acquired it. Preserve that channel information deliberately; do not infer it from the connector. [Pipedrive’s lead-source fields](https://support.pipedrive.com/en/article/lead-source-deals?ref=cascader.io)

If most sales arrive five to ten days after an inquiry, a three-day-old lead group cannot yet tell you its eventual sales rate. Nor does “most within ten days” mean all outcomes are complete on day ten. Google likewise warns that recent periods can show full spend before later conversions arrive. [Google’s explanation of conversion delay](https://support.google.com/google-ads/answer/6239119?hl=en&ref=cascader.io)

Group leads by when they entered, then compare outcomes after the same elapsed time. For example, compare what happened to each channel’s first week of leads by 14 days after **each lead** was created, if your older records show that window captures enough of the outcome to make this decision. Fourteen days is an illustration, not a waiting rule. Check later outcomes too, and allow for recording or import delay. A report of sales made this month can include old leads and therefore cannot simply be divided by leads created this month.

If you lack dated historical records, start that record prospectively and keep recent groups marked as incomplete. You can still judge whether inquiries are arriving and whether their initial CPL is tolerable. You cannot yet certify the transfer on mature quality. Cap exposure while those first groups age; continuing to acquire more leads does not make the earlier ones mature faster.

## Decide whether “quality later” is a strategy you can use

For medical opinions, the later feedback path deserves attention before connector setup.

Google’s customer-data policy applies to enhanced conversions for web and leads and prohibits sensitive conversion information, explicitly including **purchases of medical services**. Removing a diagnosis field does not establish permission to send that purchase through enhanced conversions: the purchase itself can be the sensitive information. This is a restriction on the named measurement routes, not a finding that every medical advertisement or every form of measurement is prohibited. [Google’s customer-data policy](https://support.google.com/google-ads/answer/7475709?hl=en&ref=cascader.io)

A Pipedrive/form/Zapier workflow may be technically connectable. Zapier documents a Google Ads offline-conversion action with identifiers, a conversion action and a timestamp. That tells you about transport, not permission for your medical event or evidence that the installed workflow preserves the right records. Review the actual event, fields, source and destination before sending it. A different connector or a renamed event does not resolve an event that remains prohibited. [Zapier’s offline-conversion action](https://help.zapier.com/hc/en-us/articles/8496054576909-How-to-set-up-Google-Ads-Send-Offline-Conversion-action?ref=cascader.io)

Even a permitted, successfully imported event must be configured for the campaign’s bidding objective if it is to guide optimization. Receiving data and bidding toward it are separate steps. [How Google Ads conversion goals influence bidding](https://support.google.com/google-ads/answer/10995103?hl=en&ref=cascader.io)

If the intended downstream event cannot be sent, you still have a possible Search business. Use appropriate internal cohort records to judge quality and make allocation, targeting and landing-page decisions yourself. Use only measurement and bidding inputs established as permissible for the service. **What you lose is the right to assume that Google will repair poor initial quality using that unavailable sale signal.**

In that situation, proceed only at a scale whose economics can stand without the promised automatic improvement. If internal records cannot yet connect new leads to later outcomes, repair that measurement first. A bigger acquisition test would otherwise spend money without answering the question you need settled.

## Let evidence release the next part of the budget

Choose the first allocation to test the weakest consequential assumption: affordable relevant clicks, the required conversion rate, or the mature outcome rate. Set a maximum cumulative amount the business can afford to put at risk before the first useful quality readout. Include the burden on the website and the people handling inquiries; a lead supply you cannot serve is not useful capacity.

The $25,000 point in the example is a monthly planning scenario, not a recommended test budget. A smaller capped test may be sufficient to discover that the traffic is irrelevant or the form is failing. If it cannot produce enough mature outcomes within the affordable exposure, leave the full transfer unapproved; a tiny inconclusive test does not justify a large bet.

As evidence arrives, replace assumptions with observed clicks, cost, initial leads and aged outcomes. Evaluate the added spending as well as the combined average, since earlier inexpensive traffic can make a more costly expansion look acceptable. Compare with what the displaced Meta spend was contributing; its historical average CPL is a planning reference, not proof of the precise leads you will lose by cutting its next dollar.

Also inspect weekly and daily lead counts, days with little delivery, and response capacity. A monthly average that fits the equation does not establish a steady daily flow. Google explicitly notes that daily traffic varies around its forecast averages. [Keyword Planner forecast limitations](https://support.google.com/google-ads/answer/3022575?hl=en&ref=cascader.io)

**Before using Search lost IS (budget), check the campaign’s bidding strategy.** The metric concerns Search impression share lost because of budget. Where the strategy is compatible, this campaign-level metric can indicate budget-related missed serving opportunities; it still cannot establish affordable incremental leads. [Google’s impression-share definitions and reporting](https://support.google.com/google-ads/answer/7103314?hl=en&ref=cascader.io) Google specifically identifies the column as **incompatible with Maximize conversions**: the column does not account for that strategy’s design of spending the budget. A displayed value does not make it a suitable scaling diagnostic. [Google’s Maximize conversions guidance](https://support.google.com/google-ads/answer/7381968?hl=en&ref=cascader.io)

If the metric is incompatible, missing, or its suitability is unclear, use the observations already collected. Divide actual cost by clicks to find current CPC, then divide CPC by target CPL to find the required lead rate. Compare that with the observed lead rate and mature quality under the definitions above. If they miss your requirements, hold the transfer. If they meet them, they support considering an affordable next increment—not a claim that the increment will perform equally well. Recheck its actual costs, leads and matured outcomes before releasing more.

Search earns a larger allocation when relevant demand, comparable CPL, mature quality and service capacity hold together at the expanded scope. Until then, the useful result is a conditional range with a named assumption to test. You can learn that a partial transfer works without proving that the entire $100,000 belongs there.