What the dashboard shows
Spend divided by every form fill, call or chat. It rewards whatever is cheapest to generate, including leads sales will never call back.
Lead quality · Google, Meta & LinkedIn Ads
Lead quality optimisation isn’t one setting in Google or Meta. It’s a loop: pick a signal each platform can learn from in time, send it back where configured, then decide what to change across keywords, audiences, creative, landing pages and budget, and wait when the evidence isn’t strong enough.
Definition
Short answer
It is guiding paid campaigns with evidence of which leads became worth pursuing, such as qualified RFQs, accepted demos or proposals, instead of cost per lead alone. It combines sending suitable signals to Google, Meta and LinkedIn with decisions about search terms, audiences, creative, landing pages and budget based on that evidence.
Spend divided by every form fill, call or chat. It rewards whatever is cheapest to generate, including leads sales will never call back.
Spend divided by the leads sales agrees are worth pursuing. It is the number that tells you which campaigns are actually building pipeline. Definition
The missing layer
Short answer
Importing qualified leads helps each platform’s bidding learn, but it works inside that platform only. Someone still has to decide which search terms to exclude, which audience or message attracts unqualified leads, whether the landing page filters well, and how to split budget across Google, Meta and LinkedIn. That decision layer is what Zephra adds.
Connectors move data; advice articles tell you to “optimise to revenue”. Neither decides what to change on Monday morning. Smart Bidding, Meta’s Conversion Leads and LinkedIn’s qualified leads optimisation each improve delivery inside their own auction. None of them will tell you that a search-term cluster, an Instant Form question or a LinkedIn audience is the reason sales rejects half your leads, or that the budget should move between platforms. How signals get back to each platform is covered on conversion tracking.
How Zephra works
Lead status, qualification, sales feedback and later outcomes such as quotes, proposals or opportunities, from CRM data or a spreadsheet.
For each decision, the stage that is timely, frequent and reliable enough to act on.
CRM lead stages go to Google (Data Manager API), Meta and LinkedIn (Conversions APIs), where configured.
Where quality breaks: search terms, audiences, placements, creative, landing pages or forms.
Act, review, investigate or wait, within your permissions and approvals. Every change is logged.
Early signals are checked against later wins and losses, so the chosen stage keeps predicting value.
Original framework
One disqualified enquiry adds context; it does not by itself justify a campaign change. The strength and consistency of the evidence decides the response.
| Evidence pattern | Likely cause | Response | What Zephra checks | Measure next |
|---|---|---|---|---|
| A few recent disqualified leads | Normal variation | Investigate | Whether they share a search term, audience, ad or form | The next batch of lead statuses |
| Consistent disqualification from one search-term cluster, audience or ad | Targeting reaches the wrong intent | Act (within approvals) | How much qualified volume the cluster also produces | Qualified rate after the exclusion |
| Qualified rate falling while frequency rises | Quality drift or message fatigue | Review | Reach growth, frequency, creative age | Qualified rate by ad |
| Qualified rate fine, but deals slow | Sales capacity, offer or timing, not the ads | Wait or review with sales | Time from qualified to opportunity | Opportunity rate over a matured cohort |
| No stage recorded for a large share of leads | The feedback loop is broken | Investigate | Which sources and owners leave leads unmarked | Stage-update coverage |
| Conflicting signals: cost per lead down, accepted rate down, pipeline unknown | Cheaper, weaker leads | Review | What changed: matching, forms, audiences, placements | Cost per qualified lead by channel |
Signal timing
A deeper outcome is not automatically a better signal. Each platform only learns from outcomes that arrive in time and in enough volume.
| Platform | Window | Volume guidance | Source |
|---|---|---|---|
| Google Ads | Enhanced conversions for leads uploads within 63 days of the click | At least 15 conversions in 30 days for a goal, ideally reached within about 7 days | Google Ads Help |
| Meta Ads | Optimised stage within 28 days of the lead; uploads at least daily | 200+ leads a month, with 1–40% reaching the stage | Meta for Developers |
| LinkedIn Ads | Qualified-lead data shared within 30 days | At least 5 qualified leads within two weeks recommended | LinkedIn Help |
Last verified 15 September 2026. The full argument and worked numbers: the signal gap.
In practice
An industrial automation firm in India.
| Channel | Spend | Leads | Cost per lead | Qualified | Cost per qualified lead | Quotes | Cost per quote |
|---|---|---|---|---|---|---|---|
| Google Ads | ₹3,50,000 | 230 RFQs | about ₹1,522 | 46 | about ₹7,609 | 14 | ₹25,000 |
| Meta Ads | ₹1,00,000 | 160 | ₹625 | 8 | ₹12,500 | – | – |
| LinkedIn Ads | ₹1,50,000 | 30 | ₹5,000 | 12 | ₹12,500 | 5 | ₹30,000 |
A group of Google search terms consistently produces RFQs sales rejects. Apply the exclusions within approval rules.
Meta has the cheapest leads and the lowest qualified rate, and at 160 leads a month it is below Conversion Leads’ guidance. Tighten the form.
Its qualified rate is strongest, but five quotes are too few to justify reallocating. Re-check next cohort.
A quarter of Google leads have no stage update from sales. Fix coverage before trusting the Google numbers further.
What to measure next: stage-update coverage, cost per qualified lead by channel, and quote rate. The same pattern holds for a Dubai consultancy in AED or a US SaaS team in dollars.
Getting started
A spreadsheet that sales actually updates beats a CRM nobody maintains. Keep five fields for every lead:
Agree who updates it and how often; weekly is the minimum, daily if you want Meta to learn from it. Phone and WhatsApp enquiries go in the same sheet, or they stay invisible.
By company size
| Size | Lead status source | Who defines “qualified” | Approvals | Review cadence |
|---|---|---|---|---|
| SMB | Spreadsheet or basic CRM | Founder or sales lead | Owner approves | Weekly |
| Mid-market | CRM stages | Sales and marketing together, often RevOps | By action type | Weekly, with a monthly cohort review |
| Enterprise | CRM across regions and brands | RevOps with regional sales | By account and role, with audit logs | Monthly cohorts, quarterly stage validation |
Controls
Industries
Questions, answered
No. A consistent record of which leads sales qualified or rejected, kept in a spreadsheet, is enough to start. A CRM adds stage history and automation. What matters most is coverage and timing: most leads get a status, and statuses are updated soon enough to inform the next decision.
It depends on lead volume and how quickly leads reach a qualification stage, not on a fixed timeline. A few rejected leads justify investigation, not a change. A consistent pattern across enough leads can justify review or action, and later wins and losses are used to check whether the early signal was right.
Use the same stage definition for every channel and compare cost per qualified lead or per opportunity for leads acquired in the same period. Expect LinkedIn to cost more per lead; what matters is cost per outcome. When a channel has only a handful of outcomes, treat the comparison as directional.
It may. Excluding searches and audiences that never qualify usually means fewer leads, and sometimes more expensive ones. The aim is more qualified leads and opportunities for the spend, and less sales time on leads that go nowhere.
Whatever your sales team agrees is worth pursuing, written down so everyone applies it the same way. Typical cases are an RFQ matching your specification and volume, a demo request from a company in your target size range, or an enquiry from a firm with a real project and budget.
Yes, if the outcome is recorded against the lead. For dealer or distributor networks, record the dealer’s outcome as the stage.
Zephra uses the lead status, qualification and sales feedback you provide to judge campaigns. It is not a predictive lead-scoring tool; your team’s definition of qualified stays the source of truth.
Start with your account
The free assessment shows wasted spend, structure, constraints and measurement gaps across Google, Meta and LinkedIn. Then bring your last 90 days of lead statuses to a demo.
Related: Google Ads · Meta Ads · LinkedIn Ads · Pricing · Security