For lead gen, which downstream stage should we optimize for, and how should we judge results?
Our answer
Optimize for a qualified CRM stage once you get about 200 leads a month and that stage reaches about 50 a week per ad set (Loomer plans on 25 to 50). Judge on cost per qualified lead.
- Confidence
- Medium
- Strongest proof
- 4 Meta rules
- 15 practitioners, 3 opinions or claims
- Sources
- 17 sources
- 22 statements
M
New here? Take the 1-minute tour
Why this is the answer
3 reasons and our judgmentMeta sets the bar for optimizing for a CRM stage. Its fit test asks for at least 200 leads a month, daily uploads and a stage that 1% to 40% of leads reach within 28 days.
Connecting a CRM lowers the cost of a good lead, by Meta's own figure. Instant Forms with a CRM connected showed 15% lower cost per quality lead on Meta's averages (reported by Click Here Digital).
Cheap leads can cost more in the end. In Cardinal Digital Marketing's behavioral-health case, a form with a much lower cost per lead had nearly twice the cost per new member.
Our judgment
Treat a stage reached by about a third of leads as the safe pick. Meta guidance as Zapier reports it says one-third to one-half of leads, with 5% too few, while Meta's developer guide says 1% to 40%. The two overlap only at about 33% to 40%. At 200 leads a month, about 46 a week, such a stage fires about 15 times a week, below Loomer's 50 per ad set. So accounts near the 200 line need few ad sets or a nearer stage. A stage reached by more than 40% of leads sits above Meta's range and calls for a deeper one.
Our judgment, built from Meta and Jon Loomer: our reading of these sources, not a finding from any one of them.
What the sources say
22 statements from 17 sourcesJeremy Haynes sees higher-earning webinar advertisers moving from optimizing for leads or registrations to purchases or booked calls, especially at scale, though some still make lead events work. No counts given.
“Seeing a lot of the higher earning webinar guys optimizing for purchase or schedule instead of lead or complete registration”
Jon Loomer found an undocumented Meta option that credits a lead only when a first and a final event happen in order; eligibility seemed to need about 30 optimized events in 7 days.
“if you're barely getting 50 optimized leads per week per ad set now, you may not get enough data”
Silverback Strategies, a lead gen agency, says lead-to-sale lag runs from days to months and shifts by season (one client: 70 days in September, 7 in April), so optimize to qualified leads once volume allows.
“One education client we work with sees a 70-day average lag for leads generated in September and a seven-day lag for the same programs in April”
Kalungi says qualified leads and dead signups often fire the same event, so Meta finds more of both; one B2B client's lead volume held for months as qualified conversations fell. Send a separate qualified-lead signal.
“A real, qualified lead and a completely dead signup both fire the same event, worth the same amount, to Meta's algorithm.”
Phil Graham says Meta optimizes for the event you select, which may not be your most valuable business outcome.
“Meta optimizes for the actions that you tell it to, the actions that you select.”
Nick Boddington says moving leads to Qualified or Converted in Meta's Lead Center feeds signals back to Meta that improve future lead quality.
“those actions that you're taking and putting them into those columns is feeding meta back with signals”
Jon Loomer says to start lead forms with the fewest questions and add controls like SMS verification only once data shows a quality problem, aiming for 50 optimized events per ad set per week.
“Your goal should always be to get 50 optimized events per ad set per week.”
WEBRIS's Ryan Stewart says a Texas injury law firm optimizes Meta to signed cases, not leads, since 90 to 95% of leads are unqualified; results settle in months five and six, at about $1,500 per signed case.
“If you are optimizing for leads instead of signed cases you are telling the algorithm to find you the cheapest form fills, not the best clients.”
Flighted, an agency, says B2B advertisers should send Meta a lead event and a qualified-lead event, feed deals back through the Conversions API, and not judge results before a 60 to 90+ day sales cycle plays out.
“Feed Meta two conversion signals: a Lead event and a down-funnel Qualified-Lead event.”
Elevarus, a lead-gen agency, says to judge Meta on cost per qualified appointment, with closed sales as the final test, and send qualified and closed leads back through the Conversions API.
“Cheap leads that never book are the most expensive thing in the account”
Cardinal Digital Marketing says to optimize toward the deepest signal with enough volume, at least 20 to 30 conversions per campaign a month, using CRM outcomes since some booking tracking overcounted 60%.
“The goal is always to optimize toward the best signal you can get enough volume from”
Daniel Dimsey keeps Meta lead forms on Maximize number of leads, saying Maximize number of conversion leads mainly raises cost per lead without improving lead quality.
“it doesn't actually change the quality.”
Zapier relays Meta: CRM-connected lead campaigns averaged 21% lower cost per quality lead on Instant Forms, with no sample. Optimize for a stage one-third to one-half of leads reach.
“roughly one-third to one-half of your leads”
GrowthSpree, a B2B SaaS agency, says to judge Meta on cost per sales-qualified lead over a 180-day window rather than cost per lead, because B2B journeys are long. It gives no sample.
Jon Loomer says his own tests don't show Instant Forms bringing lower-quality leads; he tags leads who click an email within 30 days as quality leads and sends that event to Meta, with mixed results.
“my own tests do not reflect this quality drop-off”
An agency relays Meta test averages: running Instant Forms and website forms together cut cost per lead about 60% and raised volume 125% against website only. No sample is given.
“Meta reported that advertisers running both Instant Form and website form campaigns together saw an average 60 percent lower cost per lead and 125% increase in lead volume”
Click Here Digital says optimizing Instant Forms for lead quality now needs CRM data shared through the Conversions API for CRM; without it you optimize for volume. Meta claims 15% lower cost per quality lead.
“now requires sharing CRM data through Conversions API for CRM”
Involve Digital says Meta's conversion leads goal fits at 200+ leads a month, and smaller B2B accounts should still send CRM stages back. Use Instant Forms for light offers, landing pages for demos and high-ticket.
“Many B2B accounts fall short of 200 leads a month; they should still send the stages back, for reporting and audiences, and switch to conversion leads optimisation when volume allows.”
Trio Media, an agency, says cheap Meta lead form leads often go nowhere, so judge lead gen on qualified leads and revenue rather than cost per lead, and feed sales outcomes back.
“The real value isn’t the cheapest lead, it’s the qualified lead.”
Says lead forms are often cheaper per lead but not always cheaper per qualified lead or opportunity, so compare lead formats on at least one downstream metric.
“Always compare formats using at least one downstream metric.”
Meta's guide says the Conversion Leads goal works only with Instant Forms and fits best with 200+ leads a month, daily CRM uploads, and a target stage reached within 28 days by 1% to 40% of leads.
“The Conversion Leads performance goal is currently only compatible with Facebook/Instagram’s Lead Ads (Instant Forms).”
Cardinal Digital Marketing reports a healthcare client's Meta lead form with higher cost per lead but much lower cost per new member. Cost per new member fell 73% in four months, with changes made in sequence.
“CPL is only useful when it correlates with patient acquisition.”
In the full answer
For readers who sign in- What to do this week3 steps
- The numbers behind it3 figures from cited sources
- The strongest disagreementThe best case against the answer, from a cited source
- How sure we areWhat its confidence rests on, and what would change the answer