Is It a Lead Quality Problem or a Follow Up Problem?

To put this workflow into practice, explore managed AI implementation. For the related checks and handoffs, use our ai sales implementation guide.

Why can't the sales team and the marketer settle this by arguing?

Because each side is arguing from a true but incomplete story. The sales lead remembers the leads that were obviously wrong: out of the service area, price-shopping with no budget, a wrong number. The marketer remembers the campaign metrics: cost per lead is down, form fills are up, the funnel looks healthy on their side of the handoff. Both are describing something real. Neither is describing the whole system.

The argument usually happens lead by lead, in a meeting where someone pulls up three bad leads and someone else pulls up the ad spend report. Individual examples do not settle a pattern question. A CRM with hundreds or thousands of disposed leads already has the answer sitting in it; the fix is to stop trading anecdotes and pull the aggregate numbers both sides can see at once.

What four numbers settle whether it's leads or follow up?

Pull these directly from the CRM for a single channel and service area over a comparable period, not from memory:

1. Contact rate. The share of leads where a rep actually reached the person, by phone, text, or a reply to an email, not just attempted to.
2. Median attempts per lead before the record goes cold. The middle attempt count when lead records are sorted by attempts, including records with no attempts. Track completed follow-up sequences separately from leads still being worked.
3. Time to first attempt. How long between the lead landing and the first outreach try, which shapes whether contact was ever likely in the first place.
4. Disqualification reason distribution. A count of every reason a lead was marked not viable, grouped into a small fixed list rather than free text.

These four numbers, read together, are what actually separate a lead problem from a follow-up problem. Any one of them alone can mislead you: a low contact rate could mean bad phone numbers or it could mean nobody called back promptly, so review attempt timing, valid contact details, and delivery outcomes before assigning a cause.

What does each pattern in the data actually mean?

High attempts, low contact. Reps are trying repeatedly and still not reaching people. Investigate contact details, delivery, timing, and channel choice: disconnected numbers, calls that do not reach the recipient, a form that does not confirm a working phone number, or outreach at unsuitable times. High attempt counts alone do not establish that the leads are poor. Repeated attempts made late or clustered together can still reflect a process problem.

Low attempts, low contact. Reps may be moving on before completing the agreed follow-up sequence. Check the lead history before blaming targeting. A low count can also reflect a recent inquiry, a request to stop, or a duplicate record. Compare leads that have had the same time to receive follow-up, and inspect whether reps honored callback preferences. If valid leads sit unassigned or the sequence ends without a reason, correct that process gap and measure contact again.

High contact, low qualification, with a consistent reason. This is stronger evidence of a targeting or intake problem. Reps are reaching people, but the same disqualification reason keeps showing up: out of area, wrong service, or no intent to buy the offered service. Review conversation notes and qualification criteria before changing targeting. When the pattern persists across comparable batches, test a clearer offer, tighter service-area filters, or a change to the channel.

Treat these patterns as investigation prompts, not a verdict. They help sales and marketing agree on what to inspect next without assuming that every missed conversation has the same cause.

How do you build reason codes that don't hide the answer?

A reason-code field that lets reps type anything hides the pattern you are trying to see. Two changes make it repeatable:

• Use a short, fixed list, not free text. A starting list might include: out of service area, wrong service or product fit, no budget or authority, not interested after a real qualifying conversation, unreachable after the full attempt sequence, duplicate or existing customer.
• Separate out-of-area and wrong-service from generic "not interested." Combining these reasons hides the change you need to test. "Not interested" after a conversation could reflect timing, the offer, or the conversation itself. "Out of area" or "wrong service" identifies a fit problem to investigate in targeting or intake. Keep each distinct so you can tell poor fit from failure to reach someone.

There is a useful discipline to borrow here from outside sales entirely. The FTC's own guidance to advertisers is that advertising claims need evidence behind them. Treat "the leads are junk" the same way internally: it is a claim, and the reason-code data is the evidence, not the meeting where someone remembers three bad calls.

What if both problems are real at the same time?

Both can occur: a business can have thin leads in one segment and slow follow-up everywhere at once. The order to fix them in still follows the numbers. Fix follow-up first if contact rate is low, because unreached leads leave you with incomplete evidence of fit. Report the reasons from actual conversations separately from unreachable records. Clear intake evidence, such as an address outside your service area, can still justify a filtering fix while you improve follow-up.

In practice this usually means: assign an owner to the process gap, then revisit the ad targeting or offer with clean data instead of a guess. Businesses that already have a sales team looking to close this gap without adding headcount can start with the guide to handling more inbound leads without hiring more reps, which covers the capacity side of the same intake problem from a different angle.

What does a weekly review with sales and marketing look like?

A one-page review, run by the owner with the sales lead and the marketer in the same room, keeps this from turning back into an argument over anecdotes. Cover, in order:

1. The four numbers, by channel, for the past week and the trailing four weeks, so a single slow week does not get overweighted.
2. The reason-code breakdown, by channel and service area, with out-of-area and wrong-service called out separately from every other disqualification.
3. One pattern call. Based on the numbers above, is this week's data or channel, process, or targeting? If it does not clearly fit one pattern, that is itself useful information, not a reason to skip the call.
4. One decision and one owner. A single change to make before the next review, assigned to a name, not to "the team." Fixing a channel, tightening intake, or adjusting attempt cadence.
5. What changed since last week's decision. Close the loop before opening a new one.

Running this weekly, with the same four numbers every time, is what turns the debate from opinion into a habit both sides trust. Logging attempts, contact outcomes, and reason codes consistently enough to support this review usually means the CRM and calling tools need to talk to each other automatically rather than depending on reps to fill in a field by hand; that is the kind of setup covered by MetaTechAi's managed AI lead response and sales services, built on the RizzDial and Beam tools MetaTechAi built and runs, for calling, texting, follow-up, and CRM workflows. Specify the fields you need and test that each outcome reaches the correct record before relying on an automated report. For a related qualification workflow, RizzDial's guide to AI lead qualification for agencies covers qualification questions and call outcomes for teams serving clients.

MetaTechAi installs AI sales and marketing systems for service businesses with sales teams, with a guarantee that conversions go up or you don't pay. If you are weighing whether to bring in outside help to fix the intake and follow-up side of this before touching targeting, the brief to prepare before hiring an AI automation agency covers what to bring to that first conversation so it starts from your real numbers instead of a guess.

What else do owners ask about diagnosing lead quality versus follow up?

How many leads do we need before these numbers mean anything?

Use a comparable group of leads that has had time to complete your follow-up sequence. Show the lead count alongside every rate, and treat a small group as a signal to investigate rather than a settled trend. Review repeated periods before changing targeting on the strength of one batch.

What is a healthy number of attempts per lead?

There is no universal attempt target. Review how many additional conversations each stage of your agreed sequence produces, while respecting callback preferences and requests to stop. If later attempts still reach suitable prospects, check whether reps are ending the sequence early. If they rarely connect, investigate timing, delivery, and contact details before adding attempts.

Should we pause ad spend while we fix follow up?

Not automatically. Review lead fit, follow-up capacity, and campaign performance together. Low contact alone does not prove targeting is wrong, but buying more inquiries while the team cannot work them may also be unhelpful. Agree on a bounded test and a review date instead of making the budget decision from anecdotes.

What if both problems are real at the same time?

Work on documented fit problems and follow-up gaps together where the evidence supports both. When contact is low, improve follow-up before generalizing from the conversations you did reach. Keep unreachable records separate from confirmed disqualifications, then compare the reason codes again after the process change.