Guide · Diagnosis
Not getting B2B leads from your ads? Diagnose in this order before you change anything
Most “no leads” months are a measurement, delivery or noise problem before they are a campaign problem. Seven checks, in a fixed order, tell you which one you have, and whether the right decision is to act, review, investigate or wait.
Why is my business not getting leads from ads?
Short answer
Check in order: whether leads are recorded correctly, whether the drop is bigger than normal variation, whether campaigns are delivering, whether demand changed, whether you reach the right searches and people, whether the page or offer blocks conversions, and whether leads arrive but get lost before sales sees them. Change nothing until the first two checks are done.
The order matters because the checks get more expensive. Confirming that a form still records submissions takes minutes. Rebuilding audiences or rewriting a landing page takes weeks, resets what ad platforms have learned, and can’t fix a thank-you page that stopped firing a tag.
Is it a volume problem or a quality problem?
Short answer
Too few leads is a volume problem, and this guide covers it. Plenty of leads that sales rejects is a quality problem, which needs a different diagnosis: start with junk leads from ads. If you can’t tell which you have, agree what counts as a qualified lead first, using lead quality.
The two feel the same in a monthly meeting (“marketing isn’t working”) and have opposite fixes. Tightening targeting helps a quality problem and worsens a volume problem. If qualified leads are the complaint, count both leads and qualified leads for the last three months before choosing a path.
Original framework
The Lead Drought Diagnosis Order
Seven checks across Google, Meta and LinkedIn. Each lists the evidence to look at and the decision it usually leads to. Stop at the first check that explains the drop, fix it, and re-check before going further.
| # | Check | Evidence to look at | Usual decision |
|---|---|---|---|
| 1 | Are leads recorded? | Form or inbox count vs CRM or sheet vs platform conversions; conversion action status; event delivery | Investigate |
| 2 | Is the drop real? | Recent weeks vs expected range at your volume; same period last year | Wait or Investigate |
| 3 | Are campaigns delivering? | Spend pacing, disapprovals, billing, recent significant edits, audience size | Act to fix, or Wait during learning |
| 4 | Has demand or competition changed? | Impression share lost to budget or rank, CPC and CPM trends, frequency | Review |
| 5 | Are you reaching the right searches and people? | Search terms, audience expansion, placements, creative fatigue, ad-to-page match | Act |
| 6 | Is the page, form or offer blocking? | Conversion rate by landing page, mobile form errors, offer friction | Review |
| 7 | Are leads arriving but getting lost? | Unassigned or unworked leads, phone and WhatsApp enquiries never logged | Investigate |
The decision labels follow the same logic Zephra uses for its findings: Act when the evidence is clear and the fix is safe, Review when a person should weigh a trade-off, Investigate when something doesn’t add up, and Wait when the data is too thin or the platform is still learning. See how it works.
Check 1: are leads actually being recorded?
Short answer
Compare three counts for the same weeks: submissions in the form tool or inbox, leads in the CRM or sheet, and conversions in each ad platform. If only the platform number fell, the ads probably didn’t break; the measurement did. Fix that first, because automated bidding reacts to missing conversions as if good traffic had stopped working.
What to look at:
- Conversion action status in Google Ads. Google marks an action “Misconfigured” when “conversions have stopped recording entirely due to a setup error or a broken tag”, and “Awaiting conversions” when none have been recorded in the last 7 days (Google Ads Help).
- Recent site changes. A new thank-you URL, a redesigned form or a moved tag manager container are the usual causes of a sudden, total drop.
- Consent banners. When visitors decline, tags may not fire, so platform conversions fall while real enquiries don’t.
- Server-side events. If you send conversions through Meta’s or LinkedIn’s Conversions API, check events are accepted rather than rejected, and not duplicated with the browser tag.
- Offline leads. Phone calls and WhatsApp messages never become platform conversions unless you record and upload them.
Setup detail is on conversion tracking. Remember the reverse too: fixing tracking restores reporting, not demand. If leads “jump” the week tracking is fixed, that is recovered reporting, not new business.
Check 2: is the drop real, or normal variation?
Short answer
Compare the last four weeks with the previous eight to twelve, and with the same period last year. Small B2B accounts swing a lot by chance: an account averaging four leads a week will see two or fewer in about one week in four. Act on sustained drops across several weeks, not on one bad week.
When leads arrive independently at a steady average rate, weekly counts follow a Poisson distribution, which makes normal variation easy to estimate. The table shows the range that holds at least nine weeks in ten at each average.
| Average leads a week | Normal weekly range | Chance of 2 or fewer in a week | Chance of a zero week |
|---|---|---|---|
| 2 | 0 to 5 | about 68% | about 14% |
| 4 | 1 to 8 | about 24% | about 2% |
| 10 | 5 to 15 | under 1% | almost none |
| 55 | 43 to 67 | almost none | almost none |
Read it this way. At four leads a week, a two-lead week is not evidence of anything. At ten a week, three leads is well outside the normal range and deserves Investigate. An Indian industrial manufacturer receiving 240 RFQs a month averages about 55 a week (240 × 12 ÷ 52 is 55.4); a 30-RFQ week there is far below the normal range and warrants Investigate the same day.
Then check the calendar. Compare with the same weeks last year, and allow for Diwali and the March fiscal year-end in India, Ramadan and summer in the GCC, and Q4 holidays and fiscal-year budget cycles in the US. The model is a method, not a platform rule: real leads cluster around trade shows, launches and budget cycles, so treat a result near the edge of the range as Review.
Check 3: are the campaigns actually delivering?
Short answer
Look at spend before performance. If a campaign spent far less than planned, the cause is delivery: disapproved ads, a billing failure, a budget cap, bids too low to win, an audience too small, or a recent edit that restarted learning. Fix what is broken; if the campaign is simply learning after a change, the decision is Wait.
- Google: after a bid strategy, setting or composition change, Google describes a learning period of up to around 50 conversion events or three conversion cycles (Google Ads Help). Several edits in a row stretch it.
- Meta: a significant edit, such as pausing an ad set or changing its optimisation event, audience or creative, counts as significant (Meta Business Help), and ad sets usually leave learning after about 50 results in the week after the last significant edit (Meta Business Help).
- LinkedIn: the minimum audience is 300 members, and LinkedIn suggests at least 50,000 to drive results (LinkedIn Help). Layered job titles at a short company list can leave a campaign unable to spend.
Deeper platform checks are in the Google Ads audit guide, why Facebook ads don’t convert and LinkedIn Ads cost.
Check 4: has demand or competition changed?
Short answer
If campaigns deliver but fewer people see or click the ads, the market may have moved. On Google, compare impression share lost to budget and lost to rank. On LinkedIn, watch frequency and cost per impression. On Meta, look for rising CPMs. Changes here call for Review, because the fix is usually budget or bidding.
Google splits lost impression share into budget, the share of time ads didn’t show because the budget ran out, and rank, the share lost to Ad Rank (Google Ads Help). Rising rank losses with steady spend often mean a competitor has arrived or raised bids. Rising frequency on a small LinkedIn audience calls for new creative or a wider audience, not more budget.
Check 5: are you reaching the right searches and people?
Short answer
Read what actually triggered the ads. On Google, the search terms report shows the real searches behind clicks. On Meta and LinkedIn, check audience expansion, placements and whether the creative has worn out. Irrelevant searches or tired ads cost clicks that never convert. This is usually an Act finding, because the evidence is direct.
The search terms report shows the searches that triggered your ads, as opposed to the keywords you bid on (Google Ads Help). Broadened match types, new automated campaign types and language drift all show up there first. Remedies are in negative keywords strategy and the creative testing framework.
Check 6: is the page, form or offer blocking conversions?
Short answer
If clicks hold steady and leads fall, look at the page. Compare conversion rate by landing page and device, submit the form yourself on a phone, and check whether the offer asks too much of the traffic it receives. A demo request can work on search and fail with cold LinkedIn audiences. Most findings here call for Review.
Common causes are a broken mobile form, a new required field, a slower page after a redesign, and ads that promise one thing while the page offers another. See the landing page guide, B2B lead forms and positioning and offer.
Check 7: are leads arriving but getting lost?
Short answer
Sometimes the leads exist and nobody sees them. Lead form downloads that nobody runs, a shared inbox nobody owns, calls to a salesperson’s mobile and WhatsApp chats never logged all make a lead drought appear in the CRM. Trace a few recent leads end to end, from ad to named owner, before changing campaigns.
Owners, deadlines and statuses are covered in the sales handoff guide.
Worked numbers
SaaS leads fell from 10 to 3 a week: walking the order
A US B2B SaaS company spends $15,000 a month across Google Search and LinkedIn. Demo requests averaged about 10 a week. Last week the platforms reported 3, and the founder asks for new campaigns.
- Recorded? The platforms show 3 conversions, but the CRM shows 7 demo requests. The website team changed the demo form’s thank-you redirect ten days ago, and the conversion tag stayed on the old page.
- Real? At an average of 10 a week, 7 sits inside the normal range of 5 to 15. A week of 3 would have happened by chance about 1% of the time, but that was the platform figure, not the real one.
- Stop. Checks 3 to 7 aren’t needed yet: the drop in real leads is within normal variation.
Move the conversion tag to the new page and confirm the status recovers. Automated bidding has seen a week of missing conversions and may have shifted spend, so check the campaigns again after the fix.
Inside the normal range. No campaign changes. Re-check once three more weeks of CRM counts are in.
The ads didn’t break; the measurement did. New campaigns would have reset learning and hidden the real cause. The same walk works for an industrial account with the RFQ numbers above.
What changes by company size?
| Size | Where droughts usually start | What to put in place |
|---|---|---|
| SMB | A founder-run account where tracking breaks silently after a website change | A weekly count of form, sheet and platform leads side by side |
| Mid-market | An agency and an in-house team, each assuming the other owns tracking | A named tracking owner and a shared change log |
| Enterprise | Many campaigns and regions, with changes made without review | Approvals, spend boundaries and alerts on spend and conversion anomalies |
Where configured, Zephra monitors spend and CPA anomalies and tracking health, records every change in an audit log, and applies approvals and spend boundaries before changes run.
Context
India, UAE and US: where lead droughts hide
India
Many enquiries arrive by phone or WhatsApp and never become conversions, so check 1 often finds the missing leads. Marketplace activity, Diwali and the March fiscal year-end shift demand. LinkedIn audiences for niche industrial titles can be too small to deliver. See marketplaces vs paid ads.
UAE and GCC
Compare Ramadan and summer weeks with the same weeks last year, not the month before. Check whether campaigns target the UAE only or the wider GCC, since a settings change can quietly cut volume. Look for Arabic-language searches in English campaigns. See GCC B2B paid ads.
US
Q4 holidays and fiscal-year budget cycles change B2B search volume. Check state-level location settings after edits. Consent banners on US sites can reduce tag firing, which shows up in check 1 as platform conversions falling faster than CRM leads.
What should you measure next? The drought log
Short answer
Keep a drought log: every time leads look low, record the date, platform, metric, which check explained it, the evidence, the decision and a re-check date. Alongside it, track weekly leads against the expected range for your volume. Once volume returns, switch attention to the qualified-lead rate, so recovered volume doesn’t hide falling quality.
| Date | Platform | Metric moved | Check # | Evidence | Decision | Owner | Re-check |
|---|---|---|---|---|---|---|---|
Cost per qualified lead is covered in the cost per qualified lead guide.
The free ad account assessment reviews wasted spend, campaign structure, performance constraints and measurement gaps across Google, Meta and LinkedIn in 10–15 minutes, with no CRM needed and no campaign changes. See what it covers.
Questions, answered
Lead drought FAQ
01Should I change my campaigns when leads drop?
Not first. Confirm leads are recorded, the drop is real and campaigns are delivering. Edits during a learning phase can make things worse, and many findings call for Investigate or Wait rather than Act. Change campaigns once the evidence points to relevance, targeting, creative or budget.
02Can broken conversion tracking cause fewer leads?
It causes fewer reported leads, and can then cause fewer real ones. If a platform stops receiving conversions, automated bidding may treat good traffic as failing and shift spend. Compare form or CRM counts with platform conversions; if only the platform number fell, fix measurement before touching targeting.
03Why is my LinkedIn campaign getting no leads?
Common causes are an audience too small or narrowly layered, bids too low to win delivery, an offer that asks cold professionals for too much, and a Lead Gen Form or Insight Tag that isn’t recording conversions. Check delivery and tracking first, then audience size, offer and creative. See LinkedIn Lead Gen Forms.
04How long should I wait before judging a new campaign?
Until it has left learning and produced enough leads for the weekly range to mean something. At a few leads a week, that is usually several weeks. Judge on the trend across those weeks, not on the first one.
05Should I add another channel to fix a lead drought?
Not until the diagnosis is done. A new channel doesn’t fix a broken tag, an unworked inbox or an offer that doesn’t fit the traffic, and it splits a small budget further. Add channels to reach buyers you can’t reach today, not to escape an unexplained drop.
06Leads fell right after we fixed tracking. Why?
Often because duplicates were removed. If a browser tag and a server-side event were both counting the same lead, fixing deduplication lowers the reported number without losing any real enquiries. Compare with CRM counts before reacting.
Your accounts
Find out whether your lead drought is tracking, delivery, targeting or the page.
The free ad account assessment takes 10–15 minutes, needs no CRM and makes no campaign changes. It covers wasted spend, campaign structure, performance constraints and measurement gaps across Google, Meta and LinkedIn.