Google Ads leads not converting? A practical recovery playbook.

Decide what is actually failing

When a leadership team says, “We are spending on Google Ads, but the leads are not converting,” several different problems may be hiding behind the same sentence. The account may be attracting the wrong buyers. The website may be losing suitable visitors. A form may be recording an interaction that never became an enquiry. Sales may be receiving genuine interest without the information or capacity to advance it. Each problem calls for a different intervention. Increasing the budget before distinguishing them makes the uncertainty more expensive.

Begin with a commercial definition of success. Describe the customer you can profitably serve, the problem they are ready to address, the minimum viable scope and the evidence that an opportunity is real. Then agree the point at which marketing hands responsibility to sales. A downloaded guide, a completed contact form, an accepted discovery meeting and a signed engagement are different stages. Combining them into a single conversion number removes the very distinction management needs to allocate capital.

This guide provides a diagnostic sequence for a business with an established offer and an active sales process. It is designed to produce decisions, owners and testable corrections. Work through it with the people responsible for advertising, the website, measurement and sales operations. Ask them to bring the same period of data and a small, documented sample of actual enquiries. The aim is a shared account of where demand is being lost, followed by a focused recovery plan.

Use the sequence even when the dashboard looks healthy. A falling cost per lead can conceal deteriorating lead quality. A higher cost per enquiry can be commercially acceptable when the enquiries are more relevant, opportunities progress and acquisition economics improve. The number is useful only when the business outcome beneath it is understood.

Establish a common funnel before reviewing campaigns

Define five stages in ordinary business language: enquiry received, enquiry accepted, opportunity created, proposal issued and customer won. Add a separate category for guide requests or other early interest. For each stage, write an entry rule, the system of record, the person responsible and the date captured. “Accepted” might require a supported geography, a relevant need and a contactable decision-maker or sponsor. Your definition should reflect how your business sells; a software trial and an enterprise consulting brief need different rules.

Select a recent cohort of enquiries and trace them through those stages. A cohort groups people by the period in which they entered, rather than mixing this month's spending with wins generated by earlier campaigns. Use a sufficiently mature period for your sales cycle. Where the cycle is long, report early indicators separately and avoid presenting incomplete revenue as a final verdict. Include the number of records in each calculation so that an apparently impressive rate does not hide a very small sample.

Create a reconciliation table with one row per unique enquiry. Record the advertising source where known, landing page, service requested, date received, owner, first human response, qualification decision and eventual outcome. An internal lead identifier can join records across systems. Keep names, emails and free-text messages inside the systems authorised to hold them. Management reports normally need the stage and source, not a copy of the prospect's personal information.

Agree how repeat enquiries will be counted. One buyer submitting twice is not necessarily two new opportunities. An existing customer asking for a different service may represent expansion revenue, which should be visible but separate from new customer acquisition. A single company can also contain several legitimate buying groups. Write the counting policy once and apply it consistently. A stable definition is more valuable than a dashboard that changes its meaning whenever someone wants a better result.

Verify the conversion signal before changing the bidding

Ask the account owner to list every conversion action, its trigger and whether it contributes to the goal used for optimisation. A click on a contact button is evidence of intent. It is not evidence that a form was accepted, stored or delivered. Similarly, visiting a thank-you URL can be misleading if the page can be reopened freely or reached without a valid submission. Start with an end-to-end test that follows a controlled enquiry from the browser into the actual destination system.

Test an empty form, an invalid email, a valid enquiry, a double click, a page refresh and a failed server response. Inspect both the website record and the measurement event. One accepted submission should create the expected number of records and one intended success signal. A failed request should show an understandable error and should not be counted as a lead. If a notification fails after the enquiry is stored, separate the delivery fault from the creation of the lead so that the record is recoverable.

Document ownership of each tag. Multiple plugins, custom scripts and imported analytics events can describe the same action. Do not remove an event simply because its name looks similar; verify the trigger, destination and reporting use first. Preserve a record of the original configuration, make a controlled change and compare the test outcome. If an advertising platform and the CRM disagree, investigate time zones, attribution windows, consent, duplicates and the event definition before assuming one is wrong.

For lead generation, a later qualification signal can be more useful than form volume alone. Google supports qualified-lead and converted-lead goals for offline outcomes. The implementation and data requirements should be checked against the current account setup. Google Ads: qualified leads and converted leads.

The practical deliverable is a one-page measurement contract. It names the primary business event, supporting events, acceptance rule, deduplication key, source system and test procedure. Use that contract when briefing an agency, changing a form or reviewing a new landing page. It keeps optimisation connected to a stable business definition.

Inspect the demand behind the enquiries

Take a sample of unsuitable enquiries and group the reasons they cannot progress. Useful categories include unsupported service, unsupported location, employment request, supplier pitch, education-only research, duplicate, spam, insufficient scope and not ready to act. Keep “unknown” available when the evidence is incomplete. A prospect who has not replied is not automatically a poor-fit lead. The sales team may have used the wrong contact method or followed up too late to establish the need.

Compare those reasons with the available search-term and campaign data. Look for repeated patterns rather than isolated phrases. An enquiry for a service you do not offer may reveal ambiguous targeting. Several applicants arriving through the same theme may justify exclusions. However, broad terms can also produce commercially valuable buyers who describe their problem differently from your team. Use the enquiry evidence to inform exclusions; do not turn every unfamiliar phrase into a negative keyword.

Review geographic targeting against where you can actually deliver. Separate a company's headquarters from the locations of its buyers. An enterprise procurement team may research from one country for a project in another. Review the current account's location settings and reports with that context. For local services, compare the area of demand with travel time, capacity and the locations you genuinely serve. A theoretical service area that the team cannot fulfil creates leads that sales will repeatedly reject.

Inspect device and schedule patterns alongside the website experience. Poor mobile performance may be a form problem rather than inferior mobile demand. An evening enquiry may be valuable even if it cannot receive an immediate human response. Before excluding a segment, ask whether the business can improve the response process. The decision should consider incremental opportunity, operational feasibility and the cost of serving the segment, rather than making the dashboard simpler at the expense of useful demand.

Align the offer with the buyer's level of commitment

A person researching a category may welcome a diagnostic guide and resist a sales meeting. A buyer comparing implementation partners may want a clear scope, relevant evidence and direct access to a conversation. A returning visitor may be ready to discuss budget and timing. When every ad leads to the same generic contact form, these different levels of readiness become indistinguishable. The resulting lead count says little about what the visitor intended to do.

Map each campaign theme to one primary offer. For urgent service demand, that may be a focused project enquiry on the service page. For a complex category, a useful guide can help the buyer frame an internal decision. Give the guide a separate conversion category and follow-up approach. Downloading an educational asset should not automatically trigger the same sales sequence as requesting an implementation proposal. The promise in the ad, the page and the first response should remain consistent.

Make the commercial scope understandable without placing a pricing barrier in every paragraph. Explain the type of engagement, the decisions made during discovery, the inputs required from the client and how implementation is scoped. A budget range can help establish fit, particularly for project work or media spending. Distinguish advertising spend from management fees and one-time implementation from ongoing support. Ambiguity at this stage creates uncomfortable conversations later and weakens the reliability the page is meant to convey.

Use evidence appropriate to the offer. A traffic screenshot may demonstrate discovery, while a CRM report may demonstrate opportunity progression. Name the metric, period and scope. If the only available evidence is platform activity, present it as platform activity. Senior buyers do not need every result to be a revenue claim. They need to see that the agency understands what each result means and can explain how the next stage will be measured.

Review the landing page as a buying decision

Open the exact destination used by the ad on a phone and a desktop. Read the first screen without the context of the campaign brief. Can a suitable buyer identify the service, recognise their problem and see the next action? Is there credible evidence near the claim that matters most? Does the page help the visitor decide whether this is a relevant professional engagement? A polished layout is helpful, but it must carry a clear commercial argument.

Keep the main conversion on the service page when possible. A visitor asking about Google Ads should not have to navigate through a general audit page and then another contact screen to explain the same need. Use a service-specific form or a clear anchor to it. If the visitor chooses Contact, carry the service context forward. This reduces repetition and gives the recipient a better starting point for the first conversation.

Review the form field by field. Ask what decision each field enables. Name and email support contact. Website provides business context. Budget helps scope the opportunity. A short message captures the actual problem. Consent explains the intended response. Additional fields may be justified for a complex request, but every requirement should serve the buyer or the qualification process. Do not make a prospect disclose a complete strategy simply to ask whether there is a fit.

Look below the hero. Explain how the work proceeds, what the business will receive, what must be available internally and how decisions will be reviewed. Add relevant questions as real headings so the page is easy to scan and navigate. Place contextual CTAs inside the explanation, particularly after a diagnostic sequence or evidence block. A reader who understands the problem at that point should be able to act without returning to the top or guessing which service to choose.

Follow the lead into the operating process

Trace the time between submission, assignment and the first useful response. An automated acknowledgement can confirm receipt, but it does not replace a human response to the prospect's actual need. Record both separately. Establish a response standard that the team can consistently meet, including coverage for different time zones and absences. The standard should be explicit enough that an unassigned enquiry becomes visible before it is forgotten.

Review the first response itself. Does it acknowledge the service requested, demonstrate that the message was read and propose a proportionate next step? A buyer who described a measurement problem should not receive a generic introduction to every service the agency sells. The response can ask one or two necessary questions and offer a useful conversation, while preserving the context already supplied. Requiring the prospect to repeat the form in a second system adds friction at the moment trust should increase.

Ask sales to record disqualification and loss reasons using a short, stable list. Separate lack of fit from lack of timing, lack of internal agreement, no response and a lost competitive decision. These categories lead to different improvements. A high rate of unsupported-service enquiries may call for targeting changes. A high rate of proposals lost on unclear scope may call for a better discovery and proposal process. Advertising cannot solve every problem after the form.

Review a small sample of records together each week. Include a successful opportunity, an unsuitable enquiry and one that is still unresolved. The discussion should examine evidence and identify a change, not allocate blame. A useful agency relationship makes it easier for marketing and sales to see the same journey. It should also reveal when the constraint is capacity, the offer or the internal decision process, so that spending is not increased into a bottleneck the organisation has not addressed.

Calculate the economics at the right stage

Cost per enquiry is advertising spend divided by unique enquiries. Cost per accepted enquiry uses the accepted subset. Cost per opportunity uses opportunities that satisfy the agreed sales rule. These measures answer different questions and should appear together when the funnel is being repaired. Keep the date basis consistent. For a mature cohort, add customers won, revenue recognised under the business's reporting policy and the relevant delivery or acquisition costs.

Consider an illustrative example, not a Prolinkage client result. Campaign A spends $10,000 and creates 100 enquiries, of which 10 become accepted opportunities. Campaign B spends $10,000 and creates 50 enquiries, of which 20 become accepted opportunities. Campaign A's cost per enquiry is $100 against $200 for B. Yet the cost per accepted opportunity is $1,000 for A and $500 for B. The cheaper enquiry has not produced the more efficient opportunity pipeline.

Do not assume that every opportunity has equal value. Segment by service, expected engagement size, margin and probability of progressing when there is enough reliable data to support the comparison. Use actual historical progression rates where possible. If management uses a forecast value, label it as a forecast and show the assumptions. Weighted pipeline is not booked revenue, and booked revenue is not necessarily cash received or contribution margin.

Before scaling, ask what happens if volume doubles. Can sales maintain response quality? Can delivery absorb additional work? Is the account dependent on a narrow set of queries or a temporary offer? What is the acceptable learning budget if the next cohort performs less well? This is where a campaign review becomes a capital-allocation decision. The purpose of measurement is to support that decision with appropriate confidence, not simply to create a more impressive performance slide.

Choose tests that distinguish one explanation from another

A useful test starts with a specific diagnosis. “Improve lead quality” is too broad. “Unsupported-service enquiries appear disproportionately in this campaign theme; clarify the service scope and review the search terms associated with those records” is more actionable. State the intervention, the expected change, the decision measure, the owner and the review date. Also state what you will preserve so that the result can be interpreted.

Avoid changing the offer, targeting, form, bidding and sales process simultaneously unless the current experience is clearly broken and requires a coordinated repair. Where several changes must happen together, document the bundle and be honest about the limits of causal interpretation. A before-and-after improvement can be useful operational evidence without proving that one headline caused the entire change. Seasonality, competition and sales capacity may also have moved.

Use guardrails that protect the business. A shorter form may increase submission volume while lowering contactability. A stronger budget question may reduce enquiries while improving accepted opportunities. An automated response may improve speed while creating incorrect promises. Decide which failure signals would stop or modify the test. The best experiment is not the one with the largest headline uplift; it is the one that helps the team make a defensible next decision.

Keep a decision log rather than a list of tasks completed. Each entry should explain what was observed, what changed, what happened and what will happen next. Over time, this becomes a record of how the account learns. It also prevents repeated tests when staff or agencies change. A leadership team should be able to ask why a campaign is structured a certain way and receive an answer grounded in accumulated evidence.

Run a focused thirty-day recovery programme

Use the first week to establish definitions and repair measurement. Reconcile a manageable sample of recent enquiries, verify the success event, inspect notification delivery and identify duplicates. Agree the service and qualification rules with sales. Produce the initial funnel table and record data gaps openly. The week is successful when the team can trust what an enquiry means and can find it in the system of record.

During the second week, investigate demand and the landing experience. Group unsuitable enquiries by reason, compare campaign themes and review the actual destinations. Fix obvious mismatches, broken forms and unsupported promises. Prioritise the changes that affect the largest relevant flow or the most valuable service. A long list of minor visual edits is less useful than one repaired path through which a suitable buyer can now complete a meaningful request.

Use the third week for a controlled test and sales-process follow-through. Confirm assignment rules, response ownership and the information passed into the first conversation. Review early outcomes without prematurely declaring success. If lead quality appears to improve but response times deteriorate, address that operational constraint. Maintain the distinction between an early signal and a sufficiently mature commercial result.

In the fourth week, present a decision brief. Show what is now measured reliably, the principal source of loss, the interventions completed, the evidence observed and the next investment recommendation. The recommendation may be to scale a focused segment, continue learning at the same budget, revise the offer or resolve an internal bottleneck before spending more. A strong review makes the decision clear even when the answer is to proceed carefully.

Use a working session to resolve the difficult cases

Bring ten to twenty representative enquiry records into a joint session with the campaign owner and sales lead. This is a practical starting sample, not a statistically representative study. Include different services, sources and outcomes. Remove personal information from the shared presentation where it is not needed. For each record, ask what the prospect appeared to want, what the business promised and what actually happened next.

Start with the difficult records rather than the easiest success stories. An enquiry may look unsuitable because the message is short, while a subsequent conversation reveals a well-funded project. Another may mention a large budget but request a service outside the business's capabilities. A third may be a genuine decision-maker who is collecting information for a later planning cycle. These distinctions help refine qualification without treating every early-stage buyer as a sales failure.

Ask the sales lead to describe the minimum evidence needed to accept an opportunity. Then ask marketing which of those signals can reasonably be collected on the page. The overlap becomes the form and routing brief. Some information belongs in the first conversation rather than the form. This negotiation is useful: it prevents a marketing team from optimising only for volume and prevents a sales team from expecting a fully qualified proposal request from every click.

End the session with no more than three decisions. Assign an owner and a date to each. For example, verify a duplicate submission trigger, clarify a service boundary on the landing page and introduce a response-time escalation for unassigned enquiries. Schedule the next review around those decisions. The discipline of completing a small number of consequential corrections is often more valuable than another broad account audit that never reaches implementation.

Prepare the brief for your next agency conversation

Provide the agency with the business objective, the services you want to grow, the geography you can fulfil and the constraints that matter. Explain the sales cycle, delivery capacity and what makes an engagement commercially viable. Share a scoped sample of qualified and rejected enquiries through appropriate access. The agency should be able to describe how it will turn that information into campaign, page and measurement decisions.

Ask for a phased scope. The first phase should resolve the most material uncertainty and establish a reliable baseline. Later phases can extend what proves useful. Require clear ownership of advertising accounts, website changes, measurement configuration and reporting access. A good working relationship should leave the business with understandable systems and a record of decisions, rather than dependence on unexplained dashboards or inaccessible accounts.

Discuss how the agency handles disappointing evidence. What would cause it to recommend a smaller scope, a different offer or a pause in scaling? How will it distinguish platform performance from sales execution? How will it communicate uncertainty without hiding behind it? These questions reveal whether the relationship is built around commercial judgement and implementation discipline. They are more informative than asking for a universal promise about lead volume.

Use the conversation to agree the next decision, not to purchase every service at once. A business that already has reliable measurement may need a focused demand and landing-page intervention. Another may need to repair the measurement and sales handoff before campaign changes can be evaluated. Start where the evidence indicates the greatest constraint, and keep the wider journey connected as that constraint improves.

Complete the enquiry-quality worksheet

Use a worksheet with one row per enquiry in the selected cohort. Begin with the unique record identifier and received date, then add campaign theme, service requested, landing page, acceptance decision and current sales stage. Include first-response time and the principal reason an enquiry did not progress. Keep a link to the authorised CRM record for investigation rather than duplicating personal details into a widely shared spreadsheet. The worksheet should make patterns visible without becoming a second, conflicting customer database.

Classify the records before calculating a rate. If two people disagree about whether a lead is qualified, discuss the actual need and record the rule that resolves the disagreement. That rule is part of the measurement definition. A consistent decision about ten ambiguous records can be more valuable than a sophisticated dashboard built on unreviewed labels. Keep unresolved cases visible until enough information is available, and avoid forcing them into a rejection category merely to complete the report.

Calculate the accepted-enquiry rate for each meaningful campaign or service group. Show the numerator and denominator beside the percentage. Then calculate the cost per accepted enquiry using the relevant spend and period. Add opportunity progression only for cohorts that have had enough time to advance. If a campaign produces a small number of records, use the worksheet to identify questions for investigation rather than declaring it the best or worst source on a thin sample.

Next, review the landing pages associated with the rejected reasons. If unsupported geography appears repeatedly, inspect what the page and ad imply about service coverage. If the scope is too small, inspect whether the page explains the type of engagement. If prospects ask for a service you do not offer, review the words used to describe adjacent capabilities. The worksheet should lead to a specific change in the experience, not simply a more detailed account of poor results.

Agree a follow-up check for each correction. The team should know which reason is expected to decline, which useful enquiries must be preserved and when the next cohort will be mature enough to review. A lower rejection count may come from lower overall traffic, so inspect rates and volume together. A higher qualification rate may still be commercially unhelpful if the campaign has become too narrow to create sufficient opportunity. Keep the business objective in view.

Use a decision table at the weekly review

If the website records fewer accepted submissions than the advertising account reports, investigate event definitions and duplicates before changing the acquisition strategy. If both systems agree but most enquiries concern unsupported services, investigate demand and offer clarity. If the enquiries are relevant but remain unassigned, fix routing and ownership. If discovery calls happen but proposals rarely progress, inspect qualification, scope and the buying process. Each observation points to a different accountable team.

Bring one proposed decision to the review rather than presenting every available metric. Explain the observation, the evidence, the alternative explanations and the recommended next action. State what would change your mind. This makes the review useful to a senior leader who has limited time but needs to understand the reasoning behind the investment. It also creates a professional standard for challenging assumptions without turning the conversation into blame.

Keep a short record of deferred questions. Some issues will require more data or access to a different system. Assign an owner and a date rather than repeatedly noting the uncertainty without resolving it. At the next review, begin with the decisions made last time and what happened after them. That continuity turns campaign management into an accumulating body of commercial knowledge.

References and working assumptions

The diagnostic sequence, sample workshop and thirty-day plan are Prolinkage's proposed operating framework. The numerical campaign comparison is illustrative and is not a client result. Apply the framework to your own sales cycle, sample size and account configuration. Platform controls and integrations evolve; verify current settings during implementation.

For platform-specific implementation, consult Google Ads guidance on qualified and converted leads, enhanced conversions for leads and Google Analytics recommended events. These sources support the platform concepts. The commercial decisions, qualification rules and operational responsibilities remain specific to the business.