You can find where leads drop off by counting five stages over the same period: enquiries received, enquiries replied to, real conversations started, quotes or proposals sent, and deals closed. Put the totals in order, calculate the conversion from each stage to the next, and inspect the largest meaningful fall before changing your marketing budget.
Most owners can name two numbers quickly. They know what they spent on marketing, and they know what revenue eventually closed.
Those numbers matter, but they sit at opposite ends of the journey. They do not show what happened between an enquiry arriving and a customer buying.
Picture a five-milestone board. A gold pawn starts at enquiry and should move through reply, conversation, proposal, and deal. A broken square can stop it before the next milestone, while the later milestones remain visible but out of reach.
That is the problem with judging marketing from spend and sales alone. A weak result can come from the campaign, but it can also come from a missed reply, a reply that did not start a conversation, a conversation that never reached a proposal, or a proposal that did not close.
When only the starting stake has a report, the campaign gets blamed. That can lead a business to increase spend, replace an agency, or rewrite an advert before checking whether the real break sits in response handling, sales conversations, proposals, or follow-up.
The three practical ideas for this week fit together:
That sequence matters. Measurement is useful when it changes the next question, not when it merely produces another dashboard.
A lead-stage handoff is complete only when the lead meets the agreed condition for the next stage and someone owns the next action. A sent message is not automatically a conversation, and a positive conversation is not automatically a proposal.
Give every stage three things: a definition, an owner, and a timestamp. The definition tells the team what qualifies. The owner prevents the lead from sitting between people. The timestamp shows how long the handoff took.
This distinction separates activity from progress. Enquiries received and replies sent are activity. Conversations and proposals show movement. Closed revenue is the result. Counting all three levels prevents a busy stage from being mistaken for a healthy pipeline.
Our view is simple: do not increase the budget until the business can explain where the current leads stop moving.
Start with five counts that most businesses can collect without buying new software.
The first is enquiries received. Count every genuine enquiry at arrival, across forms, email, telephone, direct messages, and any other channel customers use.
The second is enquiries actually replied to. Do not count an automated receipt as a reply. Count a response sent by someone who can move the conversation forward.
The third is real conversations started. A polite acknowledgement is not yet a conversation. Count the enquiries where the customer responded, answered a useful question, or agreed to a next step.
The fourth is quotes or proposals sent. This is where general interest becomes a specific commercial decision.
The fifth is deals closed. Most businesses already hold this number. Its value changes when you can see how many people entered each earlier stage.
Use one consistent definition for each stage:
Stage | Count it when | Do not count
The exact definitions may differ by business. What matters is that the team uses the same definitions for the whole measurement period.
A spreadsheet is enough for the first pass. Give each enquiry one row and each stage one column. Use a lead ID rather than copying unnecessary personal details into another system. Record the enquiry date, source, owner, reply date, whether a real conversation started, proposal date, closed status, and a simple loss reason when known.
Name one person who owns the count. If five people update it differently, the sheet becomes another broken square.
Keep the definitions simple and write them at the top. What qualifies as a genuine enquiry? What qualifies as a reply? What turns a reply into a conversation? What counts as a quote? What counts as closed?
The first month will be imperfect. That is acceptable. Missing information is itself useful because it identifies a stage your current process cannot see.
A practical starter sheet can contain these columns:
Do not turn the first sheet into a miniature CRM. Its purpose is to answer one question: where does movement stop?
The goal is not reporting theatre. The goal is to make the middle visible enough to choose the next investigation.
For every transition, divide the number that reached the next stage by the number in the previous stage, then multiply by 100.
Stage conversion rate = next-stage count ÷ previous-stage count × 100
If 40 enquiries arrived and 30 received a useful reply, the enquiry-to-reply conversion rate is 75 percent. If 30 replies produced 12 conversations, the reply-to-conversation rate is 40 percent.
Keep both the percentage and the underlying count visible. A 50 percent fall from two enquiries to one is not the same operating signal as a 50 percent fall from 100 enquiries to 50. Small volumes need a longer observation window and more caution.
Place the five totals in order and compare each stage with the one before it.
If many enquiries arrive but fewer receive a reply, inspect ownership, inbox coverage, and response handling.
If replies are sent but few become conversations, read the actual first messages. Check whether they answer the customer's question and offer a clear next step.
If conversations happen but few proposals follow, inspect qualification, handoffs, and whether the offer is being made clearly.
If proposals go out but few deals close, inspect the proposal, decision process, timing, and follow-up after the quote.
The largest fall does not prove the cause. It identifies the broken square most worth inspecting.
Do not select a stage from percentage alone. Consider four questions:
A smaller percentage drop at a high-volume stage may matter more than a dramatic swing caused by two low-value enquiries.
This is where Week 2 connects to Week 3. A technically sent reply can still fail if the prospect no longer remembers the original enquiry or cannot see why the response is relevant.
Five counts are a triage tool, not a complete measurement system.
They do not tell you whether every enquiry was equally suitable. They do not explain why a customer stopped replying. They do not separate price, timing, trust, urgency, and competitor choice. They also do not prove that the stage with the largest fall is the only stage worth improving.
A hand count can be distorted by inconsistent definitions or missing records. Small monthly volumes can make one lost deal look more important than it is. Different services may also need separate counts because their journeys are not comparable.
Use the five stages to locate the next question, not to manufacture certainty.
The count also does not provide perfect marketing attribution. A prospect may see several pieces of content, return through direct search, speak to someone offline, and close later through a different channel. Stage tracking tells you how the lead moved after it became visible. Attribution asks which touchpoints influenced the journey. Both are useful, but they are not the same question.
Start manually when definitions are still changing or the team cannot reliably name the five counts. Automating an unclear process makes inconsistent data arrive faster.
Move toward a CRM or analytics integration when:
Google Analytics currently recommends a set of lead-generation events for measuring online and offline funnel activity, including generate_lead, working_lead, qualify_lead, and close_convert_lead. That does not mean those names must replace your operating language. It shows why tracking only a form submission is not the same as measuring the full lead journey.
If advertising optimisation later needs confirmed offline outcomes, Google Ads also supports enhanced conversions for leads. That is a later implementation decision involving tagging, first-party data, consent, and Google's data terms. It should follow clean stage definitions rather than substitute for them.
Once the largest fall is visible, inspect that stage closely.
Read the messages. Follow one enquiry from arrival to its last action. Ask the person who owns the stage what makes the work slow or unclear. Look for a contained repair that is useful even if another constraint remains.
Then hold the same five counts for another month. If the stage improves, keep the repair and look at the next fall. If nothing changes, your first explanation was incomplete and the evidence has told you to investigate again.
Use a simple review rhythm:
Only add software when the manual process has shown which decision the tool needs to support. A dashboard that automates unclear definitions only redraws the same broken route faster.
The simplest first step is one sheet, five counts, and an honest look at the largest fall.
Lead drop-off happens when a prospect reaches one stage of the journey but does not move to the next. Examples include an enquiry that never receives a reply, a reply that never becomes a conversation, or a proposal that never becomes a sale.
Count enquiries, useful replies, real conversations, proposals, and closed deals for the same period. Calculate the conversion rate between each adjacent stage. Inspect the largest meaningful fall, then read the underlying enquiries before deciding on a repair.
No. A spreadsheet is enough for an initial one-month diagnostic if volume is manageable and one person owns the data. A CRM becomes useful when volume, handoffs, reminders, permissions, and repeated reporting make the manual method unreliable.
Check stuck leads weekly and compare stage totals monthly. Businesses with high volume or short sales cycles may need more frequent reviews. Low-volume businesses may need a longer period before interpreting percentages.
Not necessarily. Advertising may be the problem, but the loss may occur after the enquiry arrives. Check reply coverage, conversation quality, proposal handoffs, and closing follow-up before treating the ad account as the automatic cause.
Start with the largest meaningful stage drop that affects enough leads, carries commercial value, sits within your control, and can be tested without destabilising the whole journey.
If you want help choosing which stage deserves that closer look, the GoCoreVantage Lead Journey Quiz is a starting diagnosis, not a conclusive answer. Use it to frame the next inspection before deciding what to repair.
Take the free Lead Journey Quiz — a 3-minute scored breakdown of where your leads are actually being lost.
Take the Free Quiz