- Published
- Byline
- GoHighLevel Partners Editorial
- Filed under
- GoHighLevel Automation
A label that does not change how a lead is handled is decoration — here are observable rules for each category and what each one should trigger.
Warm and cold leads: classifying by evidence rather than by feel
Almost every sales process sorts enquiries into something like hot, warm and cold. The categories are useful because they let a limited amount of attention go where it will do most good. They are also, in most businesses, assigned by impression — a salesperson’s sense of how a call went — which makes them unreliable in exactly the situations where the sorting matters.
The fix is not a more elaborate scale. It is deciding, in advance, what observable facts put a lead into each category, and what each category actually changes about the way that lead is handled. A label that does not change the handling is decoration.
What the words normally mean
| Category | The evidence behind it | What it should change |
|---|---|---|
| Hot | Asked about price or availability, requested a call, gave a timeframe | Contact immediately, by the fastest channel they have given you |
| Warm | Engaged more than once, or matches the fit criteria without a stated timeframe | A person makes contact within a defined window, then a sequence |
| Cold | One passive interaction, or no evidence of intent yet | Automated nurture only; no manual time until something changes |
| Disqualified | Outside the service area, wrong need, cannot proceed | Removed from the pipeline and suppressed, not left to age |
The last row is the one most pipelines lack, and its absence is what turns “cold” into a bin for two different things: people who might buy later and people who will never buy at all. Those need opposite treatment, and mixing them makes every subsequent measurement wrong.
Two axes, not one
Temperature is usually treated as a single dimension, and it is really two: how well the lead fits what you sell, and how ready they are to act. A perfect fit with no urgency is a genuine future customer; an urgent enquiry from someone you cannot serve is not a lead at all.
Separating them changes the handling. High fit and high intent gets a person immediately. High fit and low intent gets patient nurture and no sales pressure. Low fit and high intent gets a fast, courteous no — which is worth more than it sounds, because it returns the time to the leads that can convert. Low fit and low intent gets suppressed.
Fit can usually be established at capture, with two or three questions on the form or through a qualifying quiz. Intent has to be observed over time, which is why these classifications should be dynamic rather than set once.
Base the classification on records, not recollection
The practical test for a good rule is whether two different people would apply it the same way. “Seemed keen” fails that; “asked for a price” passes. Write the criteria as things visible on the record: a page visited, a reply sent, a question asked, a form completed, a meeting booked.
Once the criteria are observable, they can be applied automatically, which is what lead scoring with a math operation is for: a value on the contact that moves as evidence accumulates, rather than a static tag someone set during a call. The important design choice is that the score should decay. Interest expressed six months ago is not evidence about today, and a scoring model that only ever adds points eventually marks the entire database as hot.
What each category should trigger
The label earns its place only when it changes what happens. Wire it into the routing rather than the reporting — that is the intake layer described in routing and suppression, where the classification decides which sequence, which owner and which response time applies.
- Hot: speed, and a human. The response window is the whole game here, and the standard is minutes rather than days — the discipline in a lead response playbook.
- Warm: a defined contact attempt, then automation. Two or three attempts by a person, then into a nurture sequence that ends in a reply rather than into a salesperson’s memory.
- Cold: automation only, with an escalation trigger. The value of the cold pool is that it is watched cheaply. Define what promotes a cold lead — a reply, a pricing page visit, a booking — and let that promotion happen without anyone noticing it manually.
- Disqualified: closed and suppressed. Leaving these in the pipeline distorts every rate you calculate, as measuring conversion by stage explains.
The mistakes that cost the most
Treating source as temperature. It is tempting to mark everything from a paid social funnel as cold and every referral as hot. Source correlates with quality, but a paid enquiry asking about availability tomorrow is hot by any sensible definition. Use the evidence on the record, and use source as one input rather than as the verdict.
Letting warm leads cool through neglect. Most warm leads do not go cold because they lost interest; they go cold because nobody followed up and the pipeline had no mechanism to notice. That mechanism is a stale opportunity trigger, and it should surface anything that has not moved rather than relying on someone remembering.
Never revisiting cold. A cold lead is a statement about a moment, not a person. Circumstances change, and the deliberate way to test that is a reactivation campaign rather than continued silence.
Classifying once, at capture. A tag set on day one and never updated describes a state that no longer exists. Make the classification a consequence of current evidence, and let it move in both directions.
Check whether the labels predict anything
The classification is worth having only if the categories convert differently. Measure it: take a cohort, look at what each category actually produced, and compare. If hot and warm convert at similar rates, the boundary between them is not carrying information and one of them can be removed. If cold converts almost as well as warm, your definition of cold is wrong and you are under-serving a real segment.
That comparison needs stages that mean something, which is the groundwork in stages with real exit criteria, and a record that survives someone editing a tag — exporting the cohort into a durable record of what happened keeps the history readable. The end state is unglamorous and useful: three or four categories, each with a written rule anyone can apply, each triggering a different and specific response, and each demonstrably converting at a different rate.
Frequently Asked Questions
What is the difference between a warm and a cold lead?
A warm lead has shown evidence of intent — engaged more than once, matches your fit criteria, asked something specific — while a cold lead has had one passive interaction or none. The useful distinction is not how the conversation felt but what is observable on the record.
Should leads be classified by source?
Only as one input. Source correlates with quality, but a paid enquiry asking about availability tomorrow is hot by any sensible definition, and a referral who is just browsing is not. Classify on the evidence the record carries and treat source as a contributing factor.
Why treat fit and intent as separate axes?
Because they need opposite handling. A perfect fit with no urgency deserves patient nurture and no pressure; an urgent enquiry from someone you cannot serve deserves a fast, courteous no. Collapsing both into one temperature hides which of the two situations you are in.
Should a lead score decay over time?
Yes. Interest expressed six months ago is not evidence about today, and a model that only adds points eventually marks the whole database as hot. Let the score fall with inactivity so the classification describes the present.
What should happen to cold leads?
Automation only, with a defined escalation trigger — a reply, a pricing page visit, a booking — that promotes them without anyone watching manually. The value of a cold pool is that it can be maintained cheaply and revisited deliberately later.
How do you know the classification is working?
Take a cohort and compare what each category actually produced. If hot and warm convert at similar rates, the boundary carries no information. If cold converts nearly as well as warm, the definition of cold is wrong and a real segment is being under-served.
We build your GoHighLevel account for you
CRM, pipelines, funnels, SMS and automation — configured for your business, not left for you to figure out.
- Setup (one-time)$850
- Ongoing (per month)$38.80
No obligation quote · reply within 1 business day
This site is independently operated and is not officially affiliated with, endorsed by, or sponsored by HighLevel LLC. We do not receive referral commissions — this is our own flat-fee implementation service.