What is a qualified lead, and who decides?
It is a contact that, by criteria marketing and sales agreed on, is worth a salesperson's time. The word doing the work there is “agreed”. If marketing qualifies on downloads and sales qualifies on budget, each side will see the other as the problem.
A useful definition has two parts:
- Fit. Who they are: industry, company size, job title, country. You know it from first contact and it rarely changes.
- Engagement. What they did: asked for a meeting, visited the services page several times, replied to an email. It changes over time.
A contact with plenty of engagement but no fit — a student downloading everything — is not a sales lead. One with perfect fit and no sign of interest is not one either, not yet. Keeping the two dimensions apart is what prevents that mistake. If your campaigns produce volume but few sales, it is also worth reading why your campaigns bring leads that do not buy.
Which lifecycle stages does HubSpot use, and what are they for?
According to HubSpot's official documentation, the default stages are: Subscriber, Lead, Marketing Qualified Lead (MQL), Sales Qualified Lead (SQL), Opportunity, Customer, Evangelist and Other. HubSpot describes them as the way to see where in the marketing and sales process each contact is, and how a contact passes from one team to the other.
| Stage | What it should mean in your company |
|---|---|
| Lead | Gave you their details or engaged beyond subscribing |
| MQL | Marketing considers them ready for sales by the agreed criteria |
| SQL | Sales accepted them as a possible customer |
| Opportunity | Has a deal associated |
| Customer | Has at least one closed won deal |
Two details the documentation clears up that tend to surprise people. The automatic updates HubSpot's tools make only move the stage forward; to go back to an earlier stage you have to clear the value first, by hand or with a workflow. And stages and statuses can be customized — though it is worth doing only when the real process demands it.
What is lead status for, if the stage already exists?
The stage says where the contact is in the whole journey. Lead Status spells out what is happening inside the sales stage. HubSpot documents it as the substatuses of SQL, and it ships with these values: New, Open, In Progress, Open Deal, Unqualified, Attempted to Contact, Connected and Bad Timing.
The documentation does not define each value, so define your own. “Attempted to Contact” might mean three attempts with no reply over five business days, and “Bad Timing” might require a review date. Without that precision, each rep fills the field to taste and the reports are worthless.
How does HubSpot's lead scoring work?
The lead scoring tool builds custom scores for contacts, companies and deals. Each score is stored in a property you can then use in lists, workflows and reports. Per the official documentation, there are four types:
- Engagement: points for actions, such as site visits, CTA clicks or email opens (contacts and companies).
- Fit: points for properties, such as job title, company size or annual revenue (contacts and companies).
- Combined: joins the two and creates three properties: total, engagement and fit (contacts, companies and deals).
- AI: generated from historical data, contacts only. Requires Marketing Hub Enterprise and at least 50 contacts, 25 converted and 25 not converted.
It requires Marketing Hub or Sales Hub, Professional or Enterprise. Points can be subtracted, so a score can be negative, and on engagement and combined scores you can apply decay so older actions weigh less, with intervals of 1, 3, 6 or 12 months.
How do you design the criteria without guessing?
Start from your actual customers, not from a hunch. A method that works with no AI involved:
- Take your last 10 to 20 won deals and as many lost or stalled ones.
- Write down what they had in common when they arrived: job title, size, industry, entry channel, first action.
- Turn whatever repeats into fit rules. What only the winners show adds points; what characterizes the losses subtracts them.
- Pick a handful of engagement actions — the ones that really do precede a sales conversation (asking for a meeting, visiting pricing or services, replying).
- Define the threshold that turns a contact into an MQL and test it against historical contacts before you switch the score on.
The tool lets you preview the score distribution before activating it. Use that: if 80% of your database sits above the threshold, the threshold is not filtering anything.
What makes sales ignore the score?
- Criteria sales never saw. If the commercial team was not involved, it will not trust the output.
- Too many rules. With twenty criteria, nobody can explain why a contact has 63 points.
- Empty properties. A fit score based on job title is useless if half your contacts have no job title. Check data quality first; the signs of a badly implemented CRM include exactly this.
- No follow-up on the result. If nobody compares the score against closed sales, it never gets corrected.
- Automating too soon. Assigning leads on score alone, with no initial manual review, multiplies the mistakes.
If you want to check the state of your database before starting, the CRM self-assessment and the CRM implementation guide cover the groundwork.