It’s easy to assume your HubSpot lead scoring is functioning as it should, but if sales reps are pushing back on lead quality, or you’re struggling to prove marketing influence on pipeline, your scoring isn’t accurately surfacing leads worth pursuing.
7 mins
If you’re like most customers, your HubSpot lead scoring was configured once during implementation, hasn't been updated since, or was inherited from someone who's since left the business. Now is the time to review your scoring model and adjust it to suit your current business, audience, and goals.
This post covers the HubSpot lead scoring best practices that ensure your setup is aligning teams and delivering results, from defining what your scoring criteria should be based on, through to engagement scores in HubSpot's updated lead scoring tool, automating lead routing, and using data to have conversations about marketing ROI.
You can also access the recording of our recent webinar: How to Build a Scoring Model Sales Teams Will Trust.
Lead scoring in HubSpot assigns numerical values to contacts based on two things: who they are (referred to as fit) and what they do (referred to as engagement). Fit criteria includes things like job title, company size, and industry, while engagement criteria includes actions, such as visiting a pricing page, attending a webinar, or submitting a demo request. Essentially, we want to know how interested they are in us, and how interested we should be in them.
The combination of these two scores gives you a way of identifying which contacts are likely to become customers and how ready they are for a sales conversation, as opposed to treating every lead as equally important.
HubSpot's lead scoring tool has changed considerably over the past year or so. The legacy HubSpot Score property stopped updating in August 2025, which means any team still relying on the old single-score model is working with stale data. The new tool lets you create separate fit and engagement scores, combined scores, and scoring across contacts, companies, and deals, which is a step up from the previous approach and gives you much more control over how leads are evaluated and categorised.
But the HubSpot feature is only part of the lead scoring process.
The value of lead scoring is that it forces marketing and sales teams to sit down and agree on what a "qualified" lead looks like. This shared definition can then feed into automating lifecycle stages, routing leads correctly, and marketing attribution.
Without a shared ‘lead’ criteria, marketers generate what they class as marketing-qualified leads (MQLs) only for Sales to ignore HubSpot or CRM notifications because the contacts being passed over don't match what they know will convert, or for Sales to waste hours calling leads with no intention of buying.
For instance, a professional services consultancy with a 200-person team might find that its scoring model has been flagging junior researchers who download whitepapers as MQLs, when in reality, the deals that close almost always involve a Managing Director engaging with proposal-related content. Lead scoring is active, but it isn’t aligned to the business.
The HubSpot interface makes it easy to jump in and start adding scoring criteria and assigning points, but if you haven't agreed on what you're scoring and why, you’ll end up with weak leads being passed to sales reps.
So, first things first: define your scoring criteria.
A good starting point is your closed-won data. Pull up the last 20 to 30 deals your sales team has closed and look for what those contacts had in common: their job titles, company sizes, industries, the content they engaged with, and the actions they took before entering the pipeline. This gives you an evidence base for your scoring criteria rather than assumptions.
Defining your ideal customer profile should be a joint exercise between marketing and sales teams, because the two often have different views of what "ideal" looks like.
Marketers often build personas based on research and market analysis, while sales reps have first-hand experience of who buys, and your lead scoring model needs to combine both perspectives.
It's also important to separate your ‘must-have’ qualification criteria from your scoring. If a contact absolutely has to be in a particular industry or above a certain company size to be worth pursuing, they should be filtered out in an exclusion list.
For example, a manufacturing company might meet to define lead criteria, and realise that their sales team only closes deals with procurement managers and plant directors at companies with 500 or more employees, but marketing has been scoring all form submissions equally, regardless of role or company size. The existing model hasn’t been reflecting what converts, meaning the scores are surfacing low-quality leads.
Tip: Agree on what constitutes an MQL, an SAL, and an SQL in your business, and document the score thresholds that trigger each stage change in a simple document shared between marketing and sales teams.
A guide for marketing leaders using HubSpot to prove ROI, secure budget, and run creative campaigns without the tech headaches.
Once your lead scoring criteria are defined, you can start building the scoring model in HubSpot.
HubSpot lets you score fit and engagement separately before combining them into a single view, which is a much more useful approach than the old single-score model because it allows you to distinguish between a contact who looks like a great match but hasn't shown much interest, and one who's highly engaged but isn't the right fit.
Fit scoring is about identifying whether a contact matches the demographics of someone likely to buy. The most common attributes are job title or seniority level, company size by employee count or revenue, industry, and geography, although the specifics will depend on your business and what your closed-won analysis told you.
Tip: HubSpot allows you to exclude up to five lists of contacts or companies from being scored, which is helpful for keeping competitors, partners, students, and personal email addresses out of your scoring model without needing to assign them negative points.
For example, a company selling a financial compliance platform might assign 20 points for a CFO or Head of Finance title, 15 for the financial services or insurance industry, and 10 for a company size of 100 to 1,000 employees. A marketing coordinator at a 10-person startup would score low on fit even if they're downloading every piece of content on the site, and the sales team would never pursue that contact.
Engagement scoring captures what contacts are actually doing and how they’re interacting with your brand. The key is to score actions according to how strongly they indicate buying intent rather than treating all engagement as equally meaningful.
High-intent actions could include:
pricing page visits
demo or consultation requests
case study views
ROI calculator usage
These should carry the highest scores because these are the things people tend to do when they're seriously evaluating whether to buy.
Medium-intent actions might include:
webinar attendance
multiple blog visits in a short window
email click-throughs on bottom-of-funnel content
These indicate interest without necessarily signalling readiness to buy.
Low-intent actions should include:
a single blog visit
a social interaction
a newsletter open
These might contribute to the overall picture, but they shouldn't push a contact toward MQL status on their own.
Engagement scores will vary for different companies, and the diagram below differs from the examples we gave above, but it’s a useful diagram to help view what scoring is; understanding how far along the journey a prospect is from finding out who you are to signing off on the project/product.
Another factor to consider is time decay.
HubSpot's scoring tool supports time-based criteria, which means a pricing page visit from last week can carry more weight than one from three months ago. Without decay, your database fills up with contacts sitting at artificially high engagement scores based on activity that happened months ago, and sales reps end up chasing leads no longer interested.
A 30-day decay window works well for most engagement criteria, though businesses with longer sales cycles might want to extend to 60 days.
A high-tech SaaS company might score a pricing page visit at 20 points, a demo request at 25, and a case study download at 10, while a single blog post view only earns 2 points. With a 30-day decay window in place, a contact who was very active two months ago but has since gone quiet will drop back down the scoring scale.
HubSpot's combined scoring model maps fit and engagement into a grid where A1 represents the highest fit combined with the highest engagement, and C3 represents the lowest on both dimensions. A1 and A2 contacts should go straight to Sales, B1 contacts can be added to a targeted nurture sequence, and everything else stays in general marketing programs unless something changes.
Not sure where to start with scores? A straightforward approach is a 100-point total, split evenly between fit and engagement (50 points each), with the balance adjusted based on what your data shows after the first 30 to 60 days. If high-fit contacts convert even with moderate engagement, you might shift the points toward fit; if engagement turns out to be the stronger predictor, you shift the other way.
The advantage of getting HubSpot lead scoring right is what happens after the score is calculated, because this is where it can save time and reduce friction between teams by streamlining lead handoff.
Using HubSpot workflows, automate lifecycle stage changes based on score thresholds, so that when a contact hits your agreed MQL threshold, their lifecycle stage updates automatically, and the assigned sales rep receives a notification with useful context, i.e. the contact's company, their fit score, and their most recent engagement activity.
You can also use scores to route leads to the right people. For example, A1 and A2 contacts go directly to senior account executives, while B1 contacts are routed to business development reps for further qualification. This means your most experienced reps are spending their time on the contacts most likely to convert, rather than working through a shared inbox on a first-come-first-served basis.
Tip: Use engagement score resets. After a deal is marked as closed-lost, use a workflow to reset the engagement score (using the action CRM > Reset marketing lead engagement score) so the contact will re-enter nurturing programs with a clean slate rather than sitting at a score based on old activity.
To ensure we’ve covered all bases, here's how scoring might work across three different industries that MarCloud often works with:
A professional services firm targeting Managing Directors at companies with £10M or more in revenue might assign 20 fit points for a director-level title, 15 for the right revenue bracket, and 10 for the professional services industry. On the engagement side, a proposal request earns 25 points, a case study page view earns 15, webinar attendance earns 10, and a blog visit earns 3, all with a 30-day decay window. Any contact reaching A1 or A2 is automatically routed to a named partner.
A manufacturer selling to facilities and operations teams across sectors like logistics, food production, and pharmaceuticals might weight the job title at 20 fit points for procurement managers or operations directors, company size at 15 for organisations with 500 or more employees, and industry at 10 for their core verticals. A product demo request earns 25 engagement points, a pricing page visit earns 20, a technical specification download earns 10, and an email click-through earns 5. A1 or B1 contacts are handed off to regional sales reps.
A fintech company selling compliance and reporting software to finance teams across regulated industries might score 20 fit points for CFO or Head of Finance titles, 15 for industries like banking, insurance, or healthcare where regulatory requirements drive purchasing decisions, and 10 for a company size of 100 to 1,000 employees. Engagement scoring gives 25 points for a compliance guide download, 20 for completing an ROI calculator, 15 for a pricing page visit, and 5 for newsletter engagement, with a 60-day decay window to reflect the longer sales cycle. A1, A2, or B1 contacts trigger assignment to a specialist sales consultant.
Your HubSpot scoring model needs regular attention, or else it’ll lose its effectiveness as your business, content, and audience evolve. We recommend quarterly audits as a general best practice.
The most useful metric to keep an eye on is your MQL-to-SQL conversion rate, because if sales is accepting fewer than half of the MQLs being passed over, either the threshold is set too low or the criteria don't match what sales considers qualified.
Tracking lead response time matters too. If reps aren't following up quickly on scored leads, the issue might not be the model itself but the notification and routing process around it.
Score inflation is another thing to watch for. If a large proportion of your database is accumulating high scores without converting, it usually means engagement criteria are too generous or time decay isn't doing its job properly.
HubSpot's built-in score preview and distribution tools let you see how contacts sit across score ranges, and if most of your database sits in the same narrow band, the criteria aren't differentiating well enough.
There are a few patterns that come up regularly when lead scoring isn't delivering what it should.
Scoring all actions equally: This is probably the most frequent. A blog visit and a demo request represent fundamentally different levels of intent, and weighting them the same means your pipeline fills with contacts who aren't ready for a sales conversation.
Building the model without involving Sales: Because if the sales team doesn't agree with the criteria, they'll simply ignore the scores, and the whole thing becomes an exercise that only marketing cares about.
Over-complicating the model: Start with five to seven core criteria that predict the majority of conversions, which gives you something workable that you can refine over time.
Not using time decay or list exclusions: This means your scores will include inactive contacts and irrelevant audiences, which reduces confidence in the system across both teams.
Not reviewing regularly: Your model should evolve. Marketing and Sales should align regularly to understand how the model needs adjusting; both sides are more likely to have buy-in when it's a collaborative effort.
When you have a robust scoring model in place, you can build reports in HubSpot to show conversion rates by score range, average time from MQL to closed-won by score tier, and the revenue attributed to leads that met your scoring criteria.
Being able to say that 72% of closed-won revenue last quarter came from contacts that hit your A1 or A2 threshold, and that those contacts were sourced from specific campaigns, is a concrete argument for continued investment.
Lead scoring also shows you which campaigns and content types are generating high-fit, high-engagement leads versus which are driving volume without quality, and that distinction helps you make better decisions about where you’re delivering HubSpot ROI and, therefore, where to put your creative energy and budget.
Navigate to the lead scoring tool. In your HubSpot account, go to Marketing > Lead Scoring and click Create score. Select the object you want to score (contacts, companies, or deals) and choose whether you're creating a fit score, an engagement score, or a combined score. For most teams, a combined contact score is the best starting point.
Set your score limits. Choose a maximum score range using the Score limit dropdown. A range of 0 to 100 works well for most businesses. If you're using a combined score, HubSpot will split this across fit and engagement automatically.
Build your fit criteria. Click + Add property group, then add property rules for your ICP attributes. Set the point values for each rule. Use group limits to cap how many points the fit criteria can contribute to the overall score.
Build your engagement criteria. Click + Add event group, then add event rules for the actions that signal buying intent. Assign higher point values to high-intent actions and lower values to general engagement. Use the timeframe and decay settings to ensure scores reflect recent activity.
Set up list exclusions. Add up to five lists of contacts or companies to exclude from scoring entirely.
Configure your thresholds. For combined scores, HubSpot maps fit and engagement into a grid from A1 (highest) to C3 (lowest). Go to Settings to customise the numeric ranges that define each tier.
Test your score against known contacts. Use HubSpot's test feature to check how your model scores contacts you already know well (a recent closed-won customer should score highly, and a known bad-fit contact should score low). If either doesn't seem right, adjust your criteria and point values before going live.
Build your routing workflows. Create workflows that trigger lifecycle stage changes and sales notifications when contacts hit your agreed thresholds. Include an engagement score reset workflow for closed-lost deals.
Preview your score distribution. Use the score preview tools to check how contacts cluster across your score ranges. If the majority of your database lands in the same narrow band, your criteria aren't differentiating well enough, and you'll need to adjust.
It does take some thought and cross-team alignment to get lead scoring right, but the payoff is a pipeline that sales reps trust and confident reporting, both of which make life considerably easier for marketing teams!
If you'd rather have someone help you configure it properly from the start (particularly if you're working with a complex CRM integration or multiple product lines), speak with a specialist. MarCloud partners with marketers to build lead scoring models that align with their process and reporting needs.
Lead scoring assigns numerical values to contacts based on how well they match your ideal customer profile (fit) and how they've been interacting with your content and website (engagement), so marketing and sales can prioritise the leads most likely to convert.
Navigate to Marketing > Lead Scoring, click Create score, choose your object and score type, then add property groups for fit criteria and event groups for engagement criteria with point values assigned to each.
The most effective criteria come from your own closed-won data, but common ones include job title, company size, and industry for fit, and pricing page visits, demo requests, and case study views for engagement.
Start with a 100-point total split evenly between fit and engagement, define which grid positions (such as A1 and A2) trigger MQL status and sales routing, then adjust after 30 to 60 days based on your MQL-to-SQL conversion rate.
Yes, the lead scoring tool supports contacts, companies, and deals, with company and deal engagement scores based on the actions of associated contacts.
Fit scoring evaluates whether a contact matches your ideal customer profile based on attributes like job title and company size, while engagement scoring evaluates what they've actually done, such as visiting your pricing page or requesting a demo.
Founder & CEO of MarCloud, Tom has been on both sides of the fence, client-side and agency, working with Salesforce platforms for the best part of a decade. He's a Salesforce Marketing Champion and certified consultant who loves to co-host webinars and pen original guides and articles. A regular contributor to online business and marketing publications, he's passionate about marketing automation and, along with the team, is rapidly making MarCloud the go-to place for Marketing Cloud and Salesforce expertise. He unapologetically uses the terms Pardot, Account Engagement and MCAE interchangeably.
More by Tom RyanA guide for marketing leaders using HubSpot to prove ROI, secure budget, and run creative campaigns without the tech headaches.
Download nowThere’s a trend in the conversations we’re having with marketing leaders these days. Teams that have been using Salesforce Account Engagement (Pardot) or Marketing Cloud Engagement for years are looking at alternatives, and this typically leads them to two options. Either they stay within the Salesforce ecosystem by moving to Marketing Cloud Next (one of Salesforce’s ‘on core’ marketing platforms), or they switch to HubSpot.
Tom Ryan
8 mins
If you've been using Marketing Cloud Account Engagement (Pardot) for a few years, you'll know the product roadmap has stayed very, very still. Feature requests from years ago haven't arrived, leaving workarounds as permanent fixtures. Many marketers feel as though the platform they invested in isn't keeping up with what teams need it to do.
Tom Ryan
5 mins
Most leaders reviewing marketing performance in HubSpot will eventually find themselves asking whether the data tells the full story. Even if you’re successfully tracking form submissions, original source, campaign associations, and revenue attribution, there are moments when you wish you had just a little more context around how that contact arrived in the system and what influenced them. This is where HubSpot hidden fields are useful.
Tom Ryan
5 mins
MarCloud is a team of certified Pardot, Marketing Cloud, and Salesforce specialists. We help businesses to unlock the potential of marketing automation. Join 2,000+ professionals who receive actionable insights to boost their automation performance.
Sign up to the newsletterA guide for marketing leaders using HubSpot to prove ROI, secure budget, and run creative campaigns without the tech headaches.
Download now
