B2B Lead Scoring Thresholds: 75-100 Point Handoff (2026)

B2B lead scoring thresholds cluster in the 75 to 100 point band, the range most teams treat as sales-ready for handoff. Very few teams can say why their line sits exactly where it does. The score that triggers a handoff is an output of your own pipeline rather than a figure to adopt. Here is what the number encodes, and how to derive it.

What a 75 to 100 Point Score Means in B2B Lead Scoring

A 75 to 100 point score in B2B lead scoring means a lead has cleared the top band of a 100 point scale and is treated as sales-ready, which triggers routing to a named owner instead of another nurture email. The band is a status change, not a quality grade. Crossing it moves ownership from marketing to sales and starts a clock.

That distinction matters more than it sounds. A score of 78 and a score of 96 sit in the same band and get the same treatment, because the band exists to trigger an action rather than to rank leads against each other. Ranking happens inside the queue. The threshold decides who enters it.

The Three Score Bands in a 100-Point Model

The three score bands in a 100-point model are cold, warm and sales-ready, divided at roughly 50 and 75 points. Each band commits the business to a different action, and the action is the reason the band exists.

Table 1. Score bands and the action each one triggers

Band	Range	Common label	What the score triggers	Owner after the trigger
Cold	0 to 49	Unqualified	Automated nurture only, no rep time	Marketing
Warm	50 to 74	Engaged or early MQL	Targeted nurture, manual review at the top of the band	Marketing
Sales-ready	75 to 100	MQL for handoff, or SQL	Routing to a named rep with a response window attached	Sales

Read the right-hand column first. If two bands trigger the same action in your business, you do not have two bands, you have one line drawn in the wrong place. The most common version of this problem is a warm band that quietly gets worked by reps anyway, which makes the 75 point line decorative.

Why B2B Lead Scoring Thresholds Use a 100-Point Scale

B2B lead scoring thresholds use a 100-point scale because a fixed ceiling makes weights comparable across signals, converts every score into a readable percentage, and gives marketing and sales a shared vocabulary for arguing about the line. No standards body defines it. It is a convention that spread through marketing automation templates.

The convention has a cost worth knowing about. Published scoring models place the sales-ready line anywhere from the mid-60s to 80 and above, and longer enterprise sales cycles tend to push it higher still. Copying any one of those numbers imports a threshold that was calibrated against somebody else’s conversion data, somebody else’s rep capacity, and somebody else’s definition of a qualified lead.

Why 75 Points Became the Default Sales Handoff Threshold

Quick Answer: Seventy five points became the default sales handoff threshold because it reads as three quarters of a 100 point scale and sits above the ceiling of any single scoring dimension in an even 50/50 split. It is an inherited convention from marketing automation templates rather than a figure derived from conversion data.

Seventy five points became the default sales handoff threshold long before most teams had the data to test it, and what the number encodes matters far more than where it came from. A threshold is a probability claim: above this line, a lead is likely enough to convert that a rep’s hour is better spent here than on the next name in the queue. Every input that changes that likelihood should change where the line sits.

There are four of those inputs, and most teams only ever consider the first one.

Table 2. What a handoff threshold actually encodes

Input	The question it answers	What breaks when it is ignored
Conversion evidence	At what score does closed-won density visibly rise?	The line sits at a point where nothing changes
Sales capacity	How many leads can a rep work properly each week?	Reps cherry-pick, and the score loses authority
Cost of a false positive	What does a wasted rep hour cost against a missed buyer?	The line drifts toward whichever team complains loudest
Data completeness	What share of leads have the fields the model scores?	Good-fit leads score low because fields are blank

Work through those four before touching a point value. A threshold set on conversion evidence alone will be right on paper and unworkable in practice, because it says nothing about how many leads land on a rep’s desk on a Monday morning.

The fourth input deserves an extra note for outbound teams. Where leads come from a list you built rather than a form somebody filled in, behavioural fields start empty by definition, and the threshold has to be designed around that from the outset. I work through how the list itself gets assembled in our guide to LinkedIn prospecting.

How the 100 Points Are Divided to Make 75 Meaningful

The 100 points are divided to make 75 meaningful by giving fit signals and behaviour signals a fixed point budget each, then setting the line above the ceiling either dimension can reach on its own. This is the design rule that most scoring guides leave implicit, and it is the one that determines whether the threshold works.

If fit is capped at 50 points and behaviour is capped at 50, a score of 75 cannot be reached by either dimension alone. A perfect-fit lead with no engagement maxes out at 50. A highly engaged lead with poor fit maxes out at 50. Only a lead that is both a plausible buyer and actively signalling can cross 75. The threshold is doing real work at that point, because it encodes a condition rather than a total.

Move the line to 60 on the same model and the condition disappears. A lead can now clear it on fit plus one webinar registration. That is the mechanism behind most complaints that a scoring model sends sales the wrong people.

Fit Points: The First 50

Fit points make up the first 50 of the scale and answer whether a lead resembles the accounts you already close. They are mostly static, they can be filled from enrichment data before any engagement happens, and they are the half of the model that outbound teams can actually populate.

Table 3. Illustrative fit point budget, capped at 50

Fit signal	Points	Note
Job title matches a buying role in your ICP	15	The single strongest fit predictor in most B2B models
Seniority at director level or above	10	Drop to 5 where your buyer is a practitioner
Employee count inside your target band	10	Score the band, not the raw number
Industry on your target list	8	Use your own closed-won mix, not a generic vertical list
Country or region you sell into	7	Zero, not negative, for regions you do not serve
Fit subtotal	50

The point values above are a starting structure, not a benchmark. Defining which titles, bands and industries belong in your ICP in the first place is its own exercise, and I have covered that in our B2B lead qualification framework for ICP-based scoring.

Behaviour Points: The Second 50

Behaviour points make up the second 50 of the scale and answer whether a lead is signalling now. They are dynamic, they decay, and they are worth ranking by proximity to a purchase decision rather than by how easy they are to track.

Table 4. Illustrative behaviour point budget, capped at 50

Behaviour signal	Points	Note
Demo or pricing enquiry submitted	20	Highest-intent action available on most sites
Pricing page viewed twice or more in 14 days	10	Repetition is the signal, not the visit
Reply to a sales message that asks a question	8	A question outranks a positive sentiment reply
Webinar or event attended, not just registered	6	Attendance and registration are different events
Gated asset downloaded	3	Weak alone, useful as a tiebreaker
Two or more people from one account active in 30 days	3	Account-level signal, worth more than it looks
Behaviour subtotal	50

Notice how little a download is worth here. A model that pays 10 points for an ebook and 15 for a demo request is telling reps that three ebooks beat two demo requests, which is not true in any B2B business I have seen. Points that subtract, such as competitor domains, unsubscribes and bounces, sit outside this budget and deserve their own treatment, which we cover separately in our post on negative lead scoring rules.

When to Move Away from a 50/50 Split

Moving away from a 50/50 split makes sense when one dimension carries most of your predictive signal, which is usually a function of your sales motion rather than your industry. The 75 point line still works after the shift, provided neither dimension can reach it alone.

Table 5. Fit and behaviour weighting by sales motion

Sales motion	Fit budget	Behaviour budget	Why the split shifts
Inbound, self-serve product	40	60	Volume is high and behaviour separates buyers from browsers
Inbound, considered purchase	50	50	Both dimensions carry roughly equal signal
Outbound to a built list	70	30	Behavioural data barely exists before the first reply
Account-based, named accounts	60	40	Fit is pre-qualified, engagement breadth matters most

Check one thing after any reweighting: the larger budget must stay below the threshold. A 70/30 split with a 75 point line still holds, because 70 is under 75. A 80/20 split with the same line does not, because a lead can now clear the handoff on fit alone and arrive at a rep with no evidence of interest at all.

A Worked 100-Point Model with Three Routing Outcomes

A worked 100-point model produces three distinct routing outcomes, and the instructive one is not the lead that scores highest. Running three realistic leads through the fit and behaviour budgets above shows what the 75 point line accepts, what it rejects, and where it produces an answer people argue with.

Table 6. Three leads scored against the 50/50 model

Lead	Fit	Behaviour	Total	Band	Routing outcome
VP Sales, 400-person software firm in a target region, submitted a pricing enquiry, two pricing page views	50	30	80	Sales-ready	Routed to a named rep with a response window attached
Marketing manager, 60-person agency outside the target band, three asset downloads, one webinar	22	12	34	Cold	Stays in nurture, no rep time
CRO, 900-person fintech squarely in the ICP, one email click in 90 days	50	2	52	Warm	Held for targeted nurture, flagged for manual review

The third lead is the one worth sitting with. It is a perfect fit and an obvious name, and the model refuses to hand it over. That refusal is correct on the model’s own logic, because nothing in the record suggests this person is in a buying window. It is also the exact case where a rep will override the score, and where a manual review lane at the top of the warm band stops the argument becoming a fight about whether scoring works at all.

How to Set Your Own 75-Point Handoff Threshold in Five Steps

Setting your own 75-point handoff threshold takes five steps, and none of them start with choosing a number. The number is an output. It falls out of your closed-won distribution, your capacity, and the acceptance rules you are willing to enforce.

Step 1: Score Your Last 12 Months of Closed Deals

Scoring your last 12 months of closed deals is the first step because it is the only way to learn what your scale is actually measuring. Apply the finished point model retroactively to every closed-won and closed-lost deal from the period, using the data as it stood at the moment of handoff rather than as it stands today.

That last condition does most of the work. Scoring a won deal with the enrichment and engagement history it accumulated after the deal opened will inflate every winner and make any threshold look predictive. Use the snapshot at handoff, or the exercise tells you nothing.

Step 2: Find the Conversion Cliff, Not the Round Number

Finding the conversion cliff means locating the score above which win rate rises sharply, then setting the threshold at that point rather than at a tidy figure. Sort the scored deals into 5 point buckets, calculate win rate per bucket, and look for the bucket where the rate visibly steps up.

Sometimes that step lands at 75 and the convention is vindicated. More often it lands somewhere less convenient, such as 68 or 82. Take the number your data gives you. A threshold of 68 that reflects your pipeline beats a threshold of 75 that reflects a template, and rounding it up to look tidier throws away the only evidence you have.

Step 3: Test the Threshold against Sales Capacity

Testing the threshold against sales capacity means checking how many leads clear the line each month and comparing that to how many leads your reps can genuinely work. A threshold that passes the conversion test and fails the capacity test will be ignored within a quarter.

Table 7. Illustrative capacity check for a proposed threshold

Input	Example figure	Where it comes from
Leads scored per month	1,200	Marketing automation or CRM report
Share clearing the proposed line	14%	Model output against last quarter’s leads
Sales-ready leads per month	168	Multiply the two rows above
Reps available to work them	3	Headcount
Leads one rep can work properly per month	40	Rep interviews and activity data, not aspiration
Total monthly capacity	120	Multiply the two rows above
Gap	48 leads over capacity	Subtract

A gap that size leaves two honest options: raise the threshold until volume matches capacity, or add capacity. What does not work is publishing the lower threshold and hoping the overflow gets worked, because it will not, and the leads that go cold will be indistinguishable from the ones that were never any good.

This is where volume does the most damage, and I have watched it play out directly. In my personal experience, lead scoring based on establishing thresholds always leads to higher fit and better conversations overall. One client wanted to reach six thousand prospects in the UK. At that volume it was very evident that only a small portion of the six thousand were in market for their solution. Once we created the lead scoring model for that ICP-based list, we saw a seventy two percent increase in better meetings and better quality of sales calls.

Step 4: Write the Routing Rule and the Acceptance Criteria

Writing the routing rule and the acceptance criteria turns a threshold into an operating agreement, and it has to be written down before launch rather than negotiated after the first disputed lead. The rule names who receives the lead, how fast, and on what grounds it can be sent back.

State three things in plain language: the owner assigned when a lead crosses the line, the window that owner has to accept or reject, and the closed list of valid rejection reasons. Anything outside that list is not a rejection, it is an opinion, and it will not help you improve the model.

Step 5: Recalibrate on a Fixed Schedule

Recalibrating on a fixed schedule keeps the threshold tied to the pipeline it was built from, because both the pipeline and the market underneath it move. Put a quarterly review in the calendar and re-run steps 1 to 3 against the most recent completed quarter.

Fixing the date matters more than the frequency. Reviews that happen when someone complains happen only when relations are already strained, which is the worst possible condition for a conversation about weights and lines.

The Sales Handoff: Routing Rules for 75-Point Leads

Quick Answer: Crossing 75 points should assign a named owner, start a defined response window, and create an acceptance or rejection record. The score change alone is not a handoff. A handoff exists only when a specific person has taken responsibility for the lead inside an agreed timeframe.

The sales handoff for 75-point leads succeeds or fails on three mechanics that sit outside the scoring model entirely: how fast the lead reaches a person, how that person accepts or returns it, and which signals are allowed to skip the queue. Points decide who gets handed over. These three decide whether the handover survives contact with a working week.

Routing Speed and Owner Assignment

Routing speed and owner assignment determine how much of a lead’s value survives the crossing, and the decay is faster than most handoff SLAs assume. Assignment should be automatic and immediate, to a named individual rather than a shared queue, because a queue is where accountability goes to be diluted.

Table 8. Lead response speed and qualification odds

Finding	Figure	Basis
Odds of qualifying a lead contacted within one hour	Nearly 7 times higher than contacting later	Audit of 2,241 US companies
Average first response time among companies that did respond	42 hours	Same audit
Companies that never responded to the test lead at all	23%	Same audit

Source: The Short Life of Online Sales Leads, Harvard Business Review, 2011. The audit measured inbound web leads, so treat it as directional for other channels rather than as a benchmark to copy.

The practical reading is that the gap between a good threshold and a bad one is smaller than the gap between a one hour response and a two day response. If you can only fix one thing this quarter, fix the clock before you fix the weights.

Table 9. Response window design by trigger type

Trigger	Routing action	Response window	If the window is missed
Bypass signal, such as a pricing or demo enquiry	Assign to a named rep instantly, alert in CRM and chat	1 hour	Reassign to the next available rep, log the miss
Score crosses 75 with no bypass signal	Assign to a named rep	24 hours	Reassign, log the miss
Score 50 to 74 with a rising trend	Queue for weekly manual review	5 working days	Return to nurture

Acceptance, Rejection, and the Feedback Loop

Acceptance and rejection are what separate a status change from a handoff, and the rejection path is the more valuable of the two. A lead crossing 75 points becomes an MQL by arithmetic. It becomes a sales-qualified lead only when a rep looks at it and explicitly takes it on.

Give every rejection a structured reason from a fixed list: wrong fit, wrong timing, already a customer, competitor, no contactable person, duplicate. Those reasons are the highest-value data your scoring model will ever receive, because they tell you precisely which signal is over-weighted. A cluster of wrong-fit rejections points at the fit budget. A cluster of wrong-timing rejections points at the behaviour budget and at decay.

Bypass Triggers That Override the Point Total

Bypass triggers are the signals that route a lead to sales instantly regardless of point total, and every scoring model needs a short list of them. Their purpose is to stop the arithmetic from delaying a person who has already asked to speak to somebody.

Keep the list to three or four entries. A direct enquiry through a contact or demo form, a request for pricing, a reply asking to book time, and a named account on a target list are enough for most B2B businesses. The moment the bypass list grows past a handful, the threshold stops governing anything, and you are back to routing on gut feel with extra steps.

How to Keep a 75-Point Threshold Meaningful over Time

Quick Answer: A threshold stays meaningful when the share of leads clearing it, the acceptance rate behind it, and the score distribution of won deals all stay close to where they sat when the line was set. Drift in any of the three means the line no longer describes the same pipeline.

Keeping a 75-point threshold meaningful over time is a monitoring job rather than a modelling job, and it comes down to watching the same handful of numbers you used to set the line. Scores inflate naturally as tracking improves and as more signals get added, so a line that separated leads cleanly in January can be waved through by half your database by September.

Score decay is the standard defence: behaviour points expire after a set period of inactivity so that engagement from three months ago stops propping up a current score. Set the decay window to something shorter than your sales cycle, and treat it as part of the model rather than as maintenance.

Table 10. Threshold drift diagnostics

Symptom	What to measure	Compare it against	Likely fix
Sales rejects most handoffs	Rejection rate for leads above the line	The rate in the quarter the model launched	Raise the line, or reweight the fit budget
Reps work leads below the line	Share of worked leads scoring under the threshold	Your own routing log	Lower the line, the model is under-scoring real buyers
Won deals were below the line at handoff	Score at the moment of handoff for won deals	The distribution used to set the line	A predictive signal is missing from the model
Almost everything clears the line	Share of scored leads above the threshold	The share when the line was set	Points have inflated, add decay or cap categories

Track these four every quarter alongside the recalibration in step 5. Each one has a different fix, which is why measuring them separately matters. Teams that watch only the total volume of sales-ready leads see the symptom and guess at the cause.

Why 75-Point Handoff Thresholds Fail in Practice

Quick Answer: Most 75-point handoff thresholds fail for reasons that have nothing to do with the number. Scoring activity instead of intent, missing data fields, and an absent acceptance loop will each break a correctly placed line, and none of them are fixed by moving it.

The reasons 75-point handoff thresholds fail in practice cluster into four failure modes, and each one needs a different repair. Identifying which one you have matters far more than adjusting the line by five points in either direction.

Scoring activity instead of intent. Email opens, ebook downloads and repeat blog visits measure engagement with your marketing, not readiness to buy. A model built on those inputs will faithfully surface the most curious people on your list, which is a different group from the people with a budget and a deadline.

Incomplete data caps the score. A lead whose employee count and industry fields are blank cannot reach the fit points it deserves, so it sits in the warm band regardless of how good a buyer it is. Before reweighting anything, measure what share of your leads have every field the model scores. If that share is low, the fix is enrichment, not arithmetic.

No acceptance loop. Where crossing the line changes a field but nobody accepts or rejects anything, there is no record of whether the threshold is working. The model then gets judged on anecdotes, and anecdotes always favour whoever tells the story more forcefully.

The score describes a person, the decision belongs to a group. This is the failure mode that scoring cannot solve by design, and it is worth being honest about.

Table 11. Buying group composition and its effect on the handoff

Finding	Figure	Basis
B2B buyer teams showing unhealthy conflict during the decision process	74%	Survey of 632 B2B buyers, August to September 2024
Size of a modern buying group	5 to 16 people across as many as four functions	Same survey
Buying groups that reach consensus, against those that do not	2.5 times more likely to report a high-quality deal	Same survey

Source: Gartner Sales Survey, May 2025.

A single contact clearing 75 points tells you one member of a group that may run to sixteen people is engaged. That is worth acting on, and it is not the same as an account being in market. Adding an account-level view, where the scores of everyone from the same company are visible together, is the cheapest correction available. It is also why I treat a booked meeting rather than a scored lead as the real unit of progress, and why I start from account fit before any individual gets a point value at all.

Where a lead scoring model is the thing telling you which conversations to prioritise, the quality of the underlying list decides how much the model has to work with. A list built to fit produces a distribution where the top band is genuinely small and genuinely good. A list built for volume produces a distribution where the threshold is the only thing standing between your reps and six thousand names. Getting the list right first is most of the job, and it is the part we handle before any scoring conversation starts in our LinkedIn outreach programmes.

Setting a B2B Lead Scoring Threshold Your Sales Team Will Trust

B2B lead scoring thresholds in the 75 to 100 point band are a reasonable place to start and a poor place to stop. The number is only a summary of four decisions underneath it: what your closed-won deals actually scored, how many leads your reps can work, what a false positive costs you, and how complete your data is. Get those right and the handoff number almost picks itself. Copy someone else’s and you inherit their pipeline maths along with it.

The threshold conversation also has a limit. Scoring tells you which conversations to have first. It does not create the conversations, and it cannot rescue a list that was built for volume. That is the part I care most about, and it is why we build fit-first target lists and run human-written outreach rather than automated sequences, with marketing-qualified meetings on your calendar as the deliverable rather than a longer list of scored names. Across that work we have booked 4,000+ qualified meetings, generated $40M+ in pipeline and contributed to $4.5M+ in closed revenue for our clients. If you want to see how we approach it end to end, our LinkedIn lead generation guide covers the full picture, and both our done-for-you and team training engagement models are built around the same principle.

If your B2B lead scoring thresholds are surfacing leads your reps do not want, the problem usually sits upstream of the 75 to 100 point handoff line rather than in it. Book a Pipeline Fit Call and I will look at where your handoff threshold sits, what your list is feeding it, and whether the meetings coming out the other side are the ones you actually want.

Frequently Asked Questions