Almost every CRM and inbox tool ships a lead-routing setting, and "round robin" is usually the box that's already ticked: new lead in, next rep out, everyone gets an equal turn. Picking it takes ten seconds. What happens to the lead after it lands in someone's queue is a completely separate question — and, on the evidence, the one almost nobody actually checks.
We went looking for the research behind two different claims that circulate in this space. First, the specific percentages — 15-30% higher close rates, 18% higher conversion, and similar — that get cited for routing leads by skill or performance instead of by turn order. Second, a much older and more specific number: that sales reps simply never pursue a large share of the leads marketing hands them. One of those turned out to be real, peer-reviewed, and about something routing settings alone can't fix. The other turned out to be unsourceable, repeated across dozens of blogs with no study behind any of them.
Key takeaways
- The one rigorous, peer-reviewed source in this space is Sabnis, Chatterjee, Grewal and Lilien's 2013 Journal of Marketing study. It opens by framing "the sales lead black hole — the 70% of leads generated by marketing departments that sales representatives do not pursue" — and then uses survey data from 461 sales reps to study what actually predicts whether an assigned lead gets worked: how well it was pre-qualified, whether a manager tracks follow-up, how many leads the rep is already carrying, and the rep's own experience and performance.
- "36% of leads are never called," a number that still circulates from a 2012 Forbes piece, traces to further work by the same researcher-and-vendor pairing — James Oldroyd and InsideSales.com — behind the well-known 2007 "100x more likely to qualify" lead-response study. Its author, Ken Krogue, was InsideSales.com's own co-founder at the time. Same lineage, same reason to read it as vendor-produced rather than independent.
- The specific conversion-lift percentages that circulate for "smarter" routing over plain round-robin — 15-30% higher close rates, 18% higher conversion on routed leads, 12-18% better first-contact resolution — could not be traced to any locatable, methodology-disclosed study anywhere we looked. Treat them as unverifiable marketing claims, not evidence.
- Round-robin's actual, guaranteed mechanical promise is narrow: every lead gets exactly one owner, and turns rotate evenly. That's it. The rule itself has no concept of time zone, business hours, a rep's current workload, language, deal size or who happens to be good at closing this particular kind of lead — which is the entire reason weighted round-robin, load-based routing and skill-based routing exist as separate methods, not add-ons to the same one.
- A lead with an owner and a lead that's actually being worked are two different facts, and the 2013 study's own findings say the gap between them is closed by process — pre-qualification and management tracking — and by the rep's own capacity and ability, not by which rotation rule happened to hand them the lead.
What are you actually choosing between when a CRM asks you to pick a routing method?
"Lead routing" or "lead distribution" is the rule that decides which person on a team gets assigned a new lead the moment it arrives — a form fill, an inbound WhatsApp message, a landing-page submission, a booked call. Every method answers the same one question — whose turn is it? — with a different amount of information.
- Manual / first-to-claim — a person picks it themselves, or whoever notices and grabs it first. No rule at all; entirely dependent on who happens to be paying attention.
- Strict round-robin — cycles through every rep in a fixed order, one after another, regardless of anything else happening on the team that day.
- Weighted round-robin — the same rotation, except each rep is given a set share of the total instead of an equal one, so a senior rep or a full-time hire can carry more of the queue than a part-timer without the rotation itself changing.
- Load-based / capacity routing — checks how many leads or open conversations each rep is currently carrying before deciding who's next, instead of just whose turn it is.
- Skill-based routing — matches an attribute of the lead itself (language, product interest, deal size, which ad or page it came from) to whichever rep is actually suited to it, before turn order enters the decision at all.
| Method | What it checks before assigning | What it structurally can't see |
|---|---|---|
| Manual / first-to-claim | Nothing — a person decides in the moment | Consistency. A lead sits unclaimed the moment nobody's watching, and speed of reaction wins over fit |
| Strict round-robin | Whose turn is next in a fixed order | Workload, time zone, business hours, rep skill or fit — a rep already buried in open leads still gets the next one |
| Weighted round-robin | Whose turn is next, adjusted by a pre-set share | The same blind spots as strict round-robin — the shares are fixed until someone updates them by hand |
| Load-based / capacity | How many open leads or active conversations each rep already holds | Whether the rep is actually a good fit for this specific lead — it balances quantity, not quality of match |
| Skill-based | Attributes of the lead against attributes of the rep | Fairness of workload, unless it's paired with a load check or a fallback rule for ties |
None of these is the "correct" method — they trade off different kinds of blindness. A two-person team gains almost nothing from skill-based rules it has to maintain by hand; a team fielding leads in three languages and two product lines loses real revenue to a strict rotation that can't tell those leads apart.
Where does "round robin" actually come from, and why is it every CRM's default?
The term predates software by a long way. A round-robin tournament is the format where every competitor plays every other competitor in turn, so no one gets an easier or harder draw than anyone else by luck of a bracket — a scheduling idea, not a sales one. Computing borrowed the same principle decades ago for an unrelated problem: when a single processor has to serve many waiting tasks, giving each one a fixed turn in a fixed order is the simplest rule that's provably fair without needing to know anything about what any individual task actually needs. Telecom and call-center routing borrowed it again for the same reason — distributing incoming calls or leads one at a time, in turn, needs no configuration and is trivially easy to explain to a team.
That's the whole appeal, and it's also the whole limitation. A rule that's fair by construction and needs zero information about the thing being distributed is exactly the kind of rule that becomes a default setting — and exactly the kind of rule with nothing to say about time zone, workload, or fit. The academic literature on call-center operations — surveyed in a widely-cited tutorial by Noah Gans, Ger Koole and Avishai Mandelbaum, published in Manufacturing & Service Operations Management in 2003 — treats routing discipline as a real, separate lever from staffing or speed, and skill-based routing specifically as the field's long-standing answer to a plain rotation's blindness to which agent actually fits which case. Sales tools arrived at the same idea decades later, under a different name.
Does getting a lead assigned actually mean somebody works it?
This is the question a routing rule doesn't answer by itself, and it's also the subject of the one genuinely rigorous, peer-reviewed study we could find in this entire space: "The Sales Lead Black Hole: On Sales Reps' Follow-Up of Marketing Leads," by Gaurav Sabnis, Sharmila C. Chatterjee, Rajdeep Grewal and Gary L. Lilien, published in the Journal of Marketing (Vol. 77, No. 1, January 2013, pp. 52-67).
The paper opens by naming the problem its title refers to: "the sales lead black hole — the 70% of leads generated by marketing departments that sales representatives do not pursue." That figure is the framing statistic the authors open with, not a number their own fieldwork re-derives — their actual dataset is survey data from 461 sales reps, and their question is narrower and more useful than the headline number: among leads that do get assigned to a rep, what predicts whether that rep actually spends time on them? Using a motivation-opportunity-ability framework, they find the answer comes down to four things: how well leads are pre-qualified before they reach the rep (this shapes how worthwhile a lead looks), whether a manager actively tracks follow-up (extrinsic motivation), the sheer volume of leads a rep is currently juggling (opportunity — or the lack of it), and the rep's own experience and past performance (ability).
The nuance is the useful part. More experienced reps, the study finds, respond less to managerial tracking and to raw lead volume, but respond more to how well a lead was pre-qualified — they've learned to trust (or distrust) the quality of what marketing hands them. Higher-performing reps put more effort into marketing leads overall, but are less swayed by managerial oversight than lower performers are. None of this is a routing-method finding — it's an argument that whether an assigned lead gets worked depends on process design and on who the rep is, not on which rotation rule decided whose turn it was.
70%
the share of marketing-generated leads sales reps do not pursue, per the framing given in Sabnis, Chatterjee, Grewal and Lilien's 2013 Journal of Marketing study — the opening problem the paper's own 461-rep dataset then goes on to explain, not a number the routing method a CRM picks can fix by itself.
Source: Sabnis, Chatterjee, Grewal & Lilien, "The Sales Lead Black Hole," Journal of Marketing, Vol. 77, No. 1 (2013)
This is the fact the rest of this piece follows from: an owner and a worked lead are two different things, and a routing rule only ever guarantees the first one.
What about the specific percentages people cite for "smarter" routing beating round-robin?
"36% of leads are never called" is the other number that still gets repeated from a 2012 Forbes piece by Ken Krogue, titled "The Great Marketing/Sales Disconnect." It's worth knowing where it comes from before treating it as independent evidence: Krogue was, at the time, a co-founder of InsideSales.com — the same sales-software vendor (later rebranded XANT) behind the widely-cited 2007 "Lead Response Management Study" and its "100x more likely to qualify" figure, produced with the same researcher, James Oldroyd. It's a real number from a real piece of vendor-affiliated research, in the same lineage as a claim most marketers have already learned to treat carefully in a different context.
The newer claims — that skill-based or performance-based routing beats plain round-robin by 15-30% on close rate, or by roughly 18% on conversion for higher-value leads, or that it improves first-contact resolution by 12-18% — are the ones we spent the most time trying to run down, and came up empty. They appear, largely verbatim, across dozens of lead-routing blog posts and vendor sites, consistently without a named study, a disclosed sample size, a publication, or even a specific company behind the number. One vendor-run analysis of real product usage does exist in this space — Velocify's 2013 study of leads distributed through its own platform — but by the time of writing, the original report is no longer reachable at a live URL to verify its exact figures firsthand, so we're not repeating numbers from it we couldn't confirm directly. That's the honest state of the evidence: real research exists on why an assigned lead does or doesn't get worked (the previous section), but we could not verify a single rigorous, disclosed-methodology study proving any specific routing method beats round-robin by any specific percentage.
The absence of good evidence isn't itself evidence that round-robin is fine, or that skill-based routing is snake oil — it just means the comparison table in the first section is a more honest way to choose a method than any percentage currently circulating for it.
So what should a routing rule set actually account for?
None of the above is a reason to leave the default setting alone forever, or to chase a routing method with no evidence behind it. It's a reason to design the rule around what a routing method actually controls — who owns a lead first — and to handle the rest, which the research above says matters more, separately and deliberately.
- 1Give every lead exactly one clear owner immediately, even if the rule that assigns it is simple. A shared inbox everyone can technically see is not an owner — per Sabnis et al., unclear ownership and untracked follow-up are two of the strongest predictors that a lead never gets worked, ahead of anything about the assignment rule itself.
- 2Move past strict round-robin once a team is bigger than a couple of people or the leads clearly differ in what they need — check real workload before assigning if the tooling allows it (load-based), and route by language, product or deal size when that difference is large enough to matter (skill-based).
- 3Put a clock on the lead, not just a name. A first-response target with a visible escalation path when it's missed does more for whether a lead gets worked than which rotation rule chose the owner — that's the manager-tracking factor the 2013 study found actually moves behavior.
- 4Treat an unassigned or ambiguous lead as a queue to watch, not a rule that failed once and can be forgotten. The failure mode every "black hole" story describes is a lead with no process attached to it, not simply no name attached to it — the fix is a visible backlog, not a smarter rotation formula alone.
- 5Revisit the rule as the team changes. A weighted round-robin's shares and a skill-based map are only accurate on the day someone sets them — a new hire, a rep going on leave, or a shift in what leads a team receives all quietly turn yesterday's fair rule into today's unfair one.
Where oToK fits into this directly: a contact's owner lives on the contact itself, not on a single conversation, so deals, tasks and quotes created afterward inherit it automatically instead of starting ownerless. Every inbound lead — a form, a landing page, a Meta ad, a booking, or the first WhatsApp message of a new conversation — opens an inquiry that intake rules route in order (what was submitted, who the contact is, where it came from, whether it's business hours) to a named agent or to a team by round-robin or by who's least busy right now, with a first-response target that counts down on business hours, warns the owner as it approaches, and escalates when it's missed. An inquiry routing can't resolve waits visibly in the queue instead of disappearing — which, per the research above, is exactly the part a rotation rule alone was never going to fix.
Frequently asked questions
- What does round-robin lead routing actually guarantee?
- Two things, and only two: every lead gets assigned to exactly one person, and turns rotate evenly across the team over time. It has no built-in concept of time zone, business hours, a rep's current workload, language, deal size, or which rep is actually best suited to a given lead — those are the specific gaps that weighted round-robin, load-based routing and skill-based routing each exist to close.
- Does round-robin lead routing improve conversion rates compared to other methods?
- We could not find a rigorous, disclosed-methodology study proving a specific percentage lift for skill-based or performance-based routing over plain round-robin — the 15-30%, 18% and 12-18% figures that circulate for this claim don't trace to any locatable, verifiable source. What's well-established instead is what each method structurally can and can't account for; choose based on that, not on an unverifiable percentage.
- Is it true that most sales leads never get followed up on?
- The specific "70%" figure traces to a real, peer-reviewed source — Sabnis, Chatterjee, Grewal and Lilien's 2013 Journal of Marketing study — as the framing problem the paper opens with, not a number its own 461-rep dataset re-measures. That dataset instead studies what predicts whether an assigned lead gets worked: lead pre-qualification quality, whether a manager tracks follow-up, how many leads a rep is already carrying, and the rep's own experience and performance.
- Where does the "36% of leads are never called" statistic come from?
- A 2012 Forbes piece by Ken Krogue, who was a co-founder of InsideSales.com at the time — the same vendor, working with the same researcher (James Oldroyd), behind the widely-cited 2007 lead-response study and its "100x more likely to qualify" figure. It's a real number from real vendor-affiliated research, not an independent academic study.
- What's the practical difference between round-robin, weighted round-robin, load-based and skill-based routing?
- Round-robin cycles through reps in a fixed order with no other input. Weighted round-robin does the same rotation but gives each rep a pre-set share instead of an equal one. Load-based routing checks how many open leads a rep currently holds before assigning the next one. Skill-based routing matches an attribute of the lead — language, product, deal size — to whichever rep is actually suited to it, before turn order is considered at all. Most teams outgrow a strict rotation the moment leads clearly differ in what they need or reps clearly differ in capacity.
