What The Challenger Sale Research Actually Found (And What Gets Misquoted)
Four different percentage tables, one book, and a lot of confident blog posts quoting the wrong one.
The Challenger Sale is built on CEB research covering more than 6,000 B2B sales reps, with the first data collected in 2009. The headline numbers from that study get reported wrong almost everywhere, including on the website of the company that licenses the methodology today. Here is what the primary sources actually say.
We started writing this as a straight summary and gave up about an hour in. Every blog post we checked quoted a different set of percentages, and several quoted two numbers side by side that came from two different tables. The confusion has a traceable source, and it starts at the top of the chain rather than the bottom.
What research is The Challenger Sale based on?
The book draws on research from CEB's Sales Executive Council in Arlington, Virginia. First-line sales managers rated more than 6,000 B2B reps against 44 attributes spanning attitudes, skills and behaviors, activities and knowledge. Participating companies identified their top 20% by performance against goal. CEB ran a factor analysis on the 44 variables and derived five profiles.
Two details about that study matter more than most people realize.
The data is from 2009. Not 2011, when the book came out. 2009, the trough of the recession. Noel Capon's review of the research reports that the five-group model was built from an initial sample of 700 reps, and that the database grew past 6,000 later as the economy recovered.
It didn't start as a study of salespeople. It began as a customer loyalty study. The finding that kicked everything off was that 53% of B2B customer loyalty is driven by the sales experience itself, ahead of brand, product, service and price. That number is the actual foundation of the book. The five profiles came afterwards, as an attempt to work out which reps deliver that experience.
Who wrote The Challenger Sale, and when was it published?
Matthew Dixon and Brent Adamson wrote The Challenger Sale: Taking Control of the Customer Conversation, published by Portfolio/Penguin on November 10, 2011. 240 pages. ISBN 9781591844358. Neil Rackham, author of SPIN Selling, wrote the foreword. Both authors were at CEB, where the research ran out of the Sales Executive Council.
Two things to watch for if you're citing it.
A lot of blogs say October 2011. The publisher says November 10.
And Challenger Inc's own site states that the authors "wrote the original Challenger text, Winning The Challenger Sale, in 2010." That's wrong on the title and the year. Winning the Challenger Sale is their podcast.
Where the authors ended up: Dixon is a founding partner at DCM Insights, and has since written The Effortless Experience (2013), The Challenger Customer (2015) and The JOLT Effect (2022). Adamson was at CEB and then Gartner from roughly 2003 to 2022.
Which HBR article should you cite for the Challenger numbers?
Cite "The End of Solution Sales" by Brent Adamson, Matthew Dixon and Nicholas Toman, Harvard Business Review, July/August 2012, reprint R1207C. That's the article carrying the profile percentages and the complex-sales figure. Two other HBR pieces exist: "Selling Is Not About Relationships" (September 30, 2011) and "Dismantling the Sales Machine" (November 2013).
The line everyone paraphrases: HBR reports Challengers "accounted for nearly 40% of the high performers," rising to 54% in complex, insight-driven environments.
Read that sentence carefully, because it contains both of the numbers that get scrambled downstream, and it labels them correctly. Nearly 40% is all sales. 54% is complex sales only. Everything below is what happens when that distinction gets dropped.
What are the actual Challenger Sale percentages?
Four different percentage tables come out of the CEB research, and each one answers a different question. Share of the total sales population. Share of high performers across all sales. Share of core performers. Share of high performers in complex sales specifically. Most posts quote one table and label it as another.
Table A: share of the overall sales population
| Profile | % of all reps |
|---|---|
| Challenger | 27% |
| Hard Worker | 21% |
| Relationship Builder | 21% |
| Lone Wolf | 18% |
| Reactive Problem Solver | 14% |
Sums to 101 from rounding.
Table B: share of high performers, all sales
| Profile | % of high performers |
|---|---|
| Challenger | 39% |
| Lone Wolf | 25% |
| Hard Worker | 17% |
| Reactive Problem Solver | 12% |
| Relationship Builder | 7% |
Table C: share of core performers
Relationship Builder 26%, Challenger 23%, Hard Worker 22%, Lone Wolf 15%, Reactive Problem Solver 14%. This is the weakest sourced of the four. Use it lightly.
Table D: share of high performers in complex sales
| Profile | % |
|---|---|
| Challenger | 54% |
| Lone Wolf | 25% |
| Hard Worker | 10% |
| Reactive Problem Solver | 7% |
| Relationship Builder | 4% |
Four tables. Twenty numbers. Most articles present five of them and call it "the Challenger research."
Why do the Challenger percentages disagree between sources?
Because the same profile has four different numbers attached to it and almost nobody names which table they're pulling from. The most common error pairs the complex-sales Challenger figure of 54% with the all-sales Relationship Builder figure of 7%, as though they came from one table. In complex sales the Relationship Builder figure is 4%.
Here's the part worth knowing. That specific error appears on Challenger Inc's own current website, presenting the complex-sales 54% next to the all-sales 7%. Challenger Inc is the company that licenses and sells the methodology. When the licensee garbles its own primary data, every blog that cites the licensee inherits the mistake, and every blog that cites those blogs inherits it again.
We don't think anyone is being dishonest. We think 54 versus 7 is a better headline than 54 versus 4 or 39 versus 7, and the two most quotable numbers in the dataset happen to live on different tables. That's how the error survives. It reads better wrong.
If you want the profile numbers laid out cleanly, we did that in the five types of sales reps and what the data says about each.
Is the 57% statistic from The Challenger Sale?
No. The claim that B2B buyers are 57% through the purchase process before they contact a supplier does not appear in the 2011 book. It comes from CEB's Marketing Leadership Council report "The Digital Evolution in B2B Marketing," published in 2012. The 2012 HBR article cites a different figure, nearly 60%, from a separate study of 1,400 customers.
This one is probably the most-repeated sales statistic of the last fifteen years, and it's routinely credited to The Challenger Sale, sometimes to Gartner, occasionally to both in the same paragraph. It's a real number from real CEB research. It's just from a different piece of research, by a different council, published a year after the book.
If you want a current figure on the same idea, use Adamson's January/February 2022 HBR piece, which reports that B2B customers spend only 17% of the purchase process talking to potential suppliers.
How many companies were in the Challenger study?
Harvard Business Review says 83 companies. Most sales blogs say 90. A September 2011 HBR blog post is reported to say nearly 100, but it sits behind a paywall and we could not verify it. Cite HBR's 83. It is the only company count with a source attached. If you round it off in a deck, write "roughly 90" and say that 90 is the number the blogs settled on rather than a figure any source published.
Minor point, and we're including it because "90 companies" shows up in hundreds of posts with no source attached, which tells you something about how the rest of the numbers got where they are.
What are the criticisms of The Challenger Sale research?
The most detailed critique is Noel Capon's, published in Velocity, the journal of the Strategic Account Management Association, in January 2015. Capon is R.C. Kopf Professor of International Marketing at Columbia Business School. His summary judgment: the reported evidence does "not allow us to support the authors' assertions."
His specific charges, in full, because almost nobody links to them:
- The empirical study occupies 15 of 214 pages, about 7% of the book. The remaining 93% is prescription built on top of it.
- Undisclosed variables. Only around 25 of the 44 questionnaire items are named anywhere. No scaling described, no means, no standard deviations.
- The wrong statistical technique. Factor analysis is designed to reduce variables. Cluster analysis is the conventional method for forming groups, and it comes with procedures for deciding how many groups there should be. The book never explains how it arrived at five.
- No results reported. No appendices, no factor loadings, no holdout test.
- The attribute ratings are manager judgment. All 44 attributes were scored by first-line managers describing their own reps. CEB also collected actual sales-versus-budget data and did not use it in the published analysis.
- The sample expansion is undocumented. The model came from 700 reps during a recession. The database later grew past 6,000. No evidence is offered that the model still held. Capon's sharpest question: did the Challenger advantage weaken as the economy recovered? The book splits results by high and low complexity, never by recession and recovery.
- "Complexity" is never defined or measured, despite being the moderator that the entire headline finding depends on.
- The Lone Wolf gets buried. At 25% of high performers, the Lone Wolf is the second strongest profile in the data, and the book waves it off as impossible to replicate at scale. Capon: "This argument seems like a cop out."
Capon does give the book credit for putting a new dimension into the conversation. He's not dismissing it. He's saying the evidence presented doesn't carry the weight the book puts on it.
There's also a peer-reviewed treatment: Rapp, Bachrach, Panagopoulos and Ogilvie, "Salespeople as knowledge brokers: a review and critique of the challenger sales model," Journal of Personal Selling & Sales Management, 34(4), 2014, pages 245 to 259. It won the 2014 Marvin Jolson Award. It's paywalled and we have not read the full text, so we're citing it rather than characterizing what's in it.
Has the Challenger research been independently replicated?
No. Every study cited as confirming the Challenger findings traces back to CEB, Gartner or Challenger Inc. Challenger Inc references a 2020 restudy and a 2019 study of roughly 600 sellers, and neither publishes a methodology anyone outside the company can check. The finding has never been replicated by a party that doesn't sell Challenger training.
Worth being precise about what that does and doesn't mean. It doesn't mean the finding is wrong. Large-scale sales research is expensive and awkward to run, and vendor research is most of what exists in this field. Plenty of true things have only ever been measured once.
It does mean that if you're about to restructure a sales team around it, the evidence base is one company's internal study, from a recession year, graded by the reps' own managers, with the underlying results unpublished.
What happened to CEB?
Gartner announced its acquisition of CEB on January 5, 2017 and closed it on April 6, 2017. Terms were $54.00 in cash plus 0.2284 Gartner shares per CEB share, valuing the deal at approximately $2.6 billion in transaction value, or approximately $3.3 billion in enterprise value including assumed net debt. Both figures are correct depending on the measure.
Gartner then divested the training business in 2018, and Marlin Equity Partners stood Challenger up as an independent company. In September 2024, Richardson Sales Performance, a Truelink Capital portfolio company, acquired Challenger. It now brands as "Challenger, powered by Richardson," and the CEO is John Elsey.
Neither Dixon nor Adamson runs it.
One more inherited error: Challenger Inc's own timeline dates the Gartner acquisition to 2016. Gartner announced in January 2017 and closed in April 2017. That's the second dating mistake on the licensee's site, alongside the book title and year. We mention it because it's a useful reminder that "the official source" and "the accurate source" are not automatically the same page.
The curriculum drifted too, which is worth knowing if you're citing "the Challenger skills." Today's Challenger Inc materials describe four core skills, adding constructive tension as its own named item. The 2011 book runs constructive tension through the other three as connective tissue rather than listing it separately. Both are defensible descriptions of different vintages, and we untangled the concept itself in teach, tailor and take control.
What the study measured, and what it didn't
This is the part we care about most, and it's the part that changes how you should use the framework.
What the study measured: how first-line sales managers described their reps across 44 attributes in 2009, how those descriptions clustered, and how the clusters lined up with the top 20% that each participating company had identified by performance against goal.
What the study did not measure:
- Causation. The design is cross-sectional. A rep who is winning gets described by their manager as confident, challenging and comfortable with money. The arrow could easily run backwards, and the study can't tell you which way it points.
- Whether reps can change profile. The book's whole prescription is to train reps toward Challenger behavior. The research never tested whether training moves a rep between profiles, or whether moving them changes results.
- Complexity. The moderator carrying the headline finding was never defined or measured.
- Whether it survived the recovery. 2009 data, expanded sample, no re-test reported.
- Revenue, as the input to the profiles. CEB had sales-versus-budget data on these reps and left it out of the published analysis. The 44 attribute ratings that produced the five groups came from managers describing their own people, so the descriptive half of the study rests on judgment while the harder measure sat unused. That's Capon's objection, and it's the same trap that shows up across the sales-performance literature: change the measure and the answer often flips.
Our read, plainly: the five profiles are a well-observed description of what strong reps looked like to their managers in complex B2B selling in 2009. That's genuinely useful. It's a hypothesis about behavior rather than a measured law, and the gap between those two things matters most at exactly the moment people usually forget it, which is when they're making a hiring decision.
Use the profiles as coaching language. We do. Don't use them as a screening filter, and don't quote 54% at your board without saying which table it came from.
The behaviors hold up better than the profiles, for what it's worth, and they're the part you can actually train. We wrote the implementation version separately in building a Challenger sales culture.
We took the same evidence-first approach to the broader question of how salespeople get categorized, including the frameworks that have even less behind them than this one.
Where does WingRep fit?
We're an AI sales performance team, so we have an obvious interest in the "reps need better preparation" conclusion. Worth saying out loud before we make it.
What we took from the research is narrower than what most vendors take. The part with the strongest support is the customer loyalty finding: 53% of B2B loyalty comes from the sales experience itself. That's about what happens in the conversation, which is where WingRep works. It prepares reps before a call with the account and stakeholder context they'd otherwise skim in the parking lot, assists during the call, and handles the follow-up after.
What we don't do is score reps into profiles. The measurement behind those profiles is manager opinion from 2009, and we'd rather give a manager specific behavior from an actual call than a label.
See how WingRep works on a live call
Common questions
What research is The Challenger Sale based on?
CEB's Sales Executive Council studied more than 6,000 B2B sales reps, with initial data collected in 2009. Managers rated reps against 44 attributes, and CEB ran a factor analysis to derive five profiles. It grew out of an earlier customer loyalty study that found 53% of B2B loyalty comes from the sales experience.
What percentage of top performers are Challengers?
39% of high performers across all sales, and 54% of high performers in complex, insight-driven sales, per Harvard Business Review, July/August 2012. Those are two different measurements from two different tables and they should never be quoted as one figure.
Is the 57% buyer journey statistic from The Challenger Sale?
No. It comes from CEB's Marketing Leadership Council report "The Digital Evolution in B2B Marketing," published in 2012, a year after the book. The 2012 HBR article gives a separate figure of nearly 60% from a study of 1,400 customers.
Is The Challenger Sale research still valid?
It has never been independently replicated by anyone outside CEB, Gartner or Challenger Inc. Noel Capon of Columbia Business School published a detailed methodological critique in 2015. The findings are worth using as a hypothesis about behavior in complex sales, not as settled evidence.
Who owns Challenger now?
Richardson Sales Performance, a Truelink Capital portfolio company, acquired Challenger in September 2024. It brands as "Challenger, powered by Richardson," with John Elsey as CEO. Gartner acquired CEB in 2017 and divested the training business in 2018.
How many companies took part in the Challenger study?
Harvard Business Review reports 83 companies. Most sales blogs say 90 without citing a source. The verified figure to use is more than 6,000 reps, with 83 companies if you need the company count.
Sources
- CEB press release, November 2011: prnewswire.com
- Adamson, Dixon & Toman, "The End of Solution Sales," Harvard Business Review, July/August 2012: hbr.org
- Dixon & Adamson, "Selling Is Not About Relationships," HBR, September 2011: hbr.org
- Adamson, Dixon & Toman, "Dismantling the Sales Machine," HBR, November 2013: hbr.org
- Adamson, "Sensemaking for Sales," HBR, January/February 2022: hbr.org
- Capon, "Revisiting The Challenger Sale: 'Breakthrough' Built on a Flimsy Foundation," Velocity 17(3), January 2015: business.columbia.edu
- Rapp, Bachrach, Panagopoulos & Ogilvie, Journal of Personal Selling & Sales Management 34(4), 2014: tandfonline.com
- Gartner completes acquisition of CEB, April 2017: businesswire.com
- Richardson Sales Performance acquires Challenger, September 2024: prnewswire.com
- Challenger Inc, selling profiles: challengerinc.com
- Penguin Random House, book page: penguinrandomhouse.com
WingRep is the AI sales performance team that puts this into practice on real calls: it preps the rep beforehand, nudges them live when the hard question lands, and writes the CRM update and follow-up afterwards.
