There is so much conversation about win rates. But in those conversations, there’s a lot of misunderstanding. What we measure, how we measure it, varies tremendously. Some measure win rate from very first contact. Some measure it from the moment of qualification. Some measure it on a $ basis some on a deal basis. Some attribute measure win rates within live deals, some look at past deals.
When we look at win rates, however we measure it, we tend to look at one number. But there are a variety of numbers, and analytics we can get from looking at win rates in a much richer context.
As a result, we don’t leverage the power and insights possible in win rates. It is one of the foundational and most important metrics in sales performance analysis.
This article is a tutorial, possibly a refresher, on win rates. I will start at the basics then extend them. I will leverage examples from real work we have done with our clients.
Buckle up, here we go.
What is a win rate? The win rate represents the proportion of sales opportunities we have competed in, from which we have gotten an customer commitment order. Some of you may be saying, “Well duhhh…” But let me drill down. We have to think, “what’s an opportunity?” Something becomes an opportunity when it is qualified. It is based on the customer saying, “We need to make a change and have a solution in place by this date. The consequences of not doing this by this date are…..”
We do a lot of work with customer trying to create opportunities. We talk about ideas, changes, we start narrowing to a specific area, perhaps possible solutions. We even use the term opportunity to describe these conversations. But they don’t become opportunities until they are qualified. There are huge numbers of activities that we hope to turn into opportunities, that never do. They never become real, they never become qualified, we eventually abandon or defer them. We have to do that work, because it is through this, we create qualified opportunities.
So our win rate is based on qualified opportunities.
Let me repeat myself, so many fundamental errors happen with this very first step. The win rate is based on “real deals.” Until it’s qualified, it’s not a real deal. We still have to work on those to generate real deals, but we don’t count these in our win rate calculations. We only count qualified, real deals in our win rate calculations.
How do we calculate our win rate? At the most basic level the win rate is based on historical performance of qualified opportunities that have gone through the complete buying/selling cycle. The customer has made one of three decisions: They award the business to us, they award the business to our competition, they (or we choose to) abandon the effort (for many reasons). Every qualified opportunity ends in one of those outcomes. All are used in calculating the win rate.
We may look at the trailing 12 months performance. How many deals actually went through that entire cycle? What percentage of those deals did we win? Let’s imagine 100 deals went through the cycle and we won 35 of those deals. Our win rate is 35%. We focus on the number.
But there are two other numbers we can develop that are really interesting. The loss rate. Say for those same 100 deals we lost 30. The loss rate is 30%. And, for those same 100 deals, 35 were abandoned. That shows an abandonment rate of 35%.
Now already, you can see, win rates aren’t enough. As we look at the loss and abandonment rates we can get a lot of insight. In looking to increase our win rates, we may say, “What do we do to decrease the number of abandonments?” Alternatively, we think, what do we do to reduce the losses from 30% to 15%? If all of those are converted to wins, our win rate would go up to 50%.
Some people make an error. They calculate win rates based only on wins and losses. In this example, if we did this, we would have a win rate of 54%. We might fist bump each other with the high win rate. But you can see, we are missing a huge amount, the abandonments!
What is the period of time that we use in looking at win rates? Generally, I find it most useful and accurate to look back at least one sales cycle. If we have an average 6 month sales cycle, I would want to look, minimally, at the trailing 6 months of opportunities that went full cycle. If you look at a shorter period of time, you may actually dramatically over/underestimate the win rate.
In this example, I might want to look at the trailing 12 months data, that gives me a much richer data set to analyze, we’ve been through roughly 2 sales cycles. Some people look back through all history, perhaps saying, “What’s been the win rate based on the trailing 10 years?” Generally, I think that is overkill and may be inaccurate. Things may have changed tremendously in 10 years, so much of the older data may be irrelevant, or misleading (products change, customers change, markets change). So ideally, a minimum of one trailing sales cycle, sometimes up to 2 seem to be pretty good.
What’s happening over time? A very powerful analysis is to look at trends of these numbers over time. We may want to look at an analysis of last quarter. For the deals that closed, what was the win/loss/abandon rate? How does that compare to the previous quarter, and the quarter before, and…. Do we see upward, downward trends? What is the impact of that? What do we need to do to change it?
It’s stunning to me, over the last 10-12 years in some sectors, we have seen YoY declines on win rates. Currently, down to 15-17%, and they look to be going further south. How much business are we losing because of these lowering win rates?
Which win rates do we want to look at(from this point forward, I will refer to win rates, but I am talking about the 3 rates: Win, Loss, Abandoned)? I will keep coming back to this, building the story. It is looking at this question, that we start understanding the richness and potential insights from win rates. But the first level answer is: We want to understand the individual win rates for each seller. We want to understand team win rates, organizational win rates. We may have specialists, we want to understand their individual and team won rates.
Congrats, you have successfully completed win rate 101A. Now we are moving into win rate 101B.
Which win rates do we want to look at: Now let’s look at the next level of refinement. Do we look at $ based win rates and/or Deal based win rates. in the previous example, I used a deal based calculation. “We won 35 out of 100 deals and have a 35% win rate.” But there’s another perspective, the $ based win rates. Let’s say those 100 deals represented $100M. And the 35 deals that we won represented $20M. So our $ based win rate is 20%.
Hmmm, this is interesting. Our $ based win rate is 20%, our deal based win rate is 35%, why the difference. What this shows is the team is good at winning the lower dollar value deals and bad at the big deals. If we just looked at the deal based number, we would never understand this.
Now let’s move into graduate school with win rates with win rate 201a.
Which win rates do we want to look at: (I’ll keep diving into this) I had a client that thought they were fantastic at winning $Million plus deals. They looked at the aggregate $ Value of all the deals, and found their $ Based win rates were 17% (Not great from my perspective, but they weren’t unhappy).
But then I looked at the data differently. I looked only at the qualified deals of $1M or more. (wins, losses, abandons over the trailing 18 months). It turns out that win rate was 9%! They sucked at $1M plus deals. Because they were aggregating all of these with their under $1M deals, they thought they at 17% win rate and doing pretty well. They never saw how really bad they were at big deals.
So we may want to start dis-aggregating the win rates calculations. Win rates for over $1M, win rates for under $1M. And with these we want to look at the deal based and $ based win rates for each category. We may want to divide it even further, for example, under $100K, $100K-1M, over $1M. You can see the level if insight we can get by just looking at the win rates in slightly different ways.
More win rate disaggregation: Hopefully, now you can start getting ideas of different ways to assess performance looking at win rates. Some possibilities:
- Net new account win rates. In acquiring net new customers/logos, how effective are we? How do we improve.
- Enterprise account win rates. In our enterprise accounts, we are looking to develop special relationships, driving much more business across the enterprise.
- Retention/expansion win rates. We know that we want to keep customers for life, are the renewing, are they expanding, are they upgrading. There may be a little overlap with the Enterprise accounts.
- Customer segment win rates. We may sell to SMB, Large, Enterprise, Global accounts. We learn more by looking at segment win rates than an overall win rate.
- Product line win rates. We are accountable for selling the entire product line. I’ve seen so many sellers and organizations make their numbers focusing on one or two product lines, ignoring the rest. Tracking win rates by product line help us understand some areas where we may have huge opportunities for win rate improvement.
- Industry/market segment win rates. Like the product lines, we want to see how we perform in our key industries and markets. Are we underperforming in some, can we improve.
- Regional/geographic win rates. How are we doing across the geography? I have one client that has 60%+ win rates in North America. In ROW, it’s under 35%. Tremendous opportunity for improvement.
- Channel partner win rates. So much of what we do requires channel partners. How are we winning over channel partners, and how are we, working together, performing with the end customer.
- Win rates for our new sales people, win rates for our senior sales people, how do we grow our people’s capability to win. (We might fairly expect our newer sales people to have lower win rates, but we want to train, coach them to improvement.) But if our more experiences sales people have lowering win rates, there’s a problem.
- Program or initiative win rates. Perhaps we are launching a major new initiative. We want to track the win rates of those initiatives.
- ….and more…..
You can see the possibilities are numerous. Not all of these will be useful to every organization, and there may be many others that I’ve not mentioned. Again, traditionally we have treated win rates as a single monolithic number that applies to all parts of our business. But in reality we may have widely varying win rates as we drill down.
As we are looking at how/where we might focus our performance improvement efforts, drilling down on our win rates can help us better focus on performance improvement areas. Let’s look at an example. All of these are Deal based metrics. Let’s say across the organization we have a 50% win rate–on 100 deals. Not bad!
But when we drill down, we discover for the 50 deals in product line A, we had a 75% win rate. For the 50 deals in product line B, we had a 25% win rate. All of a sudden we have a huge “Aha” moment. We are really underperforming in product line B. What can we do to raise the win rate? Look at the leverage we have just by focusing on that. Product line A is good, but we can drive a lot more improvement, just by focusing on product line B.
But if we only look at the single win rate number, we are unlikely to have the deep understanding of actual performance and where we might focus performance improvement efforts.
Moving on to win rate 201B.
The perils of average: When we look at win rates at a team or organizational level, we create an average win rate across the organization. But sometimes this average may be very misleading. Let’s look at an example:
- Our team of 10 sellers competed for 10 deals each, 100 total and won 50 total. So we have a 50% win rate! Fist bumps all around.
- 3 of your sellers each won 10 deals. Those sellers each had 100% win rate (OK, this is just an example).
- 7 of your sellers won roughly 3 deals each for a total of about 20. Those sellers had a 30% win rate.
You can see the problem. When we look at the average across the organization, we see a 50% win rate, completely missing the fact that 70% of our sellers are underperforming.
There are a couple ways we might start understanding this. We might look at the average win rate across the organization, then compare individual win rates with the average. Those above the average, fist bumps all around. Those below the average, how do we get them to improve. And guess what, when we get them to improve, the overall average win rate also improves.
Another way of looking at this from an organizational level is to look at both average and median win rates. If the median win rate is significantly different from the average win rate there is a performance issue-rather a performance improvement opportunity.
Errors we make in thinking about win rates, here is you PhD level course.
The win rate has nothing to do with our current pipelines! “Wait a minute Dave, we need the win rate to understand pipeline health!” Absolutely, once we have determined our win rates, we leverage them to help us understand pipeline health. But we don’t calculate our win rates in our current active pipelines! It’s amazing the number of organizations trying to do this. But think a moment. We don’t know whether something has been won, lost, abandoned until it has gone full cycle, until a customer has made a decision! If you are trying to determine your win rate based on your current open pipeline, stop! Use the win rate to help understand the health of your qualified pipeline.
The win rate is only on qualified deals. Yeah, I’m repeating myself. But I see so many people start counting a deal as real based on a customer opening an email (I’m not kidding). I don’t want to diminish the importance of prospecting and all the other lead/demand/awareness/social programs we may be conducting. We need to do those to create new qualified opportunities. But it until it is qualified, it is a possibility. Once qualified, it’s a real deal that we do everything possible to win. There are other metrics and conversion measures important to understand and track, but they aren’t win rates.
There is no such thing as a marketing win rate. In recent years, I’ve seen some in marketing trying to create a win rate metric. I’m not clear that I understand the logic. Much of it seems to be focused on measuring activity and conversions that lead to developing a qualified opportunity. It’s important to understand these activities and the results they produce, but these are not win rates. Before I alienate all my marketing friends, I’m not saying marketing does not contribute to our win rates! Marketing is very important to our win rates. Marketing contributes to the full buying cycle, now only awareness/lead/demand gen, but marketing contributes in important ways to helping us win qualified deals. So win rates are important for marketing to track. And with sales we need to figure out how marketing might help to improving our win rates.
Rev-ops, where are you? Rev-ops can provide a huge amount of leadership in drilling down into these win rates, identifying areas of performance improvement. Rev-ops can help managers better understand the performance of each of their people, by equipping managers with the few win rate numbers critical to them and their teams.
Enough already! OK, I’ll stop here. There are all sorts of other ways we can look at win rates to better understand performance. Hopefully, this provokes your curiosity.
The majority of organizations don’t understand the power and insight provided by this single metric. Too many get the basic measurement wrong, or are measuring the wrong things. Too many rely on the single number, “What % of deals do we win,” and in focusing only on that, they miss so much more.
But that number is just the starting point!
Rev-ops leaders need to drill deeply into this, identifying helping managers identify and address performance gaps. Every front line manager needs to understand these numbers for each person on their teams.
Afterword: Normally, I would provide a link to an AI generated discussion of this post. I attempted to generate one and reached the limits of this type of AI tool. This post leans to an analytic, problem solving process. These tools simply can’t handle this type of discussion of, “What are the equations, why are they important, how do we make the calculations?”
This discussion was so bad, it didn’t even have comic value. So I’m not publishing it.
Leave a Reply