So far, I’ve directly worked with over twenty CMOs and CROs from companies ranging from 50 to 1,000 employees, right from seed stage to Series D. One thing they all kept saying to me was, “We don’t know if our metrics are good because there’s no benchmark data in B2B SaaS.” These companies usually rely on bits of data from Gartner reports, competitor analysis from consultancy firms, or the experiences of employees who worked at other companies. There’s never been a real go-to source, until now.
I’ve collaborated with HockeyStack this month to create the benchmark report I’ve always wanted, a go-to guide for all the B2B leaders out there.
This report has two parts. The first is all about marketing, diving into how budgets are split across channels, what the metrics within these platforms look like, and the conversion rates they’re seeing.
The second part focuses on revenue, breaking down the best and worst months for MQLs, opportunities, closed-won deals, and revenue. We also take a look at sales cycles and ACVs.
Creating this report was a great experience, and we hope it becomes a 2024 benchmark for all B2B SaaS companies. Hope you find it insightful and enjoyable too!
Last week, we published the Self Reported Attribution Report 2024, examining the relationship between last-touch sources and self-reported attribution. From over 8,000 responses, we found out that nearly half of the buyers still use search engines for product research.
Reflecting this trend, we’ve observed that in 2023, a significant portion of ad budgets went to Google, accounting for 47.72% of total marketing spend. Notably, Linkedin was a close second, getting 45.84% of the spend.
This year’s budget allocation marks a shift from previous patterns. Traditionally, paid search received a larger share, but in 2023, the combined spend on Linkedin and Facebook, paid social, exceeded 50% of the total budget.
Additionally, we’re also seeing that Bing started to get more budget after Q1-2023. This uptick indicates that B2B companies started to advertise on Bing after the launch of Chatgpt for Bing.
Bing’s budget share grew from 1% in Q1 to an average of 1.8% in subsequent quarters, totaling 1.6% for the year. This signifies an 80% increase in Bing’s budget following the Chatgpt launch.
While some clients experimented with other ad platforms like Youtube, Reddit, and Capterra, we’ve excluded these platforms from this analysis since the combined spend was below 1%.
Throughout 2023, Bing had the best CTR with over 3%, followed by Google at 2.2%, Facebook at 0.96%, and Linkedin at 0.88%.
Does this mean that Linkedin is not performing well? Our hypothesis is different.
Considering that Bing’s total budget allocation was just above 1.5%, it’s likely that companies are using Bing for flagship campaigns, such as brand or high intent keyword campaigns, which could be why we’re seeing a higher CTR for Bing compared to Google.
The difference in CTRs between paid search and paid social is to be expected. Paid search is more about capturing existing demand, so naturally, the intent is higher.
But why is Facebook’s CTR higher than Linkedin’s? This might seem odd at first, especially given Linkedin’s B2B targeting capabilities. However, considering the budget allocation, with Facebook receiving less than 5% of the budget, it indicates that companies are primarily using Facebook for retargeting campaigns. They aren’t leveraging Facebook as a major channel for demand creation or brand awareness. Hence, the higher CTR on Facebook can be attributed to an audience that is already familiar with these companies, rather than a completely new audience, which explains the higher CTR.
In this part, we can see the actual power of Linkedin targeting. Linkedin has, by far, the best MQL to Opportunity conversion rate compared to other channels. We’re seeing that 36.2% of the Linkedin MQLs became an opportunity compared to 22.7% on Facebook, 22.3% on Bing, and 21.7% on Google.
To calculate MQL, we have used a position-based attribution model because the position-based model can accurately reflect the impact of various touchpoints in the customer journey. Unlike first touch or last touch models, which credit only the initial or final interaction, the position-based model acknowledges the importance of both the initial engagement and the decisive action leading to conversion. It also avoids the oversimplification of uniform or linear models, which may distribute credit too evenly, potentially overlooking key interactions. This model is particularly effective in understanding the high conversion rates from platforms like Linkedin, where multiple touchpoints play critical roles in driving MQLs to become opportunities.
These conversion rates justify why Linkedin is more expensive. It’s also interesting to see the conversion rate of Facebook MQLs, as Facebook is mainly used for retargeting. Yet, the retargeting MQLs don’t have high conversion rates – in fact, their rate is just slightly higher than the paid search conversion rates.
Although the definition of MQLs might vary from company to company, we have used high-intent form submissions as our MQL definition. High-intent form submissions include forms like demo requests, pricing, talk to an expert, et cetera.
While some of the figures were expected, some were surprisingly different. Our initial expectation was for December to be the worst month for MQLs, along with July and August, when people usually go on holiday. However, the data showed differently.
December is still one of the worst months, but it ranks third after January and February, with January being the worst month in 2023, according to our data.
Here, it’s important to note that the spend figures didn’t change much amongst these months. For instance, the total ad spend difference between Q1 and Q2 is less than 15%, and the difference across all four quarters is not more than 20%. The most significant difference was in Q4, as our clients decreased their ad spend by an average of 35% in December – which Linkedin doesn’t recommend, but we strongly do.
One might suggest that in January, buyers are returning from holidays, spending their first couple of weeks catching up and launching plans for the new year. This could explain the significant difference between January and other months. From January to February, there’s a 22% increase in MQLs. However, compared to other months, this increase in February still isn’t enough.
February is 3 days shorter than January. Considering there were also 8 weekend days, February was effectively a 20-day month. Could this be the reason? Not quite.
Although April was a 30-day month, it actually had 5 weekends hence 10 weekend days, making it have 20 working days just like February. However, we saw an 84% increase in MQLs in April. Therefore, the number of days doesn’t seem to be the primary factor.
The most logical reason for the lower numbers in January and February is that companies are still figuring out their 2024 budgets, plans, and KPIs. Ideally, companies would have this done by the end of the previous year, but it seems like it doesn’t happen that way.
We can validate this assumption by looking at the month-on-month data, this data suggests that the worst quarter for generating MQLs is Q1, with only 17.5% of the total MQLs generated; then we see a massive uptick in Q2, most likely when companies have their plans and budgets ready and are more inclined to spend.
Almost 2x more MQLs were generated in Q2 compared to Q1.
Precisely, 31% of the total MQLs generated in 2023 were in Q2.
May appears to be the best month for generating MQLs, followed by July and June.
Despite common assumptions, July is not a dead month for marketing; it ranks as the second-best month for generating MQLs, but we see the impact of summer in August.
Although numbers improve in October, the first month of Q4, they are still significantly behind the numbers in April, the first month of Q2.
We would normally expect to see higher MQL numbers in the first month of quarters; comparing the first months, the best is July (Q3), then April (Q2), followed by October (Q4), and January (Q1).
So, what can be done with this data? Our recommendation is to invest heavily in brand/demand creation activities in Q1. During this period, as buyers are sorting out their internal plans and KPIs, maximizing your reach is key. This way, when Q2 comes and buyers have their budgets ready, they’ll already be familiar with your brand.
On the pipeline side, we’re seeing a completely different picture. Although Q1 was the worst month for the number of MQLs, it turns out to be the best for the pipeline, where 30% of the total pipeline was generated.
What could be the reason for this? Considering the sales cycles and cohort data is important. Let’s look at this in conjunction with the state of MQLs.
We are seeing shorter MQL:Opportunity cycles in Q1 and Q2, meaning MQLs generated in Q2 become opportunities within the next 30 days, within the same quarter.
Although Q3 seemed good for generating MQLs, the impact is different at the pipeline level. 26% of the total MQLs were generated in Q3, but only 18% of the total pipeline was generated in this time period.
Compared to Q2, we’re seeing that MQL:Opportunity duration gets around 1.5x longer.
So, what does this mean?
– MQLs generated in Q2 become opportunities in Q2.
– MQLs generated in Q3 become opportunities in Q4.
– MQLs generated in Q4 become opportunities in Q1.
– MQLs generated in Q1 also become opportunities in Q1.
Therefore, the best pipeline happens in Q1.
On a monthly view, November is the best for pipeline generation. This makes sense as sales teams push to close deals before year-end. We see a similar pattern in March, aligning with the end of the financial year for some companies: sales teams qualify MQLs massively in February and early March, so that AEs can sell in March.
Looking at the worst months for pipeline creation, December is the worst one, which is expected. It’s a month for closing, not qualifying.
Then comes September and July. Seeing July here was a bit expected due to the summer period. On the MQL level, we see the summer impact in August; on the pipeline level, it’s in July.
Here’s a thought. July is good in terms of MQL generation, but not for pipeline generation. August is good in terms of pipeline generation, but not for MQL generation. If companies qualify MQLs by seniority, does this mean decision-makers take holidays in July, and non-decision makers in August? We’re going to research this!
One surprise was seeing September as the second worst month for pipeline creation. Apparently, the ‘back to school’ concept doesn’t apply to B2B as it does in B2C.
For us, this part is the most interesting. We’ll examine this in two ways: the number of closed-won deals, and the total revenue.
In terms of the number of closed-won deals, December is the best month as expected, and Q4 is the best quarter. Almost 30% of the total deals were closed in Q4, with more than 10% in December alone.
Q4 is followed by Q2 and Q1, with Q3 being the worst for closing deals. The closed-won data aligns with our expectations: December leads, followed by March (end of Q1), June (end of Q2), and September (end of Q3).
Despite Q3 being the worst quarter for closing deals, September has a solid record – almost 10% of the total closed-won deals happened in September. We might not see good MQL or pipeline numbers in September, but deals are definitely getting closed.
Here’s where it gets interesting – looking at the revenue data, everything changes!
While Q4 had the most deals closed, it falls behind Q1 and Q2 in total revenue.
Q1 is the best quarter for ARR, generating 29% of the total revenue, followed by Q2 with 27%.
In Q4, although 28% of the deals were closed, only 25% of the revenue was generated.
This means companies heavily discount their prices in Q4. We’ll discuss this in the last part, the State of ACV.
Ironically, ACVs are higher in the first two quarters with shorter sales cycles, and lower in the last two quarters with longer sales cycles.
On the monthly level, the best month for revenue is March, followed by December, June, and May.
March is a key driver for Q1’s revenue, being 1.5x higher than January and February combined.
July is the worst month for revenue, followed by October, April, and January. This doesn’t surprise us considering that all these four months are the first months of quarters.
Every B2B company has their own benchmarks, some still using Gartner research data from ten years ago, others relying on information from their peers. We’re putting an end to this now.
From our analysis of dozens of customers, the average MQL to closed-won rate in 2023 was 5.94%, and the average opportunity to closed-won rate was 22.85%.
For MQLs, the highest win rates were seen in Q1, particularly in March. In March, 10% – or one out of every ten MQLs – became closed-won deals, while in November, this rate dropped to 3.85%.
On the opportunity front, the highest win rates were in May, March, and September.
May led with an opportunity to closed-won rate of over 35%, followed closely by March at 34.96%, and September at 30.34%. The lowest win rate was in December, at 11.64%, but it’s worth noting that opportunities generated in December are likely to convert in Q1-2024. Hence, the actual closed-won rate of December’s opportunities will only be clear by the end of Q1.
This data indicates that by maximizing the number of MQLs generated in early March, these MQLs can potentially convert into revenue within the same month.
TLDR: The average sales cycle in B2B SaaS companies in 2023 was 69 days.
In the previous part, we covered the average time from MQL to opportunity. Whilst MQLs from Q1 and Q2 usually become opportunities in the same quarter, those from Q3 and Q4 typically become opportunities in the next quarter. Now, let’s shift gears to the opportunity to closed-won cycles.
December turns out to be the slowest month when it comes to the opportunity:closed-won cycles. If an opportunity is created in December, it takes an average of 108 days to close it, so it basically becomes next year’s deal (As of January 20th, 2024 – this might change depending on the number of deals closed in the next 10 days).
January seems to be the second slowest month, but considering the approval of the yearly budgets and KPIs, this makes sense. Opportunities created in January end up closed won deals in 97 days.
In general, May seems to be the best month for everything; we’re seeing the best number of MQLs, best revenue, and best win rates in May. On the sales cycle side, it’s also the best month where opportunities created in May become closed won in 51 days.
On the quarterly level, we’re seeing that Q2 is the fastest quarter to close deals. The data suggests that if you maximize your pipeline in early Q2, you can maximize your revenue by the end of Q2.
TLDR: The average ACV of B2B SaaS companies in 2023 was $20.8K.
Do ACVs really change that much from quarter to quarter? Yes. Big time.
It looks like deals in Q1 and Q2 are bringing the highest ACV, while deals in Q4 are bringing the lowest. This pretty much confirms that companies are slashing prices big time towards year’s end. Just like we pointed out earlier in the ARR section, Q4 is the best quarter in terms of the number of closed won deals but not in terms of the total added revenue. We can understand this better by looking at the month-on-month change in average ACVs.
While in January, the average ACV was $38K; in December it was almost half of it, $22K. What’s more striking is how average ACVs steadily drop after Q2.
We mentioned that July was the second best month in terms of the opportunity:closed won sales cycle, only 54 days, this data makes more sense when we look at the average ACV in July, which is only $10.9K. By far the lowest month in terms of ACV.
Here’s another twist: we thought higher ACVs would mean longer sales cycles, but that’s not always what we’re seeing. Turns out, there’s not a significant link between higher ACVs and longer sales cycles (keeping it simple here, we’re not diving into the exceptions).
I think this report can help with two important things: understanding the current landscape better, and managing expectations.
As I said in the beginning, one thing I kept hearing from all the CMOs and CROs I worked for was that there was no benchmark, and they struggle with showing their team or their managers if their numbers are good.
We’ve analyzed tens of thousands of data points from dozens of B2B SaaS companies to reveal the benchmark data for the most important metrics I kept hearing about for years. I hope this report will be a go-to guide for all the B2B leaders out there.
The second important thing is managing expectations, especially when it comes to dealing with finance teams. Most of the time, finance teams consider the customer journey as a linear line where if you put the same number every month, you’ll get the same result. Reality is often quite different from this; even if every variable stays the same, the results can drastically change due to external factors. This report is solid proof of these changes; for example, if your ACV is getting lower or the sales cycle is getting longer, this might not necessarily be about unqualified opportunities.
Understanding what’s been happening in the market is as important as understanding the actual impact of your marketing spend. HockeyStack has already been helping its customers to measure their ROAS without any blindspots, and now HockeyStack customers will receive these benchmark data every month to better understand the market landscape and use their budget more effectively.
.svg)
.png)
.png)
