The Final Verdict: Demand Gen vs. Lead Gen

I can’t believe why we haven’t written about this before, but better late than never. When I joined Cognism back in 2022, it was solely because of demand generation—Cognism was doing something incredible at scale. I was watching, reading, and listening to Cognism and Refine Labs everywhere, and was trying to implement their tactics in my company; however, it wasn’t easy to break that leadership resistance. Nobody wanted to take any risks and change their lead gen engines. When I saw a job opening, I texted Liam without a second thought. One thing I really wanted to learn was if the conversion metrics were really that different from lead gen. Things went well; fast forward, I joined Cognism and started playing with Salesforce. And yes, everything was true—the conversion metrics were amazing, and the pipeline and revenue ROI were indeed way healthier than the lead gen metrics.

When it comes to sharing demand generation tactics and data, Cognism has been pretty open about it, as HockeyStack and Metadata (back in 2022). Similarly, Refine Labs shared lots of reports and success stories about their clients. Although they didn’t intend to, Metadata’s report from last year inadvertently highlighted how unprofitable lead gen was. However, I thought we could take things even further- like maximizing the sample size and publishing one final verdict on demand generation vs. lead generation. There hasn't been research at this scale before, and I'm so happy that it's coming from HockeyStack Labs.

Methodology: 

Demand Gen MQLs: Direct form submissions such as demo, pricing page, and contact us. 

Lead Gen MQLs: Any form submission that is defined as an MQL in companies' CRMs—which includes form submissions such as demo, pricing page, and contact us, as well as ebook downloads, webinar registrations, etc.

SQL: Pipeline created. Every company has different definitions, but we unified this on the backend and used the SQL definition for when the pipeline is actually created.

Pipeline ACV: Total contract value, deal size, deal value by the number of SQLs. 

Revenue ACV: Total new business revenue added divided by the number of closed-won deals (also known as Average Contract Value, Deal Size, Selling Price, Deal Value)

CW: Closed won deals, net new business, new revenue. 

Sample Description:

Sample Size: 87 B2B SaaS companies with SLG motions.

- 48 Demand Gen

- 39 Lead Gen 

Data Range: From October 1st, 2023 to March 31, 2024. 

Attribution Model: Position-based because it offers a balanced view on the importance of each touchpoint in a customer's journey towards a conversion. This method assigns more credit to the first and last interactions and distributes the remaining 20% among the middle interactions. 

- Using HockeyStack at least since September 1st, 2023. 

- From $5M ARR to $2B ARR; average ACV from $10K to $150K. 

- 75% NAM, 20% UK, 5% Europe 

More than:

- Quarter million MQLs, to be exact, 270K

- $103.9M ad spend

- Half a billion pipeline, to be exact $559M

- $72M revenue 

Caveat: This doesn't mean that the ad spend in these two quarters resulted in a 4.79x pipeline ROI and a barely positive revenue ROI. It's important to consider the length of sales cycles. The ad spend in Q4 not only impacts the pipeline for Q4-23 but also impacts the pipeline in Q1-24. Similarly, the ad spend in Q1-24 will not only influence the pipeline in Q1-24 but it will also influence the pipeline and revenue in Q2-24.

Edit: One thing I forgot to mention was how we defined DG and LG.

DG: Companies that don't spend any budget on lead generation campaigns

LG: Companies that spent at least 50% of their budget on lead generation campaigns.

Companies that spent less than 50% of their budget on lead generation campaigns were not included.

Part I: MQLs

Between October 1st, 2023, and March 31st, 2024, our sample set of 87 companies generated a total of 270,811 MQLs. 

Companies with a demand generation motion generated 106,727 MQLs, whereas companies with a lead generation motion generated 164,084 MQLs.

This means that, on average, the first group of companies generated 371 MQLs per month, compared to the second group's 701 MQLs per month. As expected, companies with the lead generation motion had 1.9x more MQLs on a monthly basis.

On the spend side, the first group of companies spent a total of $60.88M within the timeframe, which translates to $1.27M per company in two quarters and $211.4K per month.

The second group of companies spent a total of $43.06M million within the same time period, which translates to $1.1M per company and $184K per month.

This means that companies with the demand generation motion had an average cost per MQL of $570, while those with the lead generation strategy had an average cost per MQL of $262. 

So, companies with the lead generation strategy not only generated 1.9x more MQLs on average, but they also had a 54% cheaper cost per MQL… 

At first glance, lead generation looks way more efficient, doesn't it? Almost dreamy... The ultimate goal of every mediocre executive…

Part II: Deals

Time to go back to reality now, let's take a look at the actual pipeline data. 

23K SQLs were generated out of 106K demand generation MQLs, meaning that the MQL:SQL conversion rate was 21.55%.

Although there were 53% more lead generation MQLs, only 8,087 SQLs were generated out of 164K MQLs, meaning that the MQL:SQL conversion rate was 4.93%. 

This shows that demand generation MQLs have a 4.37x better conversion rate than lead generation MQLs. 

Furthermore, the cost per SQL for companies with the demand generation motion was $2.64K, while it was $5.32K for companies with the lead generation motion. Although cost per lead was 54% cheaper, and the second group of companies generated 1.9x more leads, the cost per SQL for these companies was actually 2x higher—because the lack of intent in these “MQLs” means they simply don't convert.

This also meant that the first group of companies generated an average of 80 deals per month, while the second group generated only 24 deals.

To summarize:

- Companies with the demand generation motion generated $405M pipeline from 106K MQLs

- Companies with the lead generation motion generated $153M pipeline

However, when it comes to the pipeline ACV side, we're not seeing a massive difference—which is fair, considering that the main problem is converting lead generation MQLs into deals. What I mean is that we're seeing a huge problem when it comes to converting from MQL to SQL because the intent in these lead generation MQLs is low—but the ones with intent convert (even the intent is lower than the demand generation MQLs—or maybe it’s just because SDRs are pushy). Hence, we're seeing an average of $20.1K in ACV for demand generation MQLs and $19K in ACV for lead generation MQLs. 

To simply put, companies with the demand generation motion have 5% higher ACV than the companies with the lead generation motion. 

Part III: Revenue

Although we didn't see a massive difference on the ACV part, again it could be because of legitimate intent or SDRs pushing hard or prospects just being curious about pricing, there's still an undeniable difference in SQL:CW conversion rates.

For the demand generation MQLs, the SQL:CW conversion rate is 14.1%, while it is 11.21% for the lead generation MQLs. This means that companies with the demand generation motion have a 26% better close rate than companies with the lead generation motion.

This also means that from 106K MQLs, the first group of companies closed 3.23K deals whereas the second group of companies closed only 907 deals from 164K MQLs. 

On the revenue ACV side, the first group of companies discounted 12% to their final contacts - which meant that the deal ACV was $20.1K and the revenue ACV was $18K; the second group of companies gave 13% discount on average to their final contacts, and their revenue ACV was $16.5K.

In summary, companies with the demand gen motion closed $57M revenue, while companies with the lead generation motion closed $14.9M revenue. 

One last crucial thing is the sales cycles, as expected companies with the demand generation motion had faster deal cycles. The average sales cycle for the first group companies was 54 days, whereas the average sales cycle for the second group companies was 88 days. 

Conclusion

I suppose this is another nail in the coffin. Companies doing lead generation indeed generate more MQLs with lower costs, and this could come across as success in the eyes of investors and executives who just want to save the day.

After analyzing over $100M spend, quarter million MQLs, half a billion pipeline and more than 4K closed won deals from 87 B2B SaaS companies with SLG motions; the data is, once again, clear. I know, especially for digital marketing people, generating cheap and many leads is somewhat addictive - but you gotta keep in mind that this just saves the day. I don’t know if there’s much to say after all these data points; break free from lead generation.