From Unmatched Data to Hundreds More Meetings: Snowflake's AI-Driven Marketing Breakthroughs

Check out how Snowflake’s marketing intelligence team uses Snowflake AI — from automating global data translation to implementing a 99% accurate spam filter — resulting in hundreds of new meetings and a significant improvement in lead quality.

We’ve seen how Snowflake AI tools are transforming outcomes for our customers. From strengthening supply chain operations to improving healthcare staffing outcomes and saving 54% of costs, AI in Snowflake is making a real impact on businesses around the world. 

That same transformative power is at work within Snowflake, too. Here, we dive into how Snowflake’s marketing intelligence and developer relationships teams are using Snowflake AI tools to drive significant improvements in how they connect with customers and allocate resources.

As a company with customers across the globe, Snowflake must be able to connect across regions, languages and even alphabets. But when customers and prospects fill out forms for gated content such as ebooks or webinars using non-English characters such as Hangul or Japanese, the listings may not match in English-based databases like Salesforce — hindering accurate lead matching and account management. So the marketing intelligence team decided to put Snowflake to work.

Using Snowflake Cortex Complete, the team automated the translation of Korean and Japanese company names. Then they would do some additional checks for accuracy and write it back to Salesforce. This enabled previously incompatible data to flow seamlessly into the lead generation system without requiring extra manual translation or costly external services. 

“The actual implementation was very easy since we update our Snowflake tables and push information into Salesforce every day,” says Daniel Chow, Senior Data Scientist at Snowflake. “The beauty of Snowflake Cortex is that it works in line with SQL. If you’re already updating your tables in SQL, it’s just another line.”

The effort led to the integration of 8,000 previously unmatched accounts. With the translation system automated, it’s been running without the team’s intervention and has led to more than 300 additional meetings booked to date.

In marketing, there are plenty of ways to connect with interested customers. But enterprises want to be sure that when they’re reaching out, they’re actually going to reach humans who want to be reached. This can be a challenge sometimes when trying to filter spam personas. To help the marketing and sales teams focus their efforts on genuine prospects, the marketing intelligence team used Snowflake AI to implement a persona screening process. 

Chow and his team used Snowflake Cortex to identify patterns that indicate whether a lead is spam — like a first name of “Test” or “XXXX” — to help improve lead scoring and prevent time wasted on outreach to illegitimate personas. By fine-tuning the model over many thousands of samples, the team was able to create a model that could screen out spam personas with 99.8% accuracy — and help determine an effective lead score. “Ultimately, Cortex can provide a high enough accuracy and low enough false negative rate to reduce meeting cancellation rate,” says Chow. Now, sales reps no longer have to waste time pursuing leads that have a very low likelihood of getting results. 

Over time, the consistently low conversion rates from leads flagged as spam validated the model's effectiveness, leading to its integration as a direct filter. The result is a significant enhancement in lead quality. “We've all struggled with the boundless ‘creativity’ of people and bots who fill out marketing forms with novel, irrelevant information, like Bruce Wayne or unmentionable expletives,” says Travis Henry, Senior Director of Go-to-Market Operations at Snowflake. “Hard-coded filters don't solve the full problem and obvious junk (to humans) still slips through to sales. This solution effectively puts a 24/7, intelligent reviewer on every single lead to help ensure the junk is thrown aside and only relevant leads get scored, routed and followed up with.” 

By preventing the processing of likely spam inquiries, Snowflake optimizes its resources and helps make sure that sales and marketing efforts are directed toward genuine prospects. 

To learn more about how customers are using generative AI with Snowflake for real-world results, download the Secrets of Gen AI Success report or explore Snowflake’s AI solutions.