Twilio on Twilio: Using Segment to optimize ad campaign performance

August 07, 2023
Written by
Lyssa Test
Reviewed by

The last few years have been hard on marketing teams and advertisers. Rising customer acquisition costs paired with the gradual depreciation of third-party cookies have made finding and converting qualified leads harder and more expensive than ever before.

Twilio’s Demand Generation (DG) team wasn’t immune to these trends. In fact, they noticed a few additional hurdles holding their team back from optimizing ad spend and campaign performance, including:

  1. Limited in-platform ad targeting capabilities: Twilio runs ad campaigns across many channels, but not all ad platforms have the specific targeting capabilities Twilio needs. Our DG team wanted a reliable way to ensure our ads reached the right audiences, not wasting impressions on low-quality leads.
  2. Slow manual processes: Twilio’s DG team spent a lot of time manually uploading audience lists to various ad platforms. From building custom audiences to reformatting them for each ad platform and manually uploading lists, this process was a huge time drain for the team.
  3. Poor quality conversions: In-platform audiences often led to poor quality conversions. Our DG team wanted a way to ensure their budget only went to qualified leads interested in Twilio products and services.
  4. Lack of accurate conversion tracking: An in-platform conversion doesn't necessarily translate to a qualified lead, pushing the DG team to look for a new way to safely and compliantly share customer relationship management (CRM) data directly with third-party ad platforms.

To tackle these challenges, the DG team turned to Twilio Segment for help.

Look-alike audiences

Different ad platforms have different targeting capabilities, with some better than others. As a B2B company targeting specific personas and job titles, advertising to the right audience on channels like Facebook and Instagram can be a challenge for Twilio. To help focus our campaigns on the right individuals, our DG team often leverages look-alike audiences or a group of users with similar characteristics to existing Twilio customers.

Before Segment, look-alike audience creation was a flawed manual process. Unqualified leads would inevitably infiltrate these lists and convert, which would lead to wasted ad dollars and a high volume of unworkable leads.

Now, with Segment on our side, our DG team can build much more qualified audiences modeled off of sales accepted leads or opportunities to target better audiences (workable leads) on the platform.

Audience retargeting

Retargeting is always a great way to re-engage qualified, motivated audiences and ensure your business stays top of mind as they progress along their customer journey. But this seemingly simple strategy proved to be a logistical challenge for the DG team. To build these retargeting audiences across more than seven ad platforms and Twilio’s martech stack, they had to install pixels on the Twilio domain for Google, Facebook, LinkedIn, Bing/Microsoft, Reddit, Capterra, and more. This led to numerous unintended consequences, such as:

  • Slowing website performance and speed
  • Creating more work on the operations side
  • Leading to potentially redundant retargeting pools across the channels

Not to mention, the list-building process could take the DG team up to two weeks to build.

Looking for a more efficient way to build and maintain retargeting audiences, the DG team turned to Segment. “Segment now acts as an air traffic controller for our retargeting data,” says Brandon Borden, Twilio’s former Principle Demand Generation Manager. “Now, when we want to build a retargeting list, we're confident that it’s backed by strong first-party data. We can seamlessly build lists and push them to any of our ad platforms—from one place.”

Segment seamlessly enrolls users into these lists who complete any high-intent actions, like landing on a specific page, filling out a demo request, or downloading a piece of content, so the DG team no longer has to maintain them manually. This new list-building automation has also been particularly helpful for cross-sell and upsell purposes.

Since using Segment and incorporating these stronger retargeting pools, Brandon says the team has reduced cost per acquisition by 10%. “We've also unlocked the chance to build automated, multichannel campaigns across paid, web, and email with retargeting lists powered by Segment. This provides us the opportunity to drive more impact to the bottom line by influencing product adoption, acceleration, and more,” he continues.

Audience exclusions

Next, our DG team wanted to reduce inefficient budget spend. They knew their campaigns inadvertently targeted unqualified leads and wasted ad dollars on users who’d never convert. Or as Twilio’s former Group Manager of Demand Generation Anneke Cavness puts it, “The more we can exclude users we don’t want to hit, the more efficiently we can spend our budget on the users we want to get in front of.”

So who didn’t make it? Our DG team wanted to exclude unworkable marketing qualified leads (MQLs) or leads rejected due to bad contact data, fraudulent/spam sign-ups, or previously unsubscribed from Twilio. Depending on the campaign, they also would exclude users who converted, like someone who completed an account sign-up or who is a Twilio customer. That way, they could invest in driving new business, not targeting existing users.

With Segment, the DG team can now feed insights directly back into ad platforms to filter out these poor-performing audiences. “We’re filtering them out to ensure our ad campaigns are being as efficient as possible with our spend. This has also helped us improve our lead quality from paid since we’re no longer hitting those unqualified audiences,” explains Alexis Cattermole, Twilio’s Director of Demand Generation.

The DG team has also used these improvements for Twilio SendGrid. “We’re operating a semi-complex operation from a DG perspective because we acquire customers for multiple products. With Segment, we can sync those audiences across our different website domains and databases. We can filter out the poor-quality audiences we find on the SendGrid side from our Twilio campaigns, which would be impossible without a solution like Segment,” Alexis adds.

The best part? The impact of these exclusions will only increase over time. “When you apply those audience exclusions to all of our Google and Facebook campaigns, it has a material impact. Over time, these audiences will continue to get larger, so we anticipate the impact of these Segment exclusions to only increase over time,” says Anneke.

Conversion tracking

Since Apple implemented the App Tracking Transparency (ATT) framework with the iOS 14.5 update, which allowed users to opt out of being tracked across applications and websites, many advertisers have struggled to understand their mobile campaign performance. Unfortunately, Twilio is one of those advertisers.

“Many of our ad platforms have seen diminishing performance in the last few years. Apple’s iOS update really killed the performance of our Facebook campaigns and channels that are reliant on look-alike audiences,” shares Alexis. “Essentially, the platforms are losing visibility into conversions, meaning we no longer have a clear understanding of who is engaging with our ads and turning into meaningful customers for us,” she continues.

That’s where Segment helped change the game. “What Segment has allowed us to do is essentially backfill bottom-funnel conversion data into our advertising platforms and build look-alikes using our own first-party data. That’s helped us improve some of the performance gap that we were seeing from some of those channels reliant on look-alike audiences last year,” she adds.

Now, the DG team uses Segment to track ad conversions and share that data from our CRM directly to Facebook. That helps the DG team train the ad platform’s algorithm to go after quality conversions rather than all conversions.

Measuring impact: how Segment helped drive $250,000 in influenced pipeline in just 2 months

Before using Segment, the DG team saw a significant number of their campaign-generated leads disqualified due to fraud, spam, bad contact data, or unqualified contacts.

But almost immediately after implementing Segment, the Twilio DG team saw significant improvements in campaign performance and lead quality. In fact, within 30 days of implementing Segment, the DG team’s percentage of workable leads increased from 73.3% to 76.4%, a percentage increase of 4.2%. More importantly, the number of unworkable leads decreased considerably by 20.5%. Both trends continue upward when looking 90 days out as well.

Metric30 days pre-Segment30 days post-Segment90 days pre-Segment 90 days post-Segment
% workable73.3%76.4%73%76.8%
% unworkable20.5%16.3%21.6%17%
% sales engaged6.3%7.3%5.4%6.2%

Another big win for the DG team was a major uptick in the influenced pipeline from these new, higher-quality leads. While cost per MQL remained the same (as expected), the DG team was excited to see the pipeline associated with generated opportunities increased by $250,000 in two months.

While these quantitative improvements prove the power of Segment, Anneke says the best part of using the customer data platform (CDP) is how it has improved the efficiency of her team’s work. “For us, Segment is a quality and efficiency play,” she says. “We can quickly build an audience in Segment and push it to multiple different platforms in a few clicks, all the while knowing that these audiences will automatically refresh in real time. By targeting more qualified audiences leveraging our CRM data, we can more efficiently spend our marketing dollars on the right people,” she continues.

And the DG team says they’re only scratching the surface of what Segment can do. They plan to use the platform to build personalized customer journeys across all Twilio properties, including custom webpages, bespoke email campaigns, personalized ads, and more. The DG team is excited to see what else they can achieve with Segment.

Optimize your advertising strategy with Twilio Segment

Advertisers have faced significant challenges in the last few years, like Apple’s ATT framework and the quickly approaching deprecation of third-party cookies, which have made acquiring new customers harder and costlier. Pair this with economic uncertainty and a time when marketers must “do more with less,” and you have a hard battle to win.

Luckily, having clean, organized data and the best tools available, like Twilio Segment, can help your marketing team drive results even in the face of uncertainty. Whether looking to acquire new customers or upsell existing ones, delivering a message that’s audience-, time-, channel-, and behavior-specific has never been easier, thanks to Segment.

Ready to see the leading CDP in action? Learn how Twilio Segment can help your business use first-party data to reduce customer acquisition cost and drive return on ad spend.