When marketing leaders at Linode, an established cloud computing company, reached out to Opascope, they were in the process of increasing their investment in paid advertising without seeing a related improvement in their bottom line.

We were able to help by: 

• Building out an attribution system to help track how the platform signups were moving down the funnel.
• Using this acquisition system to identify and prioritize keywords where signups were more likely to turn into paying customers.

What follows are the steps we took to optimize Linode’s funnel and, as a result, increase the company’s monthly paid signups by 660%.

But first, let’s explain why the problems Linode experienced are so common for SaaS companies using a product-led growth (PLG) model.

A common PLG issue: optimizing for lead quantity over quality

For companies using a PLG model, the primary goal of most marketing initiatives is to get as many users to try out the platform as possible and let the free version or trial do its job.

But this “cast a wide net” mentality becomes an issue when you only look at the top of the funnel. Because the first conversion that has real company value — the moment when a signup becomes a source of revenue — happens when users first pay for a service.

When you’re running a paid motion, if you simply focus on the number of new account creations, you can easily be paying for signups that have a low lifetime value (LTV) or never spend money on the service at all.

We’ve found this is a common problem with PLG companies, and it was the case with Linode.

Prior to working with us, Linode’s leadership and their former marketing agency were aligned on a primary goal being the number of account creations.

By their own KPIs, the agency was doing well. New signups were consistently coming in, and their cost per acquisition (CPA) was low.

But if you took a closer look at the pie chart of channels bringing in paid customers at the time, you’d find that paid ads accounted for an insignificant amount of new paying customers.

In other words, the signups that current campaigns were bringing in never had the intention of converting. 

This is why it’s common for PLG companies not to see the ROI they expect from paid ads.

But there’s a barrier that often gets in the way of fixing this issue: having an advanced enough attribution system in place to identify higher-quality signups.

Another common PLG issue: overly simplistic attribution

Proper attribution is always a headache for tech companies. Between tracking leads across several platforms, sessions, and devices, attribution is an imperfect art. These issues can be magnified if a company’s analytics solutions and tracking models lack the nuances to truly align with its business model.

Two things that complicated tracking attribution for Linode were a long trial period and usage-based pricing.

1. Long trial periods complicate tracking paid performance

Longer trials have their benefits: they give your prospects more time to try out the platform before making a decision, which means you have more time to win them over.

This is how Linode operates. New users can try out Linode’s services for up to 60 days before being charged. 

But when you’re tracking paid performance, this also means waiting at least sixty days to determine the quality of leads a campaign was bringing in. Other complications can delay this period even more, such as delays in monthly billing.

And this wasn’t the only challenge.

2. Usage-based pricing can make it more challenging to calculate a customer’s value

Usage-based pricing (UBP) is on the rise for SaaS in general and is common for cloud computing platforms where the value of the service depends on how much storage compute or bandwidth you use.

But, compared to other SaaS pricing models, usage-based pricing makes it more difficult to accurately track ad performance and estimate a customer’s lifetime value (LTV).

With other SaaS pricing models, such as tiers or pricing by the user, a customer’s value is fairly stable because they’re committing to a given price for a period of time.

But usage-based pricing is much more variable. The value of a user is more likely to change from month to month or year to year. 

When Opascope was brought in to partner with Linode, the company’s analytics setup wasn’t designed in a way to account for these complications.

According to Rob Yoegel, Linode’s Senior Director of Marketing, Linode was maintaining a four-step funnel “that, at a very high level, nobody really understood when it came to the efficacy of our performance marketing investment and how it was impacting those four areas of our funnel.”

To truly understand which campaigns were bringing in the highest quality signups, we needed to build out a more advanced attribution system where Linode’s marketing and revenue teams would be able to connect a campaign’s performance directly to revenue.

This would not only allow us to tailor and optimize campaigns, but it would also allow revenue teams to understand and support other initiatives as well.

Three best practices for creating a better attribution tracking system

With Linode’s trial and pricing models, as well as its large payment processing volumes, it took several months to create and implement a unified attribution tracking system and dashboard.

But when it was in place, this new system enabled the company to have significantly more transparency into how various marketing initiatives were working and what their value was — paid and beyond.

 “The piece of Opascope that I really value is the metrics and analytics side. The fact that they make it very simple to understand,” Rob said. “That allowed us to turn the dials up or down on our marketing investments, which is super important to me.”

Here are some of the factors we look at when building out an analytics system and tips for PLG companies in similar situations.

1. A data warehouse and modern analytics tools are essential

These days, there are a number of great analytics tools that work together to make granular and nuanced attribution tracking obtainable.

You’ll need a data warehouse like Snowflake or Amazon RDS, event-based analytics platforms like Segment or RudderStack, and a tool like Sigma or Tableau to go through the data and analyze it properly.

Note: Want customized recommendations on what solutions to use for your analytics setup? We can help!

2. Partner with a business intelligence specialist to build out a spec for the engineering team

Connecting attribution tools to a company’s platform often requires the partnership of their engineering team.

We’ve found that people on marketing and analytics teams often know what needs to be done to improve analytics but have trouble putting it into a spec for their engineering teams to build out.

If you don’t have a technical business intelligence specialist on staff, we recommend hiring a consultant or working with a team like ours that specializes in being able to communicate with engineering effectively. 

This requires asking questions to understand things like how user flows and billing processes work so that we can make recommendations that work inside of a platform’s existing processes.

Here’s Alex, our VP of Digital Strategy, on another problem he runs into: companies spinning their wheels trying to attribute 100% of revenue.

3. The spec needs to align with your business model

This may sound basic, but you’ll never be able to properly track attribution if your analytics setup doesn’t match your go-to-market motions.

For instance, when we were trying to estimate the potential value of one of Linode’s ad cohorts, we needed to factor in the range of value a prospect may have based on their product usage. For instance, to estimate their value we looked at things like a user’s location and whether they were using a personal or business email domain.

A paid strategy to improve lead quality for PLG companies

After we’d optimized Linode’s attribution system, the next step was using these new attribution insights to better understand which ad campaigns were bringing in the highest and lowest quality signups and adjust the campaigns accordingly.

Here’s how we did it.

1. Question what conversion you’re using when you judge paid performance

Linode’s previous performance marketing agency had reduced the company’s cost per lead by 30% to 40% while consistently bringing Linode a ton of new signups. Supposedly they were making the paid channel run more efficiently, but even at a reduced cost, paid advertising still wasn’t profitable.

Why? Because they were focused on the wrong conversion.

If the leads you’re bringing in the door aren’t converting to paid customers, they don’t have value.

So when you’re trying to assess the value and efficiency of a paid advertising program, it’s important to focus on actions that are truly adding value to the company.

In the first few months that we worked on the account, the cost per new signup actually went up 30%-60%, but the overall price per paying customer kept getting lower.

How is that possible?

Because we were pursuing higher-quality signups that were more likely to spend money on the platform.

2. Higher-cost keywords are worth it if they’re more likely to convert

The reason cost per signup was low was that the keywords being targeted were on the more affordable end of the spectrum.

For instance, if you target keywords with low-quality modifiers like free, cheap, or promo, you’re more likely to bring in leads that will rarely convert into paying customers.

In this case, your cost per lead may be fairly low, because you’re targeting cheap keywords. But the value of these leads is also low.

To illustrate this issue, let’s look at keywords related to a search for social media management platforms. Higher value SERPs may include users searching for “best social media management tools,” searches that compare popular tools or are looking for tools to manage specific social media platforms.

These kinds of searches signify a more serious buyer.

These are examples of obvious low-quality and high-quality keywords. But with a proper attribution system in place, you can get a much more granular look at the value of keywords you’re targeting in an ad campaign.

Which is exactly what we did with Linode.

As we targeted keywords with the highest likelihood of converting to a paid customer, the price per paying customer went down 80% to 90% despite the increase in the signup cost.

According to Rob, “The significant value Opascope brings to the table is maximizing the budget and maximizing the return.”

Growing paid ads into one of the main acquisition channels

In the three years since Opascope began partnering with Linode, paid ads have begun to bring in an average of 30% to 40% percent of the company’s new paying customers every month. 

But the other key benefit has been helping Linode have more transparency into their marketing investments through a more robust metrics and analytics base.

This has given Rob the ability to present trustworthy metrics to his marketing team and the rest of the organization. 

“The end result is that we know that more than four out of every five new customers that come from our product-led growth approach is attributed to a marketing investment,” he explained. “And with that knowledge, I’m able to go to other leaders and ask for more investment, tweak the investment, and change things, and Opascope is directly responsible for that help in that partnership.”