How to Automate Affiliate Qualification: A Case Study from SEMrush

The largest affiliate networks receive hundreds of thousands of requests per day. For them, automation of entries is an undisputed need.

In-house affiliate programs, which exist in many SaaS companies, manage a daily flow of hundreds to thousands of incoming requests. Processing them manually is an overwhelming task for a manager, yet businesses are often slow to consider alternatives.

Many doubts arise regarding possible optimization. Will automation spare a lot of the affiliate team’s productive time? Will the system ensure the quality of selected requests? Will it pay off for the business?

We tackled the same concerns at SEMrush’s affiliate program, and we can answer “yes” with certainty to each of these questions. Our automation solution is based on website traffic analysis. Let me explain it to you in detail.

In-House affiliate program handling incoming requests: The “Before” experience

SEMrush is an all-in-one digital marketing suite ensuring competitive intelligence for the entire team. We handle our affiliate marketing in-house, with the exclusive BeRush program. Over the years, BeRush itself built its strong brand awareness, which led us to receive 150 new registrations a day in the affiliate program.

To a manager, qualifying these registrations manually would take two hours of work a day. And that doesn’t even count the number of affiliates whose contacts are already in our database. This is 30 thousand potential partners waiting for the right moment to renew their contracts. Thousands of valuable opportunities could be profitable to the business.

Fortunately, as part of SEMrush, we’re no strangers to data technology. Having acknowledged the problem, we started looking for a solution in-house.

How we introduced traffic analytics as an affiliate qualifying solution

Step 1. Specify the metrics for scoring of all incoming submissions

We studied the workflow that our affiliate managers followed and decided on the qualifying metrics. For us, the first criterion is the amount of traffic a website receives.

A large number of visits to a website signals a high chance for us to reach a sufficient audience there and attract new users to SEMrush. Traffic volume also reflects the website owner’s investment in marketing, thus it may be a sign of a highly reliable prospective partner.

Step 2. Set up a workflow for gathering these metrics

We used SEMrush Traffic Analytics, the tool from the SEMrush suite, to collect data on our prospects’ website traffic. The interface revealed stats to use, suitable for benchmarking, and drew clear graphs that immediately showed whether the website was growing or losing visitors.

Our primary metric was Traffic Volume. The key point for us was looking at the potential affiliate’s performance in dynamics. We were not going to get involved with a partner whose time had passed.

We were completely satisfied with the data available in the tool, but we realized that we were still collecting it manually, and this meant a lot of work. We had to take one more step forward.

Step 3: Automate data gathering with websites’ bulk analysis

We moved on to the Bulk Analysis of all of the requests. Previously, we could only obtain traffic data for five websites side by side in the interface. Now we could enter a list of up to 200 domains and export them to a .csv table.

Depending on the number of incoming submissions, we could dedicate time to this procedure daily or weekly.

At that point, we already felt relief from the workload as the process was now 10 times faster.

However, we were not completely happy yet and went further with the optimization of affiliate qualification.

Step 4: Integrate API into the current affiliate scoring and monitoring system

This was our key to success: we introduced full automation of the process by connecting SEMrush Traffic Analytics via API to the internal BeRush data system.

From that point on, we no longer had to go back and forth to the tool to export traffic data. With seamless integration, as soon as the new registration was submitted to BeRush, all the needed stats were delivered to our sheets. We could quickly check Traffic Volume and get into more detail if we needed (such as, compare websites’ Bounce Rate and Traffic Sources). All we had to do was set alerts and custom triggers.

This way, a flow of hundreds of new requests turns into a prioritized list of the most valuable prospective affiliates.

Besides the scoring of new submissions, we picked up the pace of reviewing previously registered users. Some of them entered our affiliate program a year ago and now needed to be reactivated. Others signed in, eager to promote SEMrush, but didn’t succeed, so we may want to revisit our partnership with them.

With extra filters, segregating the program participants by their location and date of registration, it became more time-effective than ever.

What did we gain from automated affiliate scoring?

All in all, we processed the entire affiliate base, amounting to 30,000 partners, and determined the most (potentially) profitable among them.

BeRush managers now spend no more than 30 minutes a day on sorting requests, compared to their previous two hours. People will always use human intelligence to pursue their work goals, but technology can greatly help with routine tasks.

Ready to get started? Learn more about Tapfiliate’s automations and additional features here.

Elena Kozlova

Elena Kozlova is a content marketing strategist at SEMrush. Passionate researcher, writer, and listener.