In your GTM machine for B2B Saas Software, one of the best defenses against spam emails is to ask for a business email in your lead form. By having this initial barrier, you make it harder for prospects to use free personal email addresses to clog your funnel. You’ll need to screen for disposable domains, but finding a real prospect is much easier.
But what happens when you can’t use that filter as an initial sort or when your ideal customer uses a personal email to identify themselves? Finding a spammer then becomes a combination of detecting behavior and validating other information you get during the lead flow.
Visualize the ideal customer
To find a spammer, consider the ideal customer, then invert the behavior.
For example, a logical progression for a typical prospect might be to start a trial and make progress through that trial over a few days. When a prospect immediately becomes a customer with no prior contact in that setup, you might feel a little suspicious. However, if your GTM involves customers purchasing software on the first visit, this behavior isn’t spammy at all.
When you expect to call a prospect on their phone number because your clients typically run a business, it’s odd if the phone number they gave you doesn’t get answered. If that same prospect has an odd-sounding email and shares a phone number that’s a VOIP number rather than a wireless or business phone, your fraud radar might increase.
If your prospect is flagged by your credit card processor as a high-risk card, you have another data point toward your answer.
Does that mean that every prospect with an email of gamerguy123@freeemail.com is automatically a spam prospect? It depends how much that prospect resembles the kind of prospect who would be a good customer, multiplied by the other signals you observe.
Start by looking at the facts and guess whether it ends up being spammy. Over time, your detection and prediction will get better, tuned for your company’s GTM.
Build the data factory of tests
Your “guess” must be objective, and the rule or rules to mark something as “potentially spammy” must be as easy to determine as possible. It’s better to have a easy choice instead of a vague description of what makes spam.
Here are a few examples of behavior that suggests a spam prospect:
-
use of a “disposable domain” email address, typically with a developer tool like mailinator.com – this hides the real email address
-
submitting a website URL that is not related to your company for the purpose of passing through a form
-
use of a temporary or “burner” phone number connected to a VOIP service
These items can all be tested to be true or false, and the outcome of these simple tests is a scale from 0 (doesn’t trigger any of these tests) to 3 (triggers all of the tests). Is every prospect marked a 3 on this scale a spam prospect? Probably not, but a 3 on this scale is a lot more likely to be spammy than a 0.
Use data to refine your tests
The best way to measure your tests? Compare them after the fact to the actual results. In this (very) simple scenario, if a spam score of 3 ends up being a spam account 90% of the time, you might choose to remove those leads from your lead flow entirely, or qualify them differently to create a better mix of leads for your sales team.
Tests don’t tell you the whole story. A middling prospect with low activity in your application might look like any other middling prospect, so the critical path here is to contact the prospect and have a conversation.
What’s the takeaway? Product-led growth sometimes attracts customers that look like spam. The best defense? Contact them like any other prospect and find out what they’re up to. Adding spam signals to share with your sales team will help them know more.
