When it comes to sales development, there's always room for improvement. So I never stop looking for new ideas — brainstorming with the team, reading studies, articles, or LinkedIn posts about best practices for subject lines, email templates, CTAs, sequences, and so on.
Yet, it's important that you don't get carried away with the experiments messing things up. That's why it's important for SDRs to rely on hard data, not their gut feeling when evaluating new ideas.
For example, here is the framework I recommend:
If you want to change something (e.g. subject line), start with A/B-testing.
Focus on testing one assumption, one thing at a time — do not change email copy if you want to A/B test the subject line.
Compare a new variant to your historical data, e.g. does the new subject line perform better?.
If the new variant is successful, implement it across all your sequences.
If not, find a new idea and re-iterate — pick a new subject line and do another A/B test.
To illustrate how that works, I'd like to share my experience improving our sequence for inbound leads using this data-driven framework.
So, let's get started 🚀
Version 0
First of all, here are some stats for our inbound sequence from 2 years ago (let's call it version 0):
Open rate - 62.8%
Reply Rate - 14.5%
Interested Rate - 9.8%
As you can tell, we were pretty happy with this email-based sequence. Nevertheless, I wanted to improve it.
So we started our experiments.
Version 1
As the first step, we collected all of the top-performing email templates across our sales and SDR teams, analyzed their open and reply rates, and ranked all of them accordingly.
This allowed us to create a new sequence (version 1) which included 7 emails. I expected this sequence to generate better results than the previous version, but, surprisingly, this mixed sequence didn't work.
As you can see, all the important metrics turned out to be lower. So I had to revert the changes and start over again.
Version 2
Designing version 2 of our sequence, we combined best practices from both version 0 and version 1 sequences. Namely, we haven't changed the templates but instead added a few new steps to the sequence.
As step #1, we added a manual to-do task: before a prospect receives the first email, an SDR had to complete a to-do task and add new variables that would feed email templates on the next steps like industry, correct company name, location. Thus, we made our email even more personalized.
We also added a manual Linkedin Connect step (although there have been some inconsistencies) and slightly changed our subject lines and sequence duration.
Here are the results we've generated with our version 2 sequence:
As you can see, this time we finally were able to increase our open rates, reply rate, and, more importantly, interested rate.
So, we had been using this sequence for the next 2 months.
Version 3
Back in May 2020, we've made a major shift in product strategy to become a multi-channel sales engagement platform. Aside from making phone calls a part of sequences, we've also finally rolled out conditional variables to introduce another layer of personalization to our outreach.
This time, we decided to improve our sequence by including some new conditional variables and changing some CTAs. We've also added phone calls as sequence steps and started A/B-testing a few completely new email templates within version 3 of the sequence.
And here's how it eventually worked out:
Version 4
We had been enjoying the version 3 sequence for almost 6 months. In the meantime, Reply added a ton of new features: SMS and WhatsApp steps, semi-automated LinkedIn tasks (view profile, connect, message, and InMail).
Thus, we've decided to make a new version of our sequence — the first truly multi-channel sequence! And we were pretty excited to see how these changes would impact our results.
Here is the list of all the changes for version 4 (all while keeping sequence duration almost the same):
added an automatic email as step #1 (thus we dramatically improved our speed-to-lead metric)
made a manual to-do task as step #2
added 3 phone call steps
added 1 SMS step
added 1 WhatsApp step
added 3 LinkedIn steps (view profile, connect, message)
added new conditional variables
And here are the results of all these changes:
Version 5
Once our SDR team got used to a new sequence and workflows, I started thinking: we can definitely push things forward one more time.
Since we loved the results of version 4 (and we didn't want to mess things up), we carefully incorporated just 1 additional step in the middle of the sequence: we added 1 manual email step which contained a personalized Vidyard video.
Aside from that, we also slightly changed the order of the steps, adjusted the sequence duration, updated phone call scripts, improved the LinkedIn steps templates. And voila! Here's what the results looked like.
Having a quite polished process already, we managed to make it even more human (mostly because of that manual personalized Vidyard touch), reaching peak performance metrics:
Open rate was on the same high level - 74,6%
Reply rate was 44,1% across all channels
The interested reply rate was 16% (basically 2X of what we had before)
Version 6
While we still use version 5 as one of our key sequences, I couldn't but wonder — is there a way to make it even better?
As it turns out, yes, it's possible to do that! We've been testing some of the new email templates recently and the results have been promising so far 🤞
(although this is still a work in progress and I will be happy to share the results later!)
Wrapping up, here are the results we've been able to achieve going from version 0 to 5:
increase the open rate by 15%
3X reply rate
2X interested rate
You can also check this diagram to see the difference 👇
I hope you enjoyed this post, and let's keep decoding sales development together ⚡️
This proves that consistent, iterative, and data-backed changes can make your sales development process better 💪