Scaling Sales: The SDR Motion
The founder’s guide to SDRs (part 2)
Hi all –
This week, I’m continuing on the SDR series. I’ve talked with some amazing experts in preparing this series. Here’s what we’re covering:
Founder-led outbound (last week)
How a functional early-stage SDR function works, and do you still need SDRs in the age of AI and automation? (this week)
How to get from founder-led outbound to a functional SDR org, and when to hire SDRs (next weeks)
I’ve talked with a lot of sales-leaning founders and sales leaders. It’s interesting: People with deep expertise don’t seem to go viral, so you might not have heard of these people. I recommend following them all on LinkedIn:
Anything you read that’s smart, credit to them. Anything that doesn’t make sense is me.
Note that in this post, I use the term “SDR” - this post highlights a replicable outbound prospecting model that can be used by SDRs *and/or* AEs.
Why outbound SDRs?
If you look on Twitter, you’ll see that the SDR motion (and outbound in general) is dead because of AI. Or PLG. Or because of high-volume automated outbound, dark social, brand.
RIP.
This line of thinking is wrong. All of these things make it harder for a mediocre SDR motion to succeed, but the SDR motion is here to stay.
Why? Because well-run SDR organizations do two things that automations can’t:
They maximize the chance that you talk with any single account that’s in-market (Eg by doing multichannel outreach, personalizing, and navigating accounts / doing multi-threaded outreach within an account… but more importantly, doing *whatever it takes*).
They tinker with what’s working based on prospect feedback, so your playbooks organically evolve and get more effective (or at very least, avoid getting *less* effective with time & volume).
There’s a lot that automations and growth teams can do - and I’ll dig deeper into the role of automation / AI and growth teams, and how they can work alongside traditional SDR organizations to experiment quickly and save time and manual effort.
But the core thing to think about SDRs: If you believe that X accounts are “in market” at any given time, SDRs are your way to make sure you talk to as many of them as possible.
In short: SDRs are going to be around for a while.
What does an SDR motion look like when it works?
When it works, the SDR motion becomes a math equation:
Give an SDR X accounts per month
They do Y activities per account
They get Z results per account (demos + revenue)
Here’s a simplistic example (link to spreadsheet):
It’s a math equation that has to work from multiple angles:
Activities: How many manual activities (e.g., phone calls, personalized emails) do we generally expect an SDR to do per day?
Conversion rates: How many people do we need to put through a sequence to get a meeting? And how many of those turn to a real sales opportunity & wind up closed-won?
CAC economics: Based on what we’re paying each SDR and their results, do we have a reasonable CAC payback period?
Revenue goals: How do we build a plan to hit our revenue goals based on our conversion rates, ramp-up time & attrition rate to determine how many reps we need to hire?
When you look at this in a model, you can play around with the parameters to determine where to focus and improve. E.g., should you increase activity by investing in an auto-dialer? Or something else?
But the math doesn’t tell us what the SDR motion looks like operationally. I understood the math, but needed a framework for thinking about the SDR motion. Here’s a graphic for that:
There are 3 main steps in a functioning SDR motion:
Building out a target account list
Which companies are in our target market & fit our ideal customer profile
Ideally, which of the companies in our target are likely to consider buying now
Turning the account list into prospects (individual contacts)
Putting individual prospects into outreach sequences
The whole SDR motion seemed confusing and ambiguous until I realized that everything starts with targeting accounts, not individual prospects.
This forced me to think through which accounts are likely to be in-market now FIRST and clean our account list FIRST — and only THEN move to prospects. This discipline made our targeting and messaging better downstream. As one serial VP of Sales said: “Once you have a tight ICP and solid target account lists, everything else downstream gets easier.”
You can choose whether to allocate accounts to SDRs, and let *them* find and research leads from there. This helps them tailor their approach to each account, but does take time and likely leads to worse data than if you do this yourself. Alternatively, you can give SDRs *accounts AND leads,* then they do a little research on each account and then execute.
You allocate your accounts to each SDR based on your outbound math - many experts said either 5 accounts or 25 leads per day (with a 2-week sequence, this means an SDR manages 200 leads at any one time). You can determine whether to allocate *just* accounts, or to go the next step further and allocate accounts *plus* leads. The advantage of the latter approach is that you can then make sure you’re providing *clean data* - but the disadvantage is that it’s more work for you.
The final step in the SDR motion is taking individual prospects and putting them into different activity sequences. There’s a ton of mythology around sequences - single channel or multichannel? One prospect or many? Customized or not? Sequence length and steps? We move to that topic next.
Two main sequence approaches
There are a bunch of variables for sequences, and every expert brought in slightly different data points from their experiences:
People who sold to brick-and-mortar stores tended to recommend a call-heavy sequence, and also tended to have far less knowledge of “account demand”, so they recommended much more standardization and less personalization
People who sold enterprise tended to recommend longer, multi-channel, personalized sequences to many different people in the account
People who sold mid-market to software companies (e.g., developer tools) tended to have to invest more effort in email copy and LinkedIn/twitter outreach that didn’t feel salesy, as these end customers tend to be overwhelmed with sales outreach.
Because I couldn’t find a unifying approach or an objectively “better” approach, I realized two things:
Your sequences are going to be designed around your “market constraints”:
If you’re selling to a brick-and-mortar business like a restaurant, there’s a store phone they’re likely to pick up. Which means you should probably design your sequences to be call-heavy, and the core thing you need to figure out is how to navigate those cold calls. On the other hand, if you have low call connection rates (sub-5%), you can either optimize towards other channels *or* invest in a parallel dialer.
The looser your account targeting criteria and less “demand intent” data you have, the more you’re going to either rely on 1:1 personalization or volume.
The worse data you have on target accounts, the more you’re going to have to invest in manual data cleaning (e.g., some companies have set up entire teams dedicated to calling brick-and-mortar stores to figure out if they’re good fits, then passing THAT data to sellers).
The less standardized your target accounts (e.g., different titles might buy), the more you’re going to have to broaden your number of contacts per account to try to navigate to the right person
Multiple approaches to sequences can make outbound math work. If you have something that works and replicates, this is no longer your business constraint… go attack that business constraint.
A couple of questions that were on my mind:
How do I think about single channel vs. multichannel? Personalization vs. not? Multiple prospects per company or just one?
Multichannel outreach, personalization, and multiple contacts per company all serve one purpose: They increase the likelihood that any one person will reply. Most experts recommended multichannel outreach because they *also* recommended creating very targeted account lists that have demand. If you have conviction that certain accounts have demand, do whatever you can to increase your probability of talking to them from ~1% to 10%+. (You can also model this out and experiment with it, all with the goal of figuring out which makes your math work best.)
Can you provide a specific example of an end-to-end approach?
Sure - I’m going to anonymize this a bit. I’m working with a company that has an SDR..
I originally set this SDR up with a variety of different approaches:
Automated LinkedIn outreach to certain prospects
Automated email sequences to other prospects
Then, a call + email sequence to other prospects while the automations ran (the call list was auto-generated within Apollo)
The automations were working fine, but they couldn’t scale very far. We needed something replicable, so we pivoted to an account-based approach. After talking with the founder, we realized that a certain-sized company that had just hired a particular role was a good fit, and we could write targeted sequences to different titles in the company based on that hire’s goals and priorities.
I give the SDR 25 new accounts per week, who then prospects each company to determine who to contact. I have created 5 different multichannel sequences based on different titles (e.g., engineering leader vs. product leader). Each has a first email that can be customized, and then is a mix of calls, emails, and LinkedIn. If we have a mutual connection with one of the prospects, the SDR can go and request an introduction or change the sequence approach.
Hope this helps! Happy hunting!