The physics of sales, part 2
Seller-push strikes back
PSA: There’s a companion podcast! I go into WAY more detail on my “Physics of Startups” podcast - available on Spotify, YouTube, & Apple Podcasts.
Last week, we covered the theory behind SELLER-PUSH versus BUYER-PULL sales approaches, and the signs you’re operating in the SELLER-PUSH world. See that post HERE.
To recap: The difference is that in a SELLER-PUSH approach, we think that the main force that closes a deal is “us convincing buyer to buy”, where in a BUYER-PULL approach, we think that the main force that closes a deal is “buyer trying to accomplish something on their to-do list.”
The difference between PULL and PUSH changes how you approach every single part of your company, from sales to product to fundraising.
After last week’s post, I got comments and DMs requesting more detail:
ASK AND YOU SHALL RECEIVE.
This week: What does a BUYER-PULL sales process look like, and how is it different from what’s in most of our minds about selling? We’ll walk through each step of the sales process at a high level, comparing and contrasting PULL vs. PUSH.
Here’s a graphic of the sales process from end-to-end; we will go through it sequentially:
Note that this is a typical SaaS company’s approach for an annual contract value of $10k-$100k. Of course, this always needs to be customized for each specific company’s product and customers. It works pre- and post-PMF, and can be modified for larger enterprise deals, smaller self-service or product-led sales approaches, and selling things other than software (e.g., services).
Let’s dig in!
1: Pipeline - Getting potential customers to talk with us
A: Targeting - who do we try to talk with?
PUSH approach
Target people who could buy - they would get value out of it, they have pain points and problems.
PULL approach
Realize that most people, most of the time, would be weird to buy our product. Even if they could buy, they almost certainly won’t given what’s on their to-do list. Instead, focus on targeting people who would be weird not to buy, using the PULL framework:
(Example of PULL framework in use HERE)
B: Messaging - how do we get them to talk to us?
PUSH approach
Send hundreds / thousands of cold emails per day using salesy / spammy messaging that attempts to convince potential customers they have pain points or problems. Get a 0.2% response rate, of which most are OOO replies.
PULL approach
Realize that it’s not our messaging’s job to convince potential customers they have demand; they have demand whether or not we exist. So, pick messaging that offers a conversation that’s worth their time:
“Hey, I’m a second-time founder working on a kinda crazy AI devops engineer, would love to get your brutally honest take on it, open to chat?”
Example results from my campaigns:
(More detail on messaging HERE.)
2: Call 1 - Demand - What are they trying to accomplish?
(More detail on Call 1 HERE.)
PUSH approach
We are taught to do “discovery” to find their “pain points” and “problems.” So we waterboard potential customers with a series of questions to try to understand their current state (“tell me your team structure and roles/responsibilities / how do you do XYZ process?”), then try to surface their problems + pain points (“Isn’t X time-consuming? How much does Y cost?”)
When we do this, we are looking for the wrong things in the wrong way. To potential customers, it feels like we are poking and prodding them, cornering them, looking for their weaknesses and places we can pitch them. We are trying to build the logical case for them to buy our product, rather than helping them accomplish something they’re trying to accomplish.
PULL approach
We are simply trying to figure out their PULL!
What are you trying to accomplish right now, or this quarter? (Get the list of projects on their to-do list)
Double-click into specific projects with the PULL framework, e.g.,
Why this project now, vs. all the others you could prioritize?
What options have you considered or tried, and what do you feel like is lacking with your existing options?
Most startups do a mix of discovery questions and slides for this.
Once we understand their PULL, confirm it like this: “It sounds like you’re trying to accomplish X because Y, considered Z option but it’s not good enough because of A. Does that sound right?”
This feels like we’re on the same side of the table as them, helping them make the progress they’re trying to make.
3: Call 1 - Supply - How do we explain what we sell?
PUSH approach
When we do PUSH-based discovery/interrogation, we have a sense of their problems and pain points, now we need to convince them that our product is awesome and the solution to everything they’ve told us. We therefore describe our product as big as possible - “It’s an AI-native platform that enables any kind of workflow across enterprise applications.”
When we do this, we often take a while to explain the product (I’ve seen descriptions go on for 2-3+ mins, including the company history, product thesis, even TAM), and use words that would make VCs excited because it sounds so ambitious and big. Yet this confuses the hell out of potential customers, who have stuff on their to-do lists they’re trying to accomplish, and don’t know why they need to know about the history of agentic workflows.
PULL approach
Once we know their PULL, we can describe our product/supply simply in the way that fits their PULL. This seems “smaller” than what we’d describe to VCs (or to ourselves in the mirror), but it makes sense to potential customers as the thing that’s going to help them accomplish what they’re trying to accomplish.
Example:
PULL: “It sounds like you’re trying to move from your MVP to a production-grade product because you’ve secured some enterprise pilots; you’ve considered hiring a DevOps engineer or figuring out AWS yourself, but are worried that’s going to take too long.”
SUPPLY THAT FITS: “We offer a way for developers to build and manage a low-cost, enterprise, production-grade application in seconds from a prompt, that doesn’t require you to hire DevOps. Does that sound like it fits what you need?”
TYPICAL DESCRIPTION: “We are building the AI devops engineer for the agentic world! From a prompt, you get an architecture recommendation, cost estimates, a full repo, live infra, and forty-seven other features. We built this out of our experience…”
The latter sounds big and cool; the former fits demand and works.
4: Call 1 - Demo - What do we show?
PUSH approach
When we think that it’s our job to convince them to buy our product, we want to walk them through the product in detail - to prove that it works, and to show them how powerful, flexible, and magical it is. This winds up leading to demos with many clicks, screens, and minutes - I’ve seen demos last 10-20+ minutes!
We often rush to get to this demo, skipping past demand and supply parts of the call, thinking that they just want to see the product and will “get it” when they see the product.
This almost never works. The more product they see, the less likely they are to buy. They just can’t fit it all in their brains. Even if they directionally understand it, the more we show them in a demo, the more complicated the product seems to them, and the bigger the decision seems to buy it.
PULL approach
When we realize that it’s our job to fit their PULL - and we describe our product in the smallest possible way that fits their pull - we realize that our “demo” needs to only show what fits their PULL, and nothing else. This leads to a 30-60 second demo - which may use our actual product, may be a loom video of our product, or may simply be a conceptual example that’s not actually our product.
For example, when I’m demo’ing Waffle, I simply show the output - “based on this prompt, the AI generated this application architecture, deployed this infrastructure, and you can do XYZ from here.” I mostly show a single screen:
They have questions after this, of course, but we can go into their actual questions versus me drowning them in a demo.
5: Call 2 to Close - What’s the path to purchase?
PUSH approach
Call 2 and beyond are, in a PUSH world, a mystical combination of steps that we throw together based on what we think will get them to buy, for example:
Deep-dive demo
Group demo
Business case
Pilot planning
We think that by doing these things, they will be convinced and buy. This leads to us doing a lot of work - I’ve seen very McKinsey-esque business cases - and almost always getting ghosted after months of follow-up.
PULL approach
After call 1, the “sales” part of sales is mostly done: We understand what they’re trying to achieve, why they’re trying to achieve it, why their options suck, and that they think our supply fits their demand.
Now we’re at the “project management” part of sales. Everything after this is helping them buy - and possibly clearing up some final questions they have that are preventing them from championing this to the rest of their organization.
So the structure of Call 2 → Close is:
Whatever they need to understand to want to buy (usually in Call 2)
Whatever they need to do in order to buy (usually scaffold out the steps in Call 2, and execute after)
We are, in other words, designing our post-Call-1 process to be the minimum number of things based on what they need to see and do in order to buy from us. (Note that this might be “no Call 2” and just scheduling an onboarding session, sending a Stripe link, and getting going!)
No matter what the shape of Call 2+ is, it’s our responsibility to take as much work off our champion’s plate to help them get the deal done. Treat everything after Call 2 as “helping them accomplish their project”, not “convincing them to buy”. Examples:
A group demo is not, “It’s my job to sell to everyone”, it’s “I’m here to help my champion sell this to everyone.”
A deep-dive demo is not, “It’s my job to teach them how to use the product,” it’s “It’s my job to understand exactly what they need to see/understand in order to buy, and show them the minimum number of things.”
6: Onboarding/CS - What’s the path to “hell yes”?
PUSH approach
If we sell with the PUSH approach, we generally don’t know why they bought, which means we think about “onboarding” as “getting them to adopt the whole product.” This leads to a lot of training sessions and setup work… that may or may not lead to a successful customer. Often, it leads to them getting frustrated with our onboarding process, and this is for a simple reason: Because we don’t really know what “success” is, in their terms. We know what success means in our terms - “they actively use the product” - and we design the onboarding process to get them to what “success” is for us, which is rarely aligned with the project on their to-do list.
This leads to customer success getting wrecked - we’re not sure who is a churn risk, who is happy, and why. We’re not sure how to change churn risks to not-churn-risks. We get complaints and feature requests but aren’t sure exactly what’s going to move the needle. And so we throw a bunch of this at product + engineering, and they get wrecked too. Everyone gets wrecked!
PULL approach
When your customers buy, they are going to cancel by default. There’s a moment post-sale when they are sold, when they say, “Oh this is awesome, I would be crazy to cancel my subscription.” This is almost always when they have some indication that the project on their to-do list (the thing that caused them to buy) is going to be completed with help from your product.
When we know what needs to happen for them to say, “I’d be crazy to cancel,” we design our onboarding process as a bullet train to that moment, where they would be weird not to hit that moment. Ideally, we have an onboarding call where they hit that moment in the onboarding call. This aligns with what they want - because it’s about completing their project - so we don’t need to drag them through steps they don’t want to do.
Then CS is easier, product is easier, etc. You don’t get wrecked quite as much.
Final thoughts
Know this is a long post, but you asked for it. A few closing remarks:
This post is a starting point, but the specifics are different for each company.
What I find is that PULL & PUSH both compound: The more you need to push early on, the more you need to push later on. When you find pull early on, you tend to need less push later on.
Any moment you find yourself pushing, that’s a signal that something is off.
A way to think about product-market fit is that you are consistently repeating ONE case study (more HERE), where customers pull all the way through.
Bob Moesta’s book “Demand-Side Sales 101” is worth reading!
PS: I run a “product-market fit + sales” course & do 1:1 work with B2B founders to figure out 0-1 sales. Both are booked through September, so if you’d like to potentially work with me in Q4, reserve time on my Calendly for mid-late Sept HERE - more info about working with me is HERE.
PS: My startup, Waffle, has a few beta slots available for our new AI DevOps engineer (1-min explainer video HERE.) If your team wants access, I’m happy to include you in the beta (just email me at rob@reframeb2b.com).
This is it. The playbook to end all playbooks.
Rob, these two editions are amazing. Already forwarded it to 5 people. Great job, thanks for sharing it! Helps me sell why founders should position for demand, not ego ahah.