What I said on a podcast this week that I'd stand behind in writing too
I recorded an episode with Jonathan Bland at Omni Lab Consulting this week for their podcast, and about ten minutes in, they asked me the question I get asked in almost every ABM conversation I have.
What's actually different between ABM and demand gen? Isn't it all just targeting, eventually?
Here's the line I used, and it's the one I'd defend in front of anyone: ABM is fishing with a spear. Demand gen is throwing a wide net. One is built to hit a specific thing. The other is built for volume and reach.
The part I didn't expect to say out loud until I was already saying it: I don't start ABM with an account list. Most teams do, and it's the wrong first move.
I start with business objectives then product use cases.
At Sphera, we're selling five different product lines into large enterprises, and an account list with no use-case anchor tells a rep nothing about how to open the conversation or what the end goal is. The list is downstream of the strategy, not the starting point.
We also got into how many accounts should you actually focus on for ABM.
My rough number: about 10 accounts per rep is where real ABM personalization holds up. Fewer than that and you're not running a portfolio, you're doing one-to-one work under a different name. More than that and you're back to demand gen with extra steps, no matter what you call it in the deck.
And the demand gen campaign logic doesn't carry over either. Demand gen runs on themes, something like AI adoption, reinforced across channels all year.
ABM skips that structure completely.
There's no intermediate campaign concept. It's built directly around the outcome you want from a specific account, which is why we're piloting something specific to our org structure right now: not a demand campaign, but 48 customer accounts identified to build cross-sell pipeline around a specific product.
There's more in the full conversation, including where I think ABM has no business existing yet (hint: if you don't have a real ICP or a named-account process, skip it for now).
Worth 45 minutes if you've ever had this argument with your own team.
