Content and campaigns can wait. They take months to pay off, and most early companies do not have months to spare before they need pipeline. Talk to buyers first, learn what they respond to, then build the rest around that.
Choosing the one channel to fund and focus
Outbound
Signal in two to three weeks, pipeline in six to eight. Works when you can name specific companies who have the problem and the buyer is reachable. We usually start here. You control your actions every day, and nothing else teaches you the language your market uses this fast. Volume without a reason for contact is where it breaks.
Inbound and search
Give this six to nine months before expecting real pipeline. Lead with it only if your category already has search demand and buyers know the term to type. Otherwise let it run quietly next to outbound. The research you do for it, the questions buyers are actually typing, feeds your outbound messaging too.
Partners and community
Timing is unpredictable but conversion is high when it lands. If your buyers already gather somewhere, an accelerator, an industry association, the consultants who advise them, this will beat everything else you try. It costs relationship time, and founders almost always underestimate how much.
Paid
Good for testing a message fast. Weak as your main motion in complex B2B because the click has to survive a long buying process. Using paid to find your ICP is an expensive way to buy noise for early stage.
Do the pipeline math before you build anything
Start at the revenue target and work backward. Target divided by average deal size gives you deals needed. Deals divided by close rate gives qualified opportunities. Opportunities divided by your conversation-to-opportunity rate gives conversations needed. Look at what that means per week and ask if one person can actually produce it.
Usually the answer is no. Then the deal size has to go up, the target has to come down, or the motion has to change. Better to find that out on a page than three quarters into the year.
Where AI genuinely helps
Research and personalization at scale is the obvious one. Reading a company's public material and drafting a message that reflects their situation instead of a merge field. Account selection is the other, pulling intent signals from sources that would take a person days to check by hand.
There's also a hiring case. One BDR with the right AI tools can cover research, list building, and first-draft personalization at a volume that used to take two or three people. That doesn't replace the hire. It changes how far the hire goes before you need the next one.
A bad message doesn't get better at volume, with or without AI.
How to tell in a quarter whether to continue
- Reply rate above five percent on hand-written outbound means the message is landing. Below two percent, the target or the message is off, not the volume.
- Opportunities stalling at the same stage twice points to a process gap, not a demand gap.
- If the only replies come from companies who can't afford you, go back and take another pass at the ICP before spending anything else.
Review your numbers weekly - it will be gleaning insights at first but then it will produce trends and actionable intel. The goal is to be able to understand what levers to pull and what we can tweak to see change.
Common questions
Where should an early-stage B2B company put its first dollar?
For most early-stage B2B companies that's outbound. You can control it including your effort and time on it. Nothing beats in-person events and warm introductions but if you need to start building today, find the right channels where you can reach your prospects and provide value to them, don't just ask for 15mins.
How much pipeline do you need to hit a revenue target?
Qualified pipeline is critical to keep an eye on. This typically ranges from 2-5X depending on the industry. Work backward from your own numbers, not a benchmark. Target revenue divided by average deal size gives deals needed, divided by close rate gives qualified opportunities, divided by conversation-to-opportunity rate gives conversations. If you don't have those rates yet, that's the first gap to fix.
How long before a demand generation motion shows results?
Outbound shows reply-rate signal in two to three weeks and pipeline in six to eight. Content and SEO rarely produce meaningful pipeline before six months.
What are buying intent signals and are they worth paying for?
Intent signals are observable behaviors suggesting an account is in-market: hiring for a related role, adopting a complementary tool, funding news, research activity on relevant topics. They're worth paying for once you have a message that converts, since they change who you contact, not what you say. Other ways to gauge intent signals is who is opening your emails and clicking your links repeatedly. If you don't have intent tools, start there.
Can AI replace a demand generation hire?
No, but it changes what the hire spends time on and how far one hire goes. AI is good at research, focused list building, drafting personalized variants, and surfacing intent signals across a tech stack, often enough that one BDR with good tools produces what used to take two or three people. Deciding what to say, judging whether a reply is real interest, and running the conversation still need a person.