I speak with B2B GTM leaders everyday, and everyday I hear the same thing: Pipeline is hard again. They tell me that the channels and methods they’ve used for years no longer work, and incorporating AI-powered personalization isn’t moving the needle (yet). And, according to the latest data from Omar Akhtar, Chief Analyst at benchmarkerdata.com, they’re right: Lead volume is declining. Conversion rates are down 5 to 10 percentage points across the board. And the cost of every channel — be it email, social, or paid ads — is up.

But that’s not the whole story. There are plenty of teams out there that are seeing success today, and that’s not because the fundamentals of pipeline have changed. Successful teams are thriving for the same reasons any pipeline team has ever thrived: They’ve nailed the who, the when, the what, and the how. The only difference between the strutters and the strugglers is that the former group has adapted each of those elements to the evolving technological, market, and buyer landscape.

In this article, we’ll unpack the new-school approach to each of those elements and what you need to do about it.

As Kathy Macchi (Co-Founder and EVP, Innovation, Inverta) said on a recent episode of The Transaction, the who is the most important part of the equation. That’s why, when I was the Chief Analyst of TOPO, I spent years speaking and writing about account-based strategy. 

At its core, account-based was always about focusing efforts on the right people. And though the term may seem old-school today, the concept is more important and effective than ever. The only difference is that, nowadays, the cool kids have taken it to another level with microsegmentation.

The process involves breaking down large account lists into subgroups that are likely to engage with you, buy your product, and realize value in the same way. For example, at Scale’s recent GTM AI Summit, Flatfile’s founder David Boskovic explained: “[Our] GTM starts by creating microsegments of no more than 500 accounts who have the same use case. Once we’ve done that, it’s much easier to personalize the messaging, the outreach, and the sales process to their needs and wants.”  

Essentially, instead of merely rating accounts according to how closely they meet your ICP definitions (traditional ABM), you’re also grouping them in such a manner that you can engage them in a distinct, segment-specific way. The best part is that you probably don’t need to go out and purchase any new data to do this effectively: You can take a big step forward using the same firmographic, environmental, and behavioral data you already collect to identify ICP accounts. In Flatfile’s case, for example, the segments are defined entirely based on vertical and company size.

If the who is the most important part of the pipeline equation, the when is second, and real breakthroughs come from identifying their overlap. Frankly, if you nail these two, you’re likely to succeed even if you don’t perfect your what and how.

Successful teams optimize timing by capturing signals automatically and at scale. Signals come in several flavors (see our taxonomy for a complete list), but the two most common — and most useful — are first-party engagement data and point-in-time triggers.

Engagement data tracks prospects as they interact with your owned digital properties (e.g., when they visit your website or engage with your LinkedIn profile). Taking action on these signals couldn’t be easier: If a buyer from a target account is clicking on your site, you should reach out to them (obviously). Similarly, if a buyer “likes” your founder’s LinkedIn post, your founder should shoot them a DM to set up a meeting (note: this play converts ~40% of the time for companies I’ve spoken with). None of this is particularly revolutionary; we’ve long had the ability to track engagement.

What is fairly new is the ability to easily capture and take action on point-in-time triggers. Triggers track key business events that drive timely buying behavior. These will likely be totally unique to your product. For example, a warehouse robotics company in Scale’s portfolio tracks job ads for night-shift workers to understand which distribution centers are seeing labor shortages. The ad postings are triggers that demonstrate when their buyer is most likely to buy.

You may already know that certain triggers are particularly useful for your business. Now, with AI, you have the ability to capture them at scale without effort. 

If you’re on the other end of the spectrum, you may be wondering how to identify the specific triggers from among the thousands you could feasibly collect that are actually predictive of buying behavior. My recommendation is simple: Work backwards from your most recent “closed:won” deals. Ask your new customers about the company and industry trends that motivated their purchase, and try to identify patterns. Or, do as Jordan Crawford (the OG GTM Engineer and founder of Blueprint GTM) does: upload CRM data from recent deals into your AI model of choice, and ask it to research and identify patterns.

There are two parts to any prospecting message: the context (why you’re reaching out to this prospect at this time) and the offer (what the prospect should do next). 

When you’ve nailed the who with microsegmentation and the when with signal data, this part is easy. You already know each segment’s relevant problems and when they’re most painful; at that point, the context practically writes itself.

As Jason Vargas from RevShoppe puts it, “the signal becomes the message.” In the case of the robotics company referred to above, the job posting doesn’t just guide when they reach out. It also tells them what to say. In those cases, they begin their message by demonstrating knowledge of the labor-shortage pain, and include relevant content on running 24-hour automation in warehouses.

Context is only half the battle, though: You also have to make the right offer. Years ago, Lars Nilsson’s team at Cloudera was struggling to book meetings with prospects. Ultimately, they realized they were targeting the right accounts with the wrong offer. Once they switched from offering demos to offering 1:1 conversations with their CTO on Hadoop use cases, everything changed.

Here are the offers that work best today:

Okay. Now that you’ve figured out who to target, when to target them, and with what messaging, all that’s left to do is decide how to reach to them. 

I know some people say that this is the most difficult part: “Email doesn’t work anymore,” “No one answers calls,” etc. While it’s true that certain channels are much less effective today than they were five years ago, I’ve been hearing some version of this same problem for 30 years, and the answer has always been straightforward: 

A decade ago, during my TOPO days, I found that SDRs who used phone, email, and LinkedIn had vastly higher connect rates than those who used a single channel. Today’s successful teams use an analogous, multi-channel approach. 

Here’s how to nail each one (with one caveat: all of this is so much easier if you have a good brand!):

Pipeline fundamentals haven’t changed. The teams winning today are doing the same things we’ve always sought to do: identify the right people, figure out when to reach them, craft the best message we can, and transmit it in the right ways. The only difference between them and the rest of us is that they’re iterating and adapting faster to a changing environment. Luckily, that means that if you’re not in the lead group, you’ve got every chance of catching up. I hope this article helps you do so.

Paul Adams, CPO at Intercom, breaks down why B2B companies are failing at AI transformation. And it’s not because of models or tooling. It’s because leaders avoid the decisions required to make impactful change. Intercom went all in, and the results are hard to ignore. Read the full piece and share where you think most companies are still playing it safe.

OnlyCFO argues that SaaS isn’t dead. But the playbooks are. As AI crushes valuations, survival now depends on tough calls: retaining top talent, expanding roadmaps, rebuilding moats, and reimagining products from an AI-first perspective. Leading through this? Worth the read. 

Kyle Poyar and Maja Voje pulled together 40 real-world AI plays from GTM teams that are actually seeing results. The winners are wiring internal context directly into AI to turn insight into action across content, prospecting, and sales. If you’re trying to move past “AI slop” and into AI-led leverage, this is a dense piece on what's working now.

Become a Topline insider by joining our Slack channel.

We want to hear your feedback! Let us know your thoughts.