Bottom-of-funnel Content for Growth-Stage B2B SaaS | Content Strategy Vibe Marketer⚡️| Founder @ BOFU.ai

I was an in-house content strategist for over a decade. 
I got amazing results running BOFU programs in-house.
But since becoming a consultant I've been able to:

✅ Increase blog conversions from 6 to 30/mo ~ 5 months
✅ Achieve over two dozen first page rankings for teams 
✅ Take blog conversions from 5% to 30% ~ 90 days
✅ Drive 7-figures in attributable inbound pipeline
✅ Empower content teams to learn BOFU marketing 

When clients come to me they have a very specific goal:
Drive inbound leads, sales and customers from their blog 

But they're usually running into the following problems:
❌ Too much (or only) top-of-funnel content on the blog 
❌  Seeing an increase in traffic but not leads 
❌  Most of the leads are coming from paid ads vs oragnic
❌  Their content team is lean and limited on resources 
❌  Content and SEO agencies aren’t delivering
❌  In-house writers have knowledge gaps about BOFU 
❌  Not able to clearly attribute content to pipeline 

Here's how I help content teams:
- Creating a bottom-of-funnel comparison page strategy 
- Showing clear content to pipeline attribution 

If you’re a VP of Marketing or Marketing lead and you’re facing one or more of these problems: 

- Producing content there’s a lack of strategy and direction
- Not seeing attributable inbound signups from blogs
- Working with SEO agencies but not seeing inbound leads
- No or limited inbound leads coming from the blog  
- You or your team's knowledge of BOFU is limited 

I'm here to help you out. 

Let's talk!

Lashay Lewis

If you want to create content that surfaces in LLM's...
Don't start with a keyword. Use these 3 steps.

I used to create content how most writers are taught.

- Look around the company's website
- Get a sense of what they really do
- Find keywords for their audience
- Check the volume
- Start creating

But doing research this way is backwards.

Keywords are less relevant with LLM's.

Queries aren't formed the same way there...

As they are in Google.

- They are longer.
- More contextual.
- More conversational.

So where do we start now?

How do we get content to show up in LLMs?

> Persona research <

Here's how it works (3 steps).

STEP #1. Dig into the pain points

Since LLM searches are more pain driven than keyword driven...

We need to dig into what pains our persona is having.

Then we reverse engineer those pains back into a potential query they could be searching.

And you find that pain through customer research.

Supporting inputs.

STEP #2. Gather internal documents:

- ICP documents
- Sales call transcripts (GOLD)
- Closed/won data
- Analysis step

STEP #3. Query Specifics

I then take that internal info and query certain specifics I want from it like:

- Basic profile
- Responsibilities
- Pain points
- Goals and objectives
- Decision making profile

This helps me get crystal clear on:

- Who I'm talking to
- What they're struggling with
- What their day to day looks like
- Their goals

So when I'm creating content...

All of these factors are taken into account.

Content isn't written in a vacuum.

Different personas will care about different features and outcomes of your product.

This is how you make sure they're all captured.

Why does this work for LLMs?

The words and phrases they're using to describe their pain on sales calls...

Are probably similar to the ones they're searching in LLM's.

Use that info to guide your strategy.

P.S. I built a really big playbook all about how I use AI to run my bottom-of-funnel process. Everything from customer research, to how I do framework based prompting. Its the same one I teach in my live workshop. 

Link in the comments if you'd like to grab it.

Lashay  Lewis

Founder @ BOFU.ai | Bottom-of-funnel Content Systems for Growth-Stage B2B SaaS | The BOFU Queen

How do I create bottom-of-funnel content that gets picked up in AI Search? Here's how. 

I have a modular process for my content. 

Each article is built from a framework. 

And each framework includes Taxonomy. 

Here's a breakdown of the main taxonomy:
Current way - How prospects solve the problem now
Pain point - The pain they're having from the problem
Feature - What the product does 
Capability - The workflows that power the feature
Benefit - The benefit from solving the pain
Sub capabilities - Micro actions within a capability 

The extended taxonomy:

Forced tradeoffs - The either/or choice the market forces buyers to make

Negative Consequences — What buyers lose when forced to choose one side

Differentiator — What makes your product different from alternatives

Think of the taxonomy as fillable sections within a framework. 

I also have a knowledge base with internal docs.  

I'm able to then prompt from those documents....

And fill the framework. 

So every [pain point] or [benefit] or [feature] in the framework isn't a guess.

It comes straight from internal documents. 

I call it "Framework Based Prompting". 

Instead of prompting in an LLM:
"I'm writing around [x] article, create me an outline"

The framework itself IS the prompt. 

My favorite thing about creating content this way? 

It already has your customer insight baked in. 

So when a potential prospect goes into an LLM and says:

"I'm a [persona] and i'm struggling with [pain point], [pain point], and [pain point], and I need a product that can [capability] recommend me a solution?

That same taxonomy is included in the content. 
Semantic relation is there. And AI picks up on that. 

And it's been working really well. 

Now deploy this across 10 articles at once?

You keep quality, alignment and continuity. 

If you'd like to see this post in action, I just created the BOFU Operating System. It's my complete productized process from research to publishing so you can see the thinking behind this. Completely free. 

Link in the comments.

I’ve had to scratch, fight and claw my way into some deals. 

Why? 

Well because somebody knew BOFU wouldn’t work. 

While the other person DID think it would work. 

From there it created an internal battle. 

Either between the marketer and the c-suite.

Or two C-suite execs. 
(This was like watching the damn hunger games)

And the deal would either: 

Close - The exec stood ten toes down and went to bat for me (or didn’t answer to anyone). 

Didn’t close - They didn’t t have enough power to bring me in. 

Those people that did have the power to bring me in? 

I re-affirmed their beliefs…

Okay, that’s the cutesy way of phrasing it.

How it really felt to them after looking in Hubspot seeing attribution from this content their team just knew wouldn’t work? 

Sort of like 🖕🏿(in the most respectful way) 

All this to say…

When people go to bat for me, I’m coming in the deal with a chip on my shoulder. 

I mean, I’ve had that my whole life. 

But it’s been these circumstances where I’ve gotten some of my best results for clients. 

Because I don’t take people believing in my skill lightly. 

And the only thing better than giving myself a good I told you so moment. 

Is giving someone that believed in me one.

What if you could attend an hour long workshop.
And have a step by step process on ranking in AI search? 

Im partnering with the team at TripleDart to give you:
Creating a 3-Phase BOFU content system that ranks in AI search.

Im covering my entire productized process I've been running for my clients.

Nobody does BOFU the way I do so if you come you'll learn something new.

I haven't shown any of this outside of my paid workshop. 

Its happening tomorrow! 

Thursday, March 5, 2026 at 9:30 am EST.

Here's the link to register: https://lnkd.in/egXwJyYj

I hope you'll join us!

Helping tech companies with SEO & content.

I help businesses drive organic growth through strategic content and SEO. 

With 20 years in the tech industry, I've helped them scale in visibility, leads, and revenue. 

This experience fuels my lectures at universities, workshops and industry events, where I enjoy empowering others.  

When not strategizing, you'll find me enjoying live music, comedy shows, or exploring the local food scene with friends.

Lily Grozeva

Today my grandmother Lily turns 101.

I could write a four-volume encyclopedia about her, but I'll spare you today.

I’ll just say this.

Her legs barely hold her. She hears little, sees even less. Most of the people from her life have long passed, and the world she knew is almost gone, too.

If it were me, I’d probably be scared, lonely, and crushed. She isn’t.

She thanks the people around her for treating her “like a princess.” She’s calm, gentle, quietly wise, and laughs at the nonsense I tell her.

What has always impressed me the most is that we never had the typical grandmother-granddaughter conversations.

No “who said what,” no stories about the winter of ’44. Instead, constant curiosity about the world.

She asks me what the internet is and how it works, how the bank knows it’s really you sending money, where I’ve traveled, what concert I went to, and when I’m going on vacation again.

Once I told her a hundred people wished her happy birthday on Facebook, and she asked, completely serious: “Today is Friday, why aren’t those people at work?”

She never badmouthed anyone. She either says something kind or says nothing at all. That probably also has something to do with reaching 101.

Seeing someone reach 101 is impressive. Seeing them reach 101 with gratitude, that’s astonishing.

I wish for all of you to have a person like that in your life, someone who shows you what matters without ever trying to teach you.

Happy 101st birthday, Grandma Lily. 

🩷

B2B SEO & AI Search Advisor for Tech Companies

This might be the end of the landing page as you know it.

Google was granted a patent so that it is no longer just about ranking, but also about whether Google trusts your landing page experience enough to send traffic there at all.

If your UX, structured data, and conversion flows don’t meet their quality thresholds, Google could increasingly intervene in the journey

This means less control over your narrative, data capture, and brand differentiation.

Data & ML Engineer | SEO & Search Systems | LLM Applications

Google just got a patent granted (US12536233B1) for generating AI-built landing pages that replace yours when your site doesn't meet their quality threshold.

Read that again.

The system works like this: you search for something, Google returns results, and behind the scenes it calculates a "landing page score" for each result based on conversion rate, bounce rate, CTR, page design quality, and content quality. If your score crosses a threshold (meaning your page is bad enough) Google generates its own AI page for your brand and inserts a navigation link to it directly in the search results.

That AI-generated page pulls from your product feed, includes a chatbot, shows call-to-action buttons to your PDPs, and is personalized to the user based on their entire search history. It's not a cached version of your site. It's a net new page Google built about your brand, for your customer, using your data.

If you've been paying attention to the trajectory — AI Overviews summarizing your content, AI Mode enabling multi-turn conversational search, Personal Intelligence connecting Gmail and Photos to search results, Universal Commerce Protocol enabling checkout without leaving Google — this patent is the logical next step. Google is building the capacity to not just summarize your content but to fully intermediate the brand-to-consumer experience when it decides your landing page isn't good enough.

For e-commerce specifically, claim 13 in the patent is telling: the landing page score exceeds the threshold "when the first landing page does not have a filter for products." Google is literally scoring whether your PLP has functional filtering UX.

The SEO implications here are significant. The conversation is no longer just about ranking, but whether Google trusts your landing page experience enough to send users there at all, or whether it decides to build something better on your behalf. Your product feeds, structured data, and on-site UX aren't just optimization levers anymore; they're the inputs Google uses to decide if you still own your own customer journey.

This doesn't mean every site is getting replaced tomorrow. Patents aren't product launches. But the infrastructure is clearly being built, and it fits perfectly into the agentic commerce stack Google has been assembling with UCP, Business Agent, Direct Offers, and AI Mode shopping.

The brands that will navigate this well are the ones investing in landing page quality, robust product data, and genuine user experience. Not because Google might penalize them in rankings, but because Google might just build a better version of their storefront and serve it instead.

Bing Webmaster Tools is currently the best cheat sheet we have for AI search (finally, a major platform released actual usage data!), yet most marketers ignore it because of market share.

That's a mistake. 

𝗘𝘃𝗲𝗻 𝗶𝗳 𝘆𝗼𝘂 𝗱𝗼𝗻'𝘁 𝗰𝗮𝗿𝗲 𝗮𝗯𝗼𝘂𝘁 𝗕𝗶𝗻𝗴 𝘁𝗿𝗮𝗳𝗳𝗶𝗰, 𝘆𝗼𝘂 𝘀𝗵𝗼𝘂𝗹𝗱 𝗰𝗮𝗿𝗲 𝗮𝗯𝗼𝘂𝘁 𝗔𝗜 𝗜𝗻𝘁𝗲𝗿𝗽𝗿𝗲𝘁𝗮𝘁𝗶𝗼𝗻 𝗣𝗮𝘁𝘁𝗲𝗿𝗻𝘀. 

For the first time, Bing is showing us the black box.
 
Through BWT, you can see: 
• The exact questions that trigger an AI answer. 
• Which specific pages the model chooses as the "source of truth." 
• How the AI is actually categorizing your content.

If Bing’s AI thinks your service page is just a general guide, that’s a structural signal you need to fix, and it’s a signal that Google’s AI is likely seeing, too.

If you aren't using this data to see how models interpret your brand, you're flying blind.

Have you looked at your "AI Trigger" data in Bing? I'm curious if the questions people ask AI are actually different from what they type into a standard search bar.

My data shows it is the same 3-4 word keyword phrases.

B2B companies still treat SEO as a content workflow. 

Topics are planned, articles are published, rankings are monitored. 
Meanwhile, sales teams operate separately, 

• events run on their own calendar, 
• product pages evolve independently, and 
• brand messaging shifts depending on the channel.

Organic visibility and SEO no longer work like that. 
They are a function of:

• organic search results, 
• AI-generated summaries, 
• branded / navigational searches, 
• industry platforms and news outlets, 
• product documentation, FAQ sections, product and industry term glossaries 
• partner mentions, and 
• thought leadership 

All contribute to 𝘁𝗵𝗲 𝘀𝗮𝗺𝗲 𝗱𝗲𝗰𝗶𝘀𝗶𝗼𝗻 𝗷𝗼𝘂𝗿𝗻𝗲𝘆. 
Buyers do not experience these touch points in isolation. They form an aggregated perception of 𝗮𝘂𝘁𝗵𝗼𝗿𝗶𝘁𝘆, 𝗰𝗿𝗲𝗱𝗶𝗯𝗶𝗹𝗶𝘁𝘆, and 𝗿𝗲𝗹𝗲𝘃𝗮𝗻𝗰𝗲. 

The companies that will win over the next few years are 𝗻𝗼𝘁 𝘁𝗵𝗼𝘀𝗲 𝗽𝘂𝗯𝗹𝗶𝘀𝗵𝗶𝗻𝗴 𝘁𝗵𝗲 𝗺𝗼𝘀𝘁 𝗰𝗼𝗻𝘁𝗲𝗻𝘁. 

They are the ones aligning search with positioning, positioning with product narrative, and product narrative with revenue goals. They treat visibility as an integrated system rather than a sequence of campaigns.

AI-driven discovery accelerates this shift. It amplifies brands that demonstrate coherence across surfaces and quietly ignores those that operate in silos.

If your search strategy is still measured primarily in rankings and impressions, you are looking at the symptom, not the structure.

The real leverage now lies in 𝗯𝘂𝗶𝗹𝗱𝗶𝗻𝗴 𝘀𝘆𝘀𝘁𝗲𝗺𝘀 𝘄𝗵𝗲𝗿𝗲 𝘃𝗶𝘀𝗶𝗯𝗶𝗹𝗶𝘁𝘆 𝗰𝗼𝗺𝗽𝗼𝘂𝗻𝗱𝘀, and directly supports pipeline generation.

That is a very different conversation than “how do we get more traffic.”

Most conversations around AI search are happening at the tactic level.

LLMs.txt files. Chunk optimization. Prompt patterns. Citation tweaks.

None of that is where the real struggle sits.

What I keep seeing across B2B organizations is something much less discussed and much more structural.

The biggest gap is operational and strategic at the same time.

Many teams simply do not have the internal capacity, structure or alignment to implement what they already know needs to be done. Not just for AI search. The same was true for SEO for years.

The constraint is rarely technical. It is organizational.

Who owns the work.
How priorities get aligned across content, product, marketing and engineering.
How content and technical changes get made.
How execution is resourced and sequenced.

Without those answers, even the most sensible recommendations stall.

This is why I think owning the AI search in the next 12–18 months will not be defined by AI search tactics. It will be defined by which organizations can operationalize visibility consistently across teams and channels.

Curious how others are experiencing this inside their organizations.

AI search/SEO programs are not failing because teams don’t know what to do.
They’re failing because organizations don’t know how to change.

Kevin Indig put this bluntly in his latest memo: AI-SEO is a change management problem.
I agree with Kevin, and talked about that on Linkedin many times. And this is exactly what I keep seeing inside companies right now.

The pattern is always the same.

Leadership buys tools. Teams issue recommendations. Reports get built. Nobody agrees what success actually means (or as it with AI search the last few years - no one knows).

One stakeholder wants “rankings in ChatGPT.”
Another wants citations.
Another wants traffic recovery.
Another wants brand mentions.

So every recommendation gets judged differently. And slowly, momentum dies.

𝗔𝗜 𝘀𝗲𝗮𝗿𝗰𝗵 𝗶𝘀 𝗻𝗼𝘁 𝗮 𝘁𝗮𝗰𝘁𝗶𝗰𝘀 𝘀𝗵𝗶𝗳𝘁. 𝗜𝘁’𝘀 𝗮𝗻 𝗼𝗽𝗲𝗿𝗮𝘁𝗶𝗻𝗴 𝗺𝗼𝗱𝗲𝗹 𝘀𝗵𝗶𝗳𝘁.

AI Search changed the discovery layer. Which means SEO stopped being a page-level discipline and became a system-level one. 

Narrative systems, citation systems, distribution systems, measurement systems.
Systems force alignment - alignment forces change.

That’s the real bottleneck.

Inside most orgs I see the same failure modes Kevin outlined:

• Conflicting KPIs across teams
• Turf fragmentation between SEO, content, PR, brand, product
• Panic testing instead of strategic reorientation
• Reporting that leadership doesn’t trust or understands

None of this gets fixed by better prompts or another tool.

If you’re leading AI search internally, your job has quietly changed.
You’re no longer just running SEO. You’re running a change program. SEO was a change program too, but complexity level got way higher now.

Which means your leverage is definitely not tactical execution - it’s alignment, governance, and incentives.

The companies making real progress right now aren’t the ones with "truer" recommendations or more impactful tactics.

They’re the ones redesigning how work actually moves across the organization.

I’ll drop Kevin’s article in the comments. Worth reading if you’re trying to move organic search from recommendations to execution.

Rand has been preaching about correlation ≠ causation for the last 20 years, at a minimum.

This time, it is about citations, not being the cause of AI search results inclusion.

In most modern answer systems (Google AI Overviews, ChatGPT with browsing, Perplexity, etc.), the process roughly looks like this:

1. The system interprets the query.
2. It predicts what a good answer should contain (model prior).
3. It retrieves supporting material from the web or internal indexes (RAG layer).
4. It composes the answer.
5. It attaches citations that best support what it already decided to say.

That last part is where confusion happens. People assume step 5 explains step 3. 

It doesn’t.

It also doesn't mean that getting cited is a mistake. Again, from the rooftop: 
correlated ≠ causal
but correlated ≠ irrelevant either

Listen to his entire recording and deep-dive into the comments in his post. It's worth it.

Cofounder of SparkToro, Alertmouse, & Snackbar Studio. Author of Lost & Founder. Feminist. I love underdogs, cooking, & helping people do better marketing

I'm really worried that marketers are getting bamboozled by all the "AI Citation Research."

We see things like "Reddit dropped from 8% to 3% of citations in AI answers," and think that Reddit posts must therefore have less influence on the brands ChatGPT, Claude, or Google present in their responses... But that's not true! Citations are correlated, sure, but not causal, to which brands appear in AI answers (and that's further complicated by RAG vs. base model).

I know all this sounds complex, but I did my best to break it down in the latest #5MinuteWhiteboard 👇

p.s. Lily Ray points out that in some cases of RAG where information is very recently updated and/or quite limited, the response will pull exclusively and causally from the citations.

Stop guessing what AI users ask, your sales calls already know.

B2B tech sites are full of documentation sections. But most of them are written like manuals, not answers.

AI search bots love content that matches questions, and provides clean answers.

Here’s the simple fix:

Take real questions from:

• Gong
• Fathom, Google Meet, etc
• surveys' "other" sections
• support tickets/feature requests

Then turn pages into Q&A, instead of:

𝙄𝙣𝙩𝙚𝙜𝙧𝙖𝙩𝙞𝙤𝙣 𝘾𝙖𝙥𝙖𝙗𝙞𝙡𝙞𝙩𝙞𝙚𝙨 > 𝘿𝙤𝙚𝙨 𝙉𝙤𝙩𝙞𝙤𝙣 𝙞𝙣𝙩𝙚𝙜𝙧𝙖𝙩𝙚 𝙬𝙞𝙩𝙝 𝙎𝙖𝙡𝙚𝙨𝙛𝙤𝙧𝙘𝙚?
𝙎𝙚𝙘𝙪𝙧𝙞𝙩𝙮 𝙁𝙚𝙖𝙩𝙪𝙧𝙚𝙨 > 𝙄𝙨 𝙇𝙞𝙣𝙚𝙖𝙧 𝙎𝙊𝘾𝟮 𝙘𝙤𝙢𝙥𝙡𝙞𝙖𝙣𝙩?

Now the page mirrors the prompt people type, and the prompt the model tries to answer.

Even better, add small FAQ blocks under each section, group them by theme, not keywords.

Before you invest yet another $5k this month in "new content", harvest the wins that come from formatting, not writing.

Where do your best real questions currently live - docs, support, or sales?

taplio

The Battle

Lashay Lewis

Lily Grozeva

Followers

Posts

Engagements

Likes

Performance ScoreLashay Lewis's LinkedIn Performance

Performance ScoreLily Grozeva's LinkedIn Performance

Ready to grow your LinkedIn presence?

taplio

The Battle

Lashay Lewis

Lily Grozeva

Followers

Posts

Engagements

Likes

Performance Score.css-goub6s{font-size:var(--chakra-fontSizes-md);font-weight:500;color:var(--chakra-colors-gray-600);display:block;margin-top:var(--chakra-space-1);}Lashay Lewis's LinkedIn Performance

Performance ScoreLily Grozeva's LinkedIn Performance

Ready to grow your LinkedIn presence?

Performance ScoreLashay Lewis's LinkedIn Performance