Chris at RevDog

A paid click can look like low intent when it’s actually a bad handoff. The ad got their attention. The page left the key question unanswered. Before you rewrite the campaign, find that question and answer it on the page. revdog.ai/why-paid-traff…

@JemBourouhDE That’s brutal. Are you checking the actual Google Ads click URL across mobile, consent, and redirect states before launch?

Last 45D Google Ads have been as functional as an alcoholic 10-15% of the ads we upload are flagged as video unavailable. Despite being uploaded on YouTube and being available. 10-15% of the of the LPs launched are flagged as landing page unavailable. Despite being accessible. Fuck you Google, fix your policies

@benchgoogleads The useful answer is rarely a format-wide winner. Shorts and In-Stream can create different expectations before the click, so the next comparison is whether each visitor sees a page that continues that message. Do you break out post-click conversion by placement?

The useful part isn’t just identifying who is spending—it’s inferring which promise they are buying traffic for. Ad libraries show the angle; the next question is whether the destination resolves it. What signal tells you a competitor’s paid traffic is worth studying beyond spend?

Keyword-to-page match is necessary. It isn't the whole handoff. Two visitors can arrive from equally relevant ads but need different proof: one is checking fit, another risk. Treat the page as a decision-stage conversation. Compare it with a control before calling it a win.

@alexgroberman Paid and AI-search traffic are different acquisition paths, but the on-page job is similar: move a visitor from “this is relevant” to “this is credible for me.” Are you seeing materially different conversion paths—or buyer questions—by source?

Just my personal opinion here: SaaS companies relying entirely on paid acquisition + outbound sales are going to struggle in 2026 and beyond. Meanwhile, we're adding $30,000+ per month with high-intent traffic from Google, ChatGPT, Claude, Perplexity and Gemini. Want to know how your SaaS ranks across Google and AI search? Check here. It’s free: seo-stuff.com/free-audit Alright, let’s start at the top. In order to send you traffic + citations, Google and AI systems need to understand: What your product does Who it is for Why it should be trusted Most SaaS homepages make this unnecessarily difficult. “Transform your workflow.” “Unlock next-generation productivity.” “Empower your team to do more.” These phrases sound polished, but they tell buyers and search engines almost nothing. Compare that with: “Automated invoice reconciliation software for mid-sized accounting teams.” “Customer support analytics for B2B SaaS companies using Intercom.” “Employee scheduling software for multi-location restaurants.” Your homepage should immediately explain: What the product is Who uses it What problem it solves How it differs from alternatives What proof supports the claims If someone cannot understand your SaaS in five seconds, Google and AI systems are going to struggle too. Specific positioning makes everything else easier. If you're unclear how to implement this, try SEO Stuff (seo-stuff.com). Next, distribute authority properly. A lot of SaaS companies either build no backlinks or send nearly every backlink to the homepage. Your highest-value commercial pages should not sit several clicks deep with no links pointing toward them. The goal is to strengthen the pages that can create demos, trials and sales conversations. Now let’s talk technical SEO, which is also key for AI search. A messy SaaS website makes every other part of the strategy harder. It needs to be fast, well-organized and properly set up. This becomes especially important when your site contains feature pages, integrations, industry pages, comparisons, documentation and programmatic landing pages. Without a clear structure, those pages compete with each other or remain practically invisible. Next, map keywords to the funnel. Broad keywords often look attractive because they have high search volume. They are not always where the revenue is. A project management platform may get more value from searches like: “Best project management software for construction teams” “Project management software with client portals” “Monday dot com alternative for agencies” “Software for tracking marketing deliverables” “Best Asana alternative for small teams” “[Your Brand] vs [Competitor]” These searches reveal the buyer’s problem, audience, preferred feature or current alternative. Look for: Clear commercial intent Difficulty below 30 where possible Strong CPC Natural-language phrasing A specific audience or use case Then map each search to the right page. Category searches go to product or solution pages. Feature searches go to feature pages. Industry searches go to industry pages. Comparison searches go to comparison pages. Educational searches go to guides, templates or calculators. Build one strong page for each meaningful buyer need. Next, improve your feature and product pages. Most SaaS feature pages contain a headline, screenshot, three vague benefits and a demo button. For obvious reasons, that is rarely enough. A strong page should include: A clear explanation of the feature The problem it solves Who benefits from it Screenshots or demonstrations Relevant use cases and integrations FAQs Trust signals A clear next step Your headings should reflect how buyers talk. “Work Smarter” tells people nothing. “Automatically Reconcile Invoices Across Multiple Accounts” is much clearer. You also need to connect features to outcomes. As in, literally explain what they accomplish. Automated reports reduce manual reporting time. Custom dashboards help executives monitor the metrics that matter. Real-time alerts let teams respond before a problem affects customers or revenue. Buyers need to understand the result, not just the feature. Next, create content around purchase intent. A random blog post every week is not a strategy. Build clusters around the problems your product solves. Useful SaaS content formats include: How-to guides Templates Calculators Original research Industry benchmarks Comparison pages Alternative pages Integration tutorials Case studies The best topics sit close to the product. Content that attracts traffic but has no logical connection to the software is unlikely to generate much pipeline. Every page should naturally lead toward a feature, solution, integration, trial or demo. Now let’s talk about comparison and alternative pages since these are often some of the most valuable pages a SaaS company can create. Examples: “Best CRM software for small agencies” “HubSpot alternatives for growing SaaS companies” “[Your Brand] vs Salesforce” “Best customer analytics tools in 2026” “Intercom vs Zendesk for B2B SaaS” People searching these terms are already evaluating products. Structure these pages with: A direct answer A comparison table Pricing information where available Feature differences Best-fit use cases Limitations Customer proof A final verdict Also, be honest about who each product is best for. A believable comparison will convert better than a page claiming your platform wins every category. Next, create industry and use-case pages. “Software for everyone” is rarely convincing. A finance team, healthcare company and marketing agency may use the same product for completely different reasons. Give each important audience a dedicated page. Examples: “Reporting Software for SaaS Finance Teams” “Scheduling Software for Multi-Location Restaurants” “Customer Analytics for Subscription Businesses” “Workflow Automation for Accounting Firms” Include the audience’s specific problems, relevant features, integrations, proof and compliance requirements. This gives buyers, Google and AI systems a much clearer understanding of where your product belongs. Next, turn customer proof into permanent assets. Many SaaS companies have their best proof trapped inside sales decks, recorded calls, Slack messages and internal reports. Turn it into: Detailed case studies Industry-specific customer stories Before-and-after metrics Video testimonials with transcripts Implementation stories ROI breakdowns Do not publish a case study saying the customer “improved efficiency.” Show what changed: Time saved Revenue added Costs reduced Conversion rates improved Churn reduced Hours eliminated Specific outcomes make the story more useful and believable. You also need authority outside your own website. Useful plays include: Getting included in software roundups Publishing original industry data Appearing on podcasts with transcripts Getting founders quoted in relevant publications Building integration partnerships Sponsoring relevant events Earning links from sites already ranking for your target searches A relevant mention in a respected industry publication can be more valuable than a generic link from a larger but unrelated website. Finally, measure what creates revenue. Traffic alone is not the goal. Track: Marketing-qualified leads Sales-qualified leads Demo requests Trial signups Pipeline created Sales conversations Branded search growth AI citations and brand mentions Conversion rates Revenue influenced A comparison page attracting 300 qualified visitors may create more revenue than a broad guide attracting 20,000 readers. Connect your content and landing pages to CRM data. You need to know which pages create opportunities, which opportunities close and how much revenue those customers generate. Most SaaS companies still: Use vague positioning Send every backlink to the homepage Publish disconnected blog posts Have thin feature pages Ignore comparison searches Leave customer proof buried internally Measure traffic without connecting it to pipeline That is why so many are invisible when buyers ask Google, ChatGPT, Claude, Perplexity or Gemini which software they should use. The winning system is straightforward: Clear positioning Strong technical foundations Commercial keyword mapping Detailed feature and solution pages Intent-driven content Customer proof Third-party authority Revenue-focused analytics Google needs enough evidence to rank the product. AI systems need enough evidence to recommend it. Buyers need enough evidence to book a demo or start a trial. When those signals align, search becomes a predictable source of pipeline. The SEO Stuff Done-For-You package combines backlinks, keyword research and AI-optimized content in one package: seo-stuff.com/gold-plan-pack… And you can see how your SaaS currently ranks across Google and AI search here. It’s free: seo-stuff.com/free-audit

@keean_edward Google Ads will surface a useful distinction: some searchers want validation the product exists; others want proof it will work for them. Watch which question remains unanswered after the click before broadening spend. What’s the first job you want the landing page to do?

my SAAS side project just hit $1000! 🎉 here’s my plan to scale to $1000/mrr over the next month: so far i’ve spent $0 on marketing and mostly relied on some reddit posts i made a few months back. i think that’s mostly dried up so i plan to do more of that + google ads targeting my competitors. the biggest problem: my product is far more complex than it was back then. i need to make onboarding more clear and efficient. i’ve noticed most cancelations come within just an hour of a customer signing up - if they make it past that point they usually convert. i’m also changing my free trial from 10 days to just 3 or 4 for this exact reason. i do still have a free version of the product that i’m thinking about removing too, but unsure.

@itsivanfalco The shift isn't just faster media buying—it's making the learning loop auditable. If an agent changes targeting, creative and landing-page inputs together, attribution can flatter every change. A fixed control or staggered rollout makes the decision cleaner.

@codyschneider Agreed—the handoff matters more as targeting gets better. “Same pain point” can still hide different stages: a new-category searcher needs orientation; an alternative seeker needs proof of difference. Do you vary the page’s first proof point by intent, or keep it unified?

this feels illegal to know facebook ads are the best performing channel for B2B right now andromeda is so good at finding your target customer if the ad and the landing page speak to the pain points / outcomes of your target customer it will find the right person then the conversion event is a server side conversion api that auths the company

An average conversion rate is a weak map for paid traffic. A visitor who clicked an ad about speed is not asking the same question as one who clicked an ad about risk. Segment results by ad angle, source, and intent. That’s how you find the handoff that actually improves conversion—and the one worth proving with a holdout.

@cleancommit This is the healthy correction in experimentation: separate observed revenue from incremental revenue. A projection can be directionally useful, but a holdout is what tells a team whether to scale the change.

I used to love writing about how we made our clients $150,000+ in additional revenue from one test. As far as I knew, this was true. But as I've read deeper into revenue projections from people who know more than I do, I realise that some of these numbers were just wrong. My calculations were based on the revenue gain numbers in @shoplift and @intelligems. If the winning experiment ran for a month, hit significance and reported a revenue gain of $20,000, I'd simply multiply this by 12 months in the year and declare that I'd added an extra $240,000. But every store we worked with has ups and downs to their performance. Maybe we had a big win during an "up" month and the rest of the year would be down. Or maybe the effect of the changes would diminish over time. There's too many variables to draw a clear correlation. Here's a good example. One of our client's Meta account was compromised which cost them months of algorithm optimization. We'd recently run a great experiment for them that increased their PDP conversion rate by 22%. According to the metrics, this would have been over $300k in additional revenue. But their Shopify account told a very different story. Their sales dropped by more than $300k within 6 months. That's why these days I'm more reserved about making big dollar amount claims. To be safe, I'll usually just declare how much revenue was made by the experiment during the time it was running, because that's the only number we have confidence in... even if it doesn't look as good on social media.

@JacobRushfinn Segmentation is how tests move from verdicts to learning. For paid traffic, the most useful segments often start upstream: ad angle, source, and visitor intent. A global average can hide the exact handoff that works.

You run an A/B test, but you're missing 90% of the learnings. Everyone focuses on sample size and a well-formed hypothesis That's easy In my 12+ years running experiments for consumer apps, only the top teams do this consistently. SEGMENT YOUR EXPERIMENT RESULTS! Your user base is not all one type of person. They don't all have the same conversion rate. They don't all behave the same. You may look at the experiment data, and the results may be inconclusive. But if you dig a bit deeper, there are going to be different pockets of users that responded differently to your test. This is where the gold is. This is how you actually start to understand your audience and the different user groups and personas within it. It takes more work than just calculating whether the experiment was stat sig or not, but it's where the majority of the value and learnings are. Want more insights like this? Go get the 48 Laws of Subscription App Success

@IsmailLate52662 Yes—the destination should recognize why the visitor clicked. The goal isn’t merely fewer links; it’s carrying the ad’s promise forward and answering the next question. Then validate the lift, not just fewer bounces.

A website is an online brochure. A funnel is a high-speed machine that turns strangers into customers. Most Shopify brands spend all their time making their homepages look pretty. But here is the truth: Your homepage is for brand awareness. Your landing page is for revenue. When you run ads, you shouldn't be sending traffic to your homepage. You should be sending them to a focused, distraction-free landing page designed for one specific outcome: The Sale. No navigation menu to distract them. No links to your "About Us" page. Just the offer, the proof, and the checkout. If you’re sending paid ads to your homepage, you’re paying to let your customers get distracted. Stop "designing websites." Start building funnels.

Most marketing “wins” are attribution claims. Someone converted after seeing the ad. Someone converted after the landing-page change. But the decision-grade question is harder: Did more comparable visitors convert because of it? Attribution tells you who got credit. A holdout tells you what created incremental lift.

@PaulKoshlap “What did this cause?” is the operating question. It applies just as much to a landing-page change as a channel decision: a better-looking page is not a win unless a holdout shows it changed conversion behavior.

Best marketing tool of all time? The scientific method. Why? It’s the only way to tell what's actually true from what just looks true. And in marketing, that's the difference between real growth and lighting money on fire. Marketing’s most expensive mistakes aren't people believing obvious nonsense. They're people believing real platform data that’s reasonable, well-dressed, and dead wrong. The campaign that "drove" sales when it was really just standing next to them. The channel that looks cheap because it's quietly taking credit for demand you already paid to create. The “winning” ad with the minuscule sample size. None of those look like errors. They look like results. That's exactly what makes them dangerous. The scientist's toolkit is a set of defenses against being convinced by things that aren't real: A control group, so you can see causation, not just correlation. Statistical significance, so you don't bet the budget on noise. Causal inference when running an experiment isn’t an option. Don’t get me wrong, that platform reporting — that's data. Real, useful data. But it's missing the one thing that turns data into proof: isolating the variable. Hold everything else still, ask "what did this specific thing actually cause," and you're measuring impact. Skip that, and you're just measuring what happened nearby and hoping it's the same thing. That's the gap the scientific method closes; it’s the gap incrementality testing exists to fill. The marketers who solve this problem aren't smarter. They've just trained themselves to ask one question before they spend, scale, or celebrate: How do I know this is real, and not just convincing? Learn to answer that, and you stop making the most expensive mistakes in marketing.

@johniosifov Exactly. The useful unit of measurement isn’t “did a conversion happen after this?” It’s “did more comparable visitors convert because of this?” That distinction changes what you ship, scale, and stop.

Marketing teams are adopting AI faster than they're measuring it. 95% of enterprise marketing organizations have some form of AI automation. Only 41% can prove ROI — down from 49% two years ago. Let that math sit: adoption tripled while proof of value dropped. I've seen this pattern in every major technology wave. ERP in the 2000s. Social media in the 2010s. AI now. The adoption curve and the measurement curve almost never overlap. Companies buy before they know what success looks like. The top quartile — the 41% who can prove ROI — share one characteristic: they defined measurement architecture before deployment. Not KPIs after the fact. Not "we'll figure out attribution later." They mapped causal chains from AI action to business outcome before they turned on the tools. The specific approach varies. Holdout groups. Incrementality testing. Controlled rollouts by segment or channel. But the underlying discipline is the same: isolate the effect of the AI, don't just measure what happened in the period after you installed it. Forrester data: top quartile marketing teams running AI with proper measurement frameworks generate $8.71 per dollar invested. Bottom quartile, same AI tools, no measurement architecture: $1.24. Same tools. 7x different outcome. The AI is not the variable. The measurement discipline is. If you're deploying AI in marketing and you don't have a pre-defined holdout control group, you don't know if it's working. You have correlation. You need causation. The good news: this is fixable. The measurement architecture takes 2-4 weeks to set up. The AI implementation often takes 6-12 months. Start with measurement. Let that dictate what you deploy and how you roll it out.

@ChoyRb Retargeting is especially valuable when it preserves the original intent instead of simply buying another impression. When someone returns, the destination should pick up the question that got them there—not make them start over.

Most paid teams optimize the ad and the landing page as separate assets. But conversion happens between them. The ad creates a question. The destination needs to answer it. Then a holdout tells you whether that answer created real lift.

Most paid-acquisition teams treat the click as an outcome. It’s the beginning of the experiment. A better question than “did this page convert?”: Did it answer the reason this visitor clicked—and did it create incremental conversions? That shifts CRO from prettier pages to a measurable handoff.

@stepanecom Quiz funnels can earn their place when they replace a generic page with a useful diagnosis. The important measurement is not completion rate alone—it’s whether the added interaction creates incremental downstream conversion.

Obvi running ALL of your ad creatives to one quiz funnel… And it reportedly has an 80% completion rate. That’s insane. Yet people still argue quizzes are “too hard to build” or “not worth it.” Meanwhile they’ll spend another $30k testing new creatives into the same mediocre PDP… Then complain CAC keeps climbing lmaoooo. I’m not saying every supplement brand should send 100% of its traffic through a quiz. That’s dumb too. But look at what Obvi is actually doing. They sell products across weight loss, beauty, collagen, and daily wellness. There’s overlap between the products, benefits, and customer goals. Quiz decides: - Which product fits - Whether to recommend one SKU or a stack - Which benefit should lead - What customer should buy first So they can send 20 different ad angles into one funnel… that's a big leverage. And based on our data, quiz funnels usually outperforms any landing page on AOV. Not testing one because it sounds like “too much work” is just lazy. Will a quiz automatically fix your funnel? No. But neither will launching your 48th UGC ad into the same PDP converting at 1.5% Test more. Build more.