Discovered Labs vs GrowthPlays: AEO-first agency vs content production agency for B2B SaaS

TL;DR

• Traditional keyword rankings correlate poorly with LLM citation rates. Our research on what drives AI citations shows that information consistency and source freshness matter more than ranking position for LLM retrieval.
• Discovered Labs is an AEO-first agency for B2B SaaS that tracks actual citation rates across ChatGPT, Claude, Perplexity, Google AI Overviews, and Gemini using proprietary infrastructure, not rank trackers.
• GrowthPlays is a content production agency for B2B SaaS that builds for search, AI Overviews, and social. Their published model doesn't describe how they measure AI citations, which engines they track, or any passage-level retrieval methodology.
• The CITABLE framework structures content specifically for Retrieval-Augmented Generation (RAG) systems, producing measurable results: one client went from 575 AI-referred trials to 3,500+ in 7 weeks.

Publishing high-volume content without engineering it for passage retrieval wastes marketing budget in the era of AI search. LLMs skip unstructured content regardless of editorial quality or keyword optimization. B2B SaaS marketing leaders are now evaluating two types of agencies: those that optimize for editorial volume and keyword rankings, and those that engineer content for AI retrieval and measure the citations directly. This guide compares Discovered Labs, an AEO-first agency, with GrowthPlays, a content production agency, across methodology, measurement infrastructure, and pipeline outcomes so you can choose the model that fits where your buyers are actually researching.

AEO vs content agencies: core differences

Answer Engine Optimization (AEO) and Generative Engine Optimization (GEO) involve structuring content so that LLMs retrieve, synthesize, and cite it in AI-generated answers. Traditional SEO optimizes for ranked document lists a human clicks through. AEO optimizes for semantic passage retrieval, where a model extracts a specific passage from your content and synthesizes it into a single answer. These are different technical problems.

As I covered in Is SEO the same as AEO?, both disciplines share the same foundations: positioning, Ideal Customer Profile (ICP) clarity, technical health, and off-page information consistency. But the retrieval technology diverges in enough tactical decisions to produce a meaningful competitive gap. That gap is what this comparison is about.

The "LLM Visibility Gap" describes what happens when a brand ranks well on Google but never appears when a buyer asks ChatGPT or Claude for vendor recommendations. It is not a branding problem or a content quality problem. It is a structural and measurement problem that traditional content agencies are not built to solve.

How AEO agencies verify AI citations

Verifying AI citations requires querying LLMs directly and parsing the responses. At Discovered Labs, our AI visibility tracker runs automated prompts across ChatGPT, Claude, Perplexity, Google AI Overviews, and Gemini on a regular cadence. Each prompt reflects a real buyer query from our client's query map, ensuring the citation rate we report matches what prospects actually ask. We then parse the responses to identify which source URLs are cited, how often, and how our client's share of voice compares to named competitors.

Content production agencies typically rely on tools like Ahrefs, SEMrush, or Search Console. These tools estimate organic search volume and track keyword positions. They cannot tell you whether a specific piece of content was retrieved by an LLM, which passages were extracted, or how often your brand appeared across AI-generated answers in a given week.

We documented a related problem in our analysis of AI tracking platform measurement flaws, showing how most AI visibility tools overstate precision because they test with a limited prompt set rather than representative buyer queries at scale. For a full breakdown of what this infrastructure actually requires, see our guide on AEO agency measurement infrastructure.

Why scale alone fails AI search

Publishing more content is not sufficient for AI search visibility. LLMs retrieve content using vector embeddings and semantic matching, not keyword frequency. Research on Dense Passage Retrieval (Karpukhin et al.) showed that dense retrieval models outperformed BM25 sparse retrieval by 9 to 19 points on top-20 passage retrieval, meaning the model selects passages based on semantic relevance and structural extractability, not volume.

The practical implication: if a section cannot be lifted cleanly as a standalone passage, LLMs typically won't cite it. LLMs skip content buried in long narrative blocks written for reader enjoyment rather than passage extraction, regardless of editorial quality. In our CITABLE framework research, blocks of 200 to 400 words leading with a direct answer consistently outperform long-form narrative in citation rate.

Google's AGREE research reinforces this from a different angle: LLMs reward consistent claims across independent sources. Off-page strategy therefore shifts from "acquire do-follow links" to "keep the same accurate claim about your product live across Reddit, industry publications, comparison content, and your own site."

Tracking real source attribution in LLMs

Share of Voice (SOV) in AI answers measures the percentage of AI-generated responses to a tracked query set that mention your brand, relative to total brand mentions across your competitive set. It is the primary AI visibility KPI because it contextualizes performance against competitors rather than measuring absolute mention volume.

Our tracker structures the output by pillar topic, sub-topic, and individual prompt, so teams can see where you appear and where competitors lead. Our guide on real citation rate benchmarks explains why platform-reported numbers often understate true visibility and how to calibrate expectations before presenting to a CFO or CEO.

Comparing AEO agencies: Discovered Labs methodology

Both Discovered Labs and traditional content production agencies produce content for B2B companies with search visibility as a stated goal. The comparison sharpens when you examine the underlying infrastructure, methodology, and what each agency can actually measure and prove. Discovered Labs is built as an AEO-first agency. Traditional content production agencies focus on editorial volume for traditional search and may incorporate AI search into their editorial model.

How Discovered Labs targets AI citations

Our approach is technical and systems-first. We map a client's priority buyer queries, run those queries across five AI engines, and identify which ones the client wins, which they lose, and where competitors dominate. From there, we build content using the CITABLE framework, designed specifically for RAG system retrieval.

Our full-time AI/ML engineering team, led by CTO Ben Moore, an ex-Stanford AI researcher who built self-driving car and fraud-detection systems at Stripe, Coinbase, and Brex, runs the tracking infrastructure. This is not a reseller relationship with a third-party tool. The tracker is ours, and the citation data it produces feeds directly into what content we recommend shipping next.

Off-page work closes the information consistency gap: the same accurate claim about the client maintained across Reddit, independent publications, G2, comparison content, and the client's own site. Our research on Reddit's influence on ChatGPT answers, drawn from 144,000 AI citations, found that Reddit occupied roughly 27% of ChatGPT's internal search slots during query processing despite appearing in only 0.35% of visible citations.

How GrowthPlays scales content output

GrowthPlays builds content engines that map product- and customer-relevant topics to a content plan, with a stated goal of earning visibility across traditional search, AI Overviews, and social. They also offer content revenue attribution, tying content to pipeline and closed-won deals through CRM integration.

Their strength is editorial execution at scale with workflows that reduce production time while maintaining a quality floor that many AI-only content tools don't reach. Where the model is thinner is measurement: their published materials describe the AI-visibility goal but not how they track it. No named engines, no citation rate, no passage-level retrieval methodology. Claiming AI visibility as an outcome and measuring it are different things, and only one of those is verifiable from the outside. We looked at this distinction in detail in our piece on claiming LLM visibility vs proving it.

Core differences: AEO vs content agencies

The core tension is pipeline attribution versus visibility assumption. GrowthPlays reports on traffic, rankings, and revenue attribution through CRM integration, useful for proving traditional-channel ROI, but none of it proves whether content is actually being retrieved by LLMs. AEO-first agencies tie content decisions to actual citation data, so marketing leaders can show a CFO exactly which prompts the brand wins, which it loses, and what changed between this week and last. For a deeper look at why these retrieval systems diverge technically, our piece on how LLM retrieval works for AI search walks through the full mechanics.

How we audit AI citation and answer rates

Auditing AI visibility starts with establishing a baseline: which buyer queries trigger AI responses in your category, how often your brand appears in those responses, and what competitive share of voice looks like across engines. Our AI visibility tracker paired with the AEO content evaluator produces two outputs: where you're invisible, and why specific content isn't being cited.

How Discovered Labs tracks AI citations

We configure the tracker during client onboarding, loading in priority buyer queries and your competitive set. From that point, real buyer prompts run regularly across ChatGPT, Claude, Perplexity, Google AI Overviews, and Gemini. Each prompt runs across every engine, so we can compare share of voice engine by engine and spot drift in one before it spreads.

We structure the output by pillar topic, sub-topic, and individual prompt, so teams see exactly which content performs and which gaps need filling next. For context on how this compares to self-serve tools, our buyer's guide to AI visibility platforms covers what Profound, Peec AI, and Otterly offer versus what a managed service adds. See also our Profound vs Peec AI review for a direct platform comparison.

Measuring actual AI citation impact

We track citation rates as the leading indicator and connect them to pipeline as the lagging indicator that justifies the investment. Our tracker identifies AI-referred traffic in client analytics to track trial sign-ups, demo requests, and pipeline contributions originating from AI-generated responses. Our analysis of what drives AI citations, based on 2 million citations and 10,000 analyzed pages, provides benchmark data on what citation rates to expect at different content maturity levels. For category-specific figures, see our B2B SaaS AI citation benchmarks.

The clearest proof of what this produces comes from an anonymous B2B SaaS client who went from 575 AI-referred trials to 3,500+ in 7 weeks after implementing the CITABLE framework and systematic citation building. That result does not appear in a keyword ranking report. It only becomes visible when you track the actual source of trial sign-ups from AI-generated answers.

Quantifying the AI visibility deficit

The opportunity cost calculation is straightforward once you have baseline data. Map your 50 highest-priority buyer queries and run them across ChatGPT, Claude, and Perplexity. Count how many responses mention your brand. If your brand appears in 5 out of 50, you have 45 active gaps where competitors are being recommended instead of you. Prioritize those gaps by pipeline value, not search volume, because AI-referred buyers are typically further along in their evaluation than someone clicking a blog post from a Google results page. Our AI visibility tools vs tracking explainer covers how to distinguish between passive monitoring and active optimization services when deciding whether to self-serve or partner.

Why Discovered Labs leads in AEO performance

Our technical advantage comes from four things working together: the CITABLE framework for content structure, proprietary tracking infrastructure for measurement, original R&D that continuously updates our understanding of what drives citations, and a founding team with both AI engineering depth and B2B demand generation experience. The combination is the differentiator.

Engineering AI trust: CITABLE methodology

The CITABLE framework is a structured methodology that optimizes content specifically for RAG pipeline retrieval while keeping it readable for humans.

• Clear entity and structure: A 2 to 3 sentence Bottom Line Up Front (BLUF) opening that states the answer and names the product category explicitly.
• Intent architecture: Lead with the direct answer, then expand with evidence. Content must be quotable at the passage level.
• Third-party validation: Consistent signals across G2, Reddit, industry publications, and comparison content that LLMs cross-reference before citing.
• Answer grounding: Every claim links to a verifiable source. LLMs skip unsourced assertions in retrieval.
• Block-structured for RAG: 200 to 400 word sections, labeled clearly, in tables, ordered lists, or FAQ format. Each block answers one question independently.
• Latest and consistent: Timestamps signal freshness. Unified facts across all content sources prevent the contradictory data that causes LLMs to skip citing a brand.
• Entity graph and schema: We implement structured data with retrieval pipelines in mind, not just Googlebot.

You can score your existing content against the framework using our free AEO content evaluator. For teams implementing CITABLE workflows at scale, the Claude Code CITABLE framework guide shows how to automate extractability audits across a content library.

The incident.io case study describes the starting point: the team had been experimenting with homegrown LLM prompts but without a clear strategy for what to optimize or how to structure content for AI retrieval. CITABLE gave that process a repeatable structure.

Scaling daily content for AI retrieval

AI systems reward topical coverage. If buyers ask 50 questions about your product category and you've only published content addressing 10, you're invisible in 80% of AI-mediated research. Covering those gaps requires a higher publishing cadence than most traditional content agencies operate at. Our Establish retainer includes up to 20 SEO and AEO-optimized articles per month, and the Compete tier goes to 28, each one planned against a buyer query map and scored against CITABLE before publishing. Every piece goes through a content editor before publishing.

Validating pipeline impact with bench data

Our database of 2 million analyzed citations and 10,000 pages, detailed in our what drives AI citations research, gives us a continuously updated model of what retrieval patterns actually look like across different query types, industries, and content formats. That bench data means our recommendations reflect current retrieval patterns in production LLM systems rather than generic best practices from traditional search optimization.

Research on Retrieval-Augmented Generation (Lewis et al.) established that RAG fundamentally separates the retrieval step from the generation step, meaning the engineering decisions around passage structure and source credibility directly determine what gets cited. We build our content and off-page strategy around that technical reality.

Transparent terms for AI outcomes

All Discovered Labs retainers are month-to-month. If we stop delivering citation and pipeline results, you leave. That accountability structure is intentional. Public pricing, no annual lock-in, and weekly progress reporting on citation rates and share of voice mean the relationship stays aligned with outcomes rather than contract duration.

Why GrowthPlays leads in editorial volume

GrowthPlays has built a strong reputation for editorial content quality and positions its content engine around pipeline outcomes, not just traffic, with dedicated revenue-attribution tooling. Where they excel is high-volume, high-quality editorial execution across search and social.

Editorial volume and citation rates

GrowthPlays states AI visibility as a goal of its content engine, alongside traditional search and social. What isn't published is how that goal gets measured: no named AI engines, no citation rate tracked over time, no passage-level structure standard equivalent to CITABLE. Stating the goal and measuring progress against it are different capabilities.

Topical coverage does matter, and publishing well-written content is better than publishing nothing. However, without passage-level structure optimized for RAG retrieval, editorial quality alone underperforms purpose-built AEO content in LLM citation rates. Publishing volume without extractability structure leaves results to chance rather than engineering. If you're evaluating whether your current agency is equipped for AI search, our piece on 5 signs your content agency isn't built for AI search gives you a practical checklist.

Outdated SEO metrics for AI search

Traditional keyword rankings and AI citations are diverging. Our tracking of Ahrefs AI Overview data shows that top-10 rankers made up 76% of AI Overview citations in mid-2025 but only 38% by early 2026. LLMs pull from a broader semantic pool than Google's ranked list, and that gap is widening. For marketing leaders who need to prove AI visibility to a CEO or board, traffic growth and keyword ranking reports don't answer the core question: which AI-generated answers cited our brand this week? Our AI visibility tools guide covers the platforms that do track this directly.

Specialized AEO for B2B SaaS growth

Complex B2B SaaS products serve multiple personas and compete in categories where differentiation is often subtle. LLMs struggle to represent nuanced positioning without explicit entity mapping and structured data that defines relationships clearly. Technical products need precise schema implementation to help LLMs distinguish your product from competitors in synthesized answers.

Evaluating AEO versus content production partners

The right choice depends on what you're optimizing for and what you can actually measure. Our guide to choosing a content agency in 2026 covers the full evaluation framework if you're comparing multiple partners at once. If your primary goal is AI citation rate, AI share of voice, and pipeline attributed to AI-referred sources, you need a partner with the infrastructure to measure and optimize those specifically. If your primary goal is high-volume editorial coverage for traditional search with AI as a secondary consideration, a content production agency may fit.

When to partner with Discovered Labs

We're the right fit when:

• You're a B2B SaaS company with a complex product.
• Your marketing leader needs to show AI citation rate and pipeline impact, not just traffic growth.
• You want visibility data across ChatGPT, Claude, Perplexity, Google AI Overviews, and Gemini.
• You're willing to run month-to-month and share CRM attribution data so we can connect citations to pipeline.
• Traditional lead sources are plateauing and competitors appear in AI-generated vendor recommendations while you don't.

We're not the right fit if you need guaranteed citation positions in two to four weeks.

Ideal use cases for GrowthPlays

GrowthPlays fits well when:

• You want a high-volume editorial content engine with AI assistance for production efficiency.
• Your team wants revenue attribution through your existing CRM rather than a dedicated AI-citation tracking layer.
• You're comfortable with AI visibility as a stated goal of the engagement rather than something reported on with named engines and citation rates.

Measuring AI citation performance

Our guide on best AI visibility tools for SaaS covers how teams that want to self-serve parts of this measurement can use tools like Profound, Peec AI, and Scrunch, and where self-serve ends and managed service becomes necessary.

Quantifying your returns: AEO vs. content agency

Case study results and ROI metrics

The incident.io engagement provides clear before-and-after metrics. Incident.io competes with PagerDuty in incident response, a category where buyers heavily use AI to research and shortlist vendors. Within four months, organic meetings booked grew 22% and AI visibility climbed from 38% to 64%. One of the first articles we published became the most-cited page across all competitors in the category, as tracked in Profound.

"I have recommended you to multiple peer CMOs. There are large organizations like HubSpot and Ramp who have dedicated teams to work on large projects like AEO. For everyone else (except my competitors) there's Discovered Labs!" - Tom Wentworth, CMO at incident.io

For Sova Assessment, a skills assessment and testing platform, organic search became the number one pipeline channel, contributing more than 50% of pipeline. The anonymous B2B SaaS client went from 575 AI-referred trials to 3,500+ in 7 weeks. Full details are in our case studies.

When evaluating GrowthPlays or any content agency's results, ask for three specific proofs: citation rate across at least three AI engines, share of voice trend against named competitors, and pipeline contribution from AI-referred sources in CRM. Our full list of questions to ask any content agency about AI citations covers what to ask and what the answers should look like.

If you need to build the internal case for why those proofs matter, our B2B SaaS AEO proof requirements guide gives you the framework to present to a CFO or board. Traffic growth and keyword rankings measure traditional search performance. AI citation rates measure LLM visibility. Both matter, but they track different surfaces. As AI search takes a growing share of the buyer research journey, the gap between "we grew traffic" and "we got cited by ChatGPT when buyers asked for vendor recommendations" will widen in strategic importance.

Speed to AI citation: a 30-day outlook

With CITABLE-structured content and information consistency work running in parallel, initial citations typically begin appearing after publishing. Full optimization, where citation rate stabilizes and share of voice against key competitors becomes measurable, takes sustained effort over multiple months. That timeline reflects how LLM training data and real-time retrieval interact.

Measuring ROI: AEO vs content production

CFOs and CMOs evaluate agencies on measurable returns: cost per qualified lead, pipeline contribution, and conversion rate. Traditional content agencies report traffic growth and keyword rankings, which are inputs, not outcomes. We track AI citation rate, AI-referred pipeline, and share of voice against competitors, which connect directly to revenue. Here is how the models compare when you map them to what boards and leadership actually care about.

GrowthPlays AI citation tracking capabilities

GrowthPlays's published attribution model ties content to pipeline and closed-won revenue via CRM integration, which is a meaningful capability for proving traditional-channel ROI. What it doesn't include is AI citation tracking: querying LLMs directly, parsing which URLs get cited, or reporting share of voice by engine. So the revenue-attribution story works for search and social. It doesn't yet answer the AI-visibility question their own positioning raises.

How AEO extends traditional SEO rather than replacing it

As I covered in the full guide on SEO and AEO differences, the fundamentals are largely identical: technical health, clear positioning, off-page information consistency. But retrieval technology diverges enough to change tactical priorities. Companies optimizing only for Google rankings are losing AI visibility as the two systems diverge, and our Ahrefs tracking data shows that divergence is accelerating.

Budgeting for AEO vs content production

Discovered Labs pricing:

• Search Visibility Diagnostic: €4,370 one-off. Includes AI visibility audit across major engines, answer modelling, entity mapping, and schema implementation.
• Establish: €7,995/mo (month-to-month), €6,995/mo (6-month commitment). Up to 20 SEO and AEO articles using CITABLE, visibility tracking, competitor monitoring, structured data, backlinks and brand consistency work, and strategic Reddit engagement.
• Compete: €12,995/mo (month-to-month), €10,995/mo (6-month commitment). Establish deliverables plus up to 28 content units per month, landing pages for high-intent keywords, and quarterly business reviews.
• Enterprise: Custom scope for programmatic content at scale and original research studies.

The key budget question for a VP of Marketing presenting to a CFO: can the agency prove which AI prompts you're winning with direct citation data, or only estimate visibility based on traditional search signals?

Contract terms and renewal policy

Every Discovered Labs retainer is month-to-month. No annual lock-in. The accountability mechanism is simple: if citation rates and pipeline attribution improve, you stay. If they don't, you leave. We prefer that model because it keeps us focused on outcomes rather than retention tactics. Full pricing and terms are on our pricing page.

Conclusion

AEO is an engineering and measurement problem, not just editorial production. Agencies that prove citation rates with real retrieval data, connect citations to pipeline attribution, and structure content specifically for RAG systems will outperform those that scale volume and assume AI visibility follows. If you want to benchmark your current AI citation rate before committing to any partner, book a call and we'll run the audit and tell you honestly whether we're a fit. Or score your existing content now with the free AEO content evaluator.

FAQs

What is the cost of a Search Visibility Diagnostic with Discovered Labs?

The Search Visibility Diagnostic is a one-off payment of €4,370. It includes an AI visibility audit across major engines, answer modelling, entity mapping, and schema implementation for LLMs.

How long does it take to see initial AI citations after starting with Discovered Labs?

Initial citations typically begin appearing after publishing CITABLE-structured content. Full optimization, where citation rate stabilizes and competitive share of voice becomes measurable, takes sustained effort over multiple months.

What is the minimum contract commitment for a Discovered Labs retainer?

All retainers are month-to-month with no annual lock-in. You can scale, pause, or end the engagement each month without penalty.

Can Discovered Labs track citation rates across multiple AI engines?

Yes. Our proprietary tracker runs automated queries every five days across ChatGPT, Claude, Perplexity, Google AI Overviews, and Gemini, reporting citation rate and share of voice per engine, per topic cluster, and per individual prompt.

Why don't traditional keyword rankings predict AI citation rates?

LLMs retrieve content using vector embeddings and semantic matching rather than evaluating keyword positions. Our research on what drives AI citations shows that information consistency and source freshness matter more than ranking position for AI visibility, meaning structural extractability and consistent off-page claims drive citation rates more than domain authority.

Key terms glossary

Answer Engine Optimization (AEO): The process of structuring content to be retrieved, synthesized, and cited by AI search engines and LLMs, distinct from traditional SEO which optimizes for ranked document lists.

Generative Engine Optimization (GEO): The practice of optimizing content to be retrieved and cited by generative AI engines that synthesize answers from multiple sources, closely related to AEO but specifically focused on generative models like ChatGPT and Claude.

Citation rate: The percentage of AI-generated answers in a tracked query set that reference and link to your website as a source, measured across specific engines on a defined cadence.

Information consistency: The alignment of facts, claims, and brand details across multiple independent web sources (including Reddit, G2, industry publications, and your own site) that LLMs use to evaluate trust before citing a source.

Retrieval-Augmented Generation (RAG): A technical framework where LLMs retrieve semantically relevant document passages from an external corpus before generating a cited answer, separating the retrieval step from the generation step.

Share of voice (SOV): The percentage of AI-generated answers in a tracked query set that mention your brand relative to total brand mentions across your competitive set. It is the primary AI visibility KPI because it contextualizes performance against competitors rather than measuring absolute mention volume.