EP 263 - Stop Counting Stars: How to Mine Competitor Reviews for High-Converting Content Ideas
Learn how local brands use NLP APIs and modern scraping stacks to transform raw competitor review texts into high-converting website FAQs, operational updates, and definitive, cross-channel brand differentiators with Celeste Gonzalez of Last Mile Retail
Local SEO strategy often gets stuck playing a quantitative numbers game with review counts and velocity benchmarks. But what happens when you treat the collective text of your market's customer reviews as an ongoing, free focus group? This week, Celeste Gonzalez joins us to show how tools like Apify and Natural Language Processing (NLP) allow local agencies to scrape hidden market trends from Reddit, TikTok, and competitor Google Maps profiles. We break down exactly how to extract actionable copywriting blueprints from customer pain points, avoid confirmation biases, and keep your business answers consistent for modern AI search engines.
The Podcast Deets
- Segment 1: The Automation Ingestion Blueprint (00:00 – 13:32)
This segment addresses the technical and strategic fundamentals of gathering qualitative customer insights across disparate review platforms and social networks. It reviews structural data skew, identifies standard review-removal strategies, and outlines an enterprise scraping stack built directly upon Apify scripts, Google Colab notebooks, and the Google Cloud NLP API. - Segment 2: Organizing Raw Unstructured Data (13:33 – 21:30)
This discussion explores the process of managing unstructured language files without imposing internal confirmation biases. The conversation highlights why algorithmic, machine-learning-driven bucket classifications are significantly more accurate than manually prompted LLM architectures, and explains how to evaluate severe edge-case complaints against high-volume trend lines. - Segment 3: Turning Sentiment Into Tactical Revenue (21:31 – End)
The concluding section tracks the real-world operational translation of processed customer sentiment data into bottom-line performance improvements. Using proven studies from competitive plumbing and home construction fields, this segment shows how tracking unpublicized operational features and implementing clear pricing details builds cross-channel marketing trust.
Key Takeaways
- Review Content Rules Over Star Scores: While 65% to 80% of everyday consumers look exclusively at aggregate star averages and raw velocities, tracking underlying text profiles remains critical to modern content design and strategic brand creation.
- Algorithmic Categorization Beats Prompting: Relying on standard NLP APIs for initial review clustering avoids confirmation bias and exposes unexpected customer patterns that forced LLM prompting steps frequently overlook.
- Address Unspoken Fears Explicitly: High-performing content strategies use real competitor reviews to identify market anxieties (e.g., unexpected bills or delayed schedules) and counter them directly within core brand content.
- AI Search Requires Multi-Channel Accuracy: As search assistants leverage multi-source review records for citations, local brands must ensure their core messaging stays perfectly unified across website text, local listings, and social channels.
👇 Watch by topic:
• 00:00 | Introduction: The Review Content Goldmine
• 03:13 | Capturing Emotional Responses & Customer Sentiment
• 06:21 | The Problem with Skewed Data & Bad Review Ingestion
• 10:44 | Building the Automation Stack: Apify & Google Cloud NLP
• 13:33 | Aggregating Data: Machine Learning vs. Static Prompts
• 17:11 | Separating Critical Insights From Edge-Case Outliers
• 19:03 | Mapping the True Competitive Set via AI Discoveries
• 21:31 | Translating Raw Sentiment Data into Tactical Action
• 25:08 | Real-World Performance Case Studies: Plumbers & Homebuilders
• 27:45 | Local Brand Differentiation in a Highly Commoditized Market
• 33:46 | Cross-Channel Consistency: Aligning Operational & Marketing Realities
• 37:46 | Outro: Mining Untapped Audio Data and Call Logs
Related Links
Celeste Gonzalez
Nick Haigler
Clockworks
Interested in sponsoring this podcast or our newsletters please reach out to mblumenthal@nearmedia.co
Full Transcript -->
GREG: Hey everybody, welcome back to the Near Memo. Today we have a special guest, Celeste Gonzalez. Welcome back, Celeste! We had her on the podcast the first time to talk about TikTok and local search, which was a very illuminating discussion.
Recently, she published a piece on mining the text content of competitor reviews for fresh content and brand ideas. The term you used in the article was a "persistent focus group," which is incredibly accurate.
Mike has spent a long time at GatherUp—and both of us separately have argued for years—that there is an enormous amount of qualitative information living inside reviews that remains largely unexploited. Most of the time, local marketers focus strictly on quantitative metrics like rankings, star ratings, and the competitive thresholds required for baseline visibility.
Now, with the rise of AI search ecosystems, reviews have become even more critical as primary signals for discovery engines. Your article was fascinating because it refocused attention directly back onto the review content itself and what businesses can learn from it. Let's dive into that. Celeste, tell us what this article successfully covered.
CELESTE: Got it. Essentially, I wanted to address how SEOs typically focus on the quantitative side of reviews—things like aggregate star ratings and review velocity. It has historically been treated purely as a numbers game. When marketers did talk about review text, the advice was usually limited to "stuff your keywords in there."
Because of how search is changing with AI visibility—especially with the AI-powered summarized experiences users interact with on Google Business Profiles and Google Maps, as well as conversational tools like ChatGPT and Gemini—we are seeing that there is much more to be said about a business than just a static list of their services and products.
The goal of the article was to show that companies should actively analyze their own reviews along with competitor reviews to find their true competitive advantages and exploit them. This data can be used to run targeted SEO content tests, re-evaluate H1 and H2 header copy, and ensure that whatever a business wants to be known for is clearly addressed while resolving common customer anxieties.
GREG: What specific types of information do you hope to extract when performing this level of analysis on your own or competitor reviews?
CELESTE: You are primarily hunting for emotional responses. People who take the time to leave reviews are typically highly motivated; they either had an incredibly positive experience or an exceptionally negative one, and they feel a strong need to share it. Of course, you also have local guides and SEOs who write reviews because they understand the mechanics of the ecosystem.
When I look at review content, I analyze the specific language patterns being used and the overarching sentiment. Are there unresolved questions left behind? Is there underlying fear, anxiety, or anger driving the text? Those are the exact pain points that need to be addressed in your marketing copy. Conversely, if something is repeatedly called out as a major positive, you want to make a permanent note of that as well.
GREG: That aligns with our user behavior research. We have found that across many verticals—such as healthcare, legal, restaurants, and hospitality—a significant majority of everyday consumers do not read the actual text of reviews. They rely almost entirely on the aggregate star ratings and total review counts. Depending on the category, usually only 20% to 35% of the audience reads the text, scaling up for higher-consideration categories.
Because that metric dominates consumer behavior, the industry conversation shifted away from content and toward raw velocity and volume. But your article focused heavily on competitor reviews. Tell us what analyzing your competitor set actually reveals.
CELESTE: Looking at your competitors shows you exactly how you stack up regarding customer experience and brand perception. Just like you perform traditional technical SEO or content gap analyses, review data should be an essential part of that competitive matrix.
By mapping out their specific strengths and weaknesses, you can spot immediate opportunities. If one of their primary recurring weaknesses happens to be one of your operational strengths, you need to check if your website explicitly highlights that advantage. If a competitor's strength highlights a weakness of your own, you can look for creative ways to address it through targeted content.
MIKE B: How large of a sample size do you need to make these insights reliable? As you noted, third-party reviews are heavily skewed toward the extreme ends of the spectrum rather than following a natural bell curve. Furthermore, many businesses work incredibly hard to suppress or remove their negative reviews through various means—including persistent rumors of manipulation within Google's support facilities in India to get reviews removed for a fee.
Because public review content can be highly curated or totally skewed, unlike first-party surveys where customers offer balanced feedback in greater quantities, how do you find a sample size clean enough to back up your strategic marketing decisions?
CELESTE: While my recent article focused primarily on Google reviews, they shouldn't be your only data source. Depending on your vertical, your primary platforms will change—B2B brands, for instance, should focus on G2 or Captera.
To build a balanced sample size that isn't entirely polarized, you should expand your scope to include social media comments. While a business can try to delete comments on their own profiles, they cannot clean up external discussions across Reddit, Instagram, or TikTok. If a target competitor is being actively discussed on those channels, scraping those conversations provides a much more well-rounded data set.
GREG: Mike makes an excellent point. If you use review content to guide your website copy or brand positioning, how do you know you can trust that content? It reminds me of the political polling debates during the first Trump administration, where people questioned why traditional surveys missed the mark. The consensus was that people who voice opinions online tend to be activists on either end of the spectrum—loyalists or haters—while the quiet majority sits somewhere in the middle.
Given that online commentary is often dominated by these two vocal extremes, how do you verify that the insights you extract are truly representative before rolling out major updates?
CELESTE: It is incredibly difficult to decipher absolute truth when evaluating competitors. When analyzing your own brand, it's simple because you can validate trends against your internal first-party survey data. For competitors, it has to be handled on a case-by-case basis informed by a deep understanding of the local market.
For example, I worked with a small business client who knew for a fact that her primary local competitor was manipulation-heavy—paying to remove negative reviews while purchasing fake positive ones. By having open conversations with the client and layering in our own market research, we decided to downweight that specific competitor's data in our final analysis. You should still monitor them, but you keep their data in the background so it doesn't skew your primary strategic insights.
MIKE B: In terms of gathering information outside traditional platforms, there are plenty of automated enterprise tools designed to scrape standard review directories into a clean database. But how are you finding and aggregating relevant Facebook, TikTok, or Reddit comments at scale without relying on manual copying and pasting?
CELESTE: I use a web scraping platform called Apify (apify.com). They have pre-built scraping models specifically for Reddit, Instagram, and Facebook. I feed those data extractions into a Google Colab notebook to handle the initial data management, clean the raw text strings, and format them into spreadsheets. From there, I connect the data to the Google Cloud NLP API to run the core sentiment and entity analyses.
MIKE B: Does that API offer internal query capabilities so you can point it directly at a competitor's brand name, or how are you configuring it to find the right contextual text blocks?
CELESTE: Right now, that process relies entirely on "vibe coding"—writing custom scripts to locate specific topics, terms, and conversational buckets. I fine-tune the data extraction parameters based on the specific depth of information required for the project.
GREG: Let's talk about the exact mechanics of this process. How do you execute this framework from a step-by-step perspective?
CELESTE: I use Apify to pull the raw text data. I'm not sponsored by them; they are simply the most flexible tool I've found for this type of vibe-coding workflow. Once the data is pulled, I organize it within a master spreadsheet featuring dedicated columns for the review text, the publication date, and the source profile.
Next, I run the data through the Google Cloud NLP API. I highly recommend using a dedicated machine learning API rather than trying to prompt an LLM like Claude or ChatGPT to handle raw sentiment analysis. Standard LLMs require extensive prompt engineering and fine-tuning to categorize concepts accurately without hallucinating. The NLP API is a peer-reviewed machine learning tool that integrates directly into Google Sheets as a script add-on. It automatically outputs reliable sentiment scores and thematic buckets. You can then easily filter the sheets to surface high-priority language trends.
MIKE B: Does the NLP engine automatically aggregate those reviews into predefined buckets, or are you creating those topical categories yourself?
CELESTE: The NLP API generates its own automated thematic buckets based on its structural training data. If you want to perform deeper analysis—such as building highly specific, custom categories tailored to your business—you can write custom scripts to layer those rules on top of the initial output.
GREG: That distinction is incredibly important. If you approach a data set with rigid, preconceived notions about what you want to find, your internal biases can blind you to valuable, unexpected insights. Letting an algorithmic engine independently surface recurring themes ensures you notice patterns you wouldn't have known to look for.
MIKE B: Does the NLP API use human-readable categorizations or raw word pairs to build those thematic buckets? What does the resulting data structure actually look like?
CELESTE: The NLP API provides an extensive array of predefined, clear thematic buckets. You can look through its direct outputs to see exactly how it maps core concepts. If you want to layer on a custom metric—like tracking "brand trust," as Greg mentioned—you have to define what terms signify trust for that specific client, and script the rules to look for those indicators within the data.
GREG: Once your spreadsheet is fully populated with these NLP buckets, how do you pull out the final strategic insights?
CELESTE: I start by analyzing the predefined NLP buckets alongside the custom categories I've built. That's when you dive into the details to see what is bubbling up to the surface. I make it a point to read through the individual reviews within each bucket to understand the exact conversational language customers use.
From there, you can write additional scripts to summarize those insights into strategic recommendations. If you're working with a smaller data set, you can easily track and write them out manually.
This is the point where you make core strategic decisions. For example, if the data reveals that customers are consistently confused about how a product works, that is a clear signal to build detailed "how-to" guides, expand your website FAQs, or produce explanatory video content. Often, that information might already exist on a website but is buried where users can't find it. You can cross-reference your review findings with user behavior data to see if your helpful content is actually being reached, or if it needs to be moved to more prominent locations.
GREG: So you are hunting for common themes that represent a critical mass of users. When you evaluate whether an insight warrants a formal content recommendation versus being a simple edge-case outlier, what is your framework? Is it based purely on volume, like a ten-to-one ratio?
CELESTE: Volume is an important metric, but it isn't the only consideration. If nine people leave standard comments and only one person leaves a highly specific complaint, you have to look at the severity of that single review. If it highlights a harsh or critical failure point, it warrants a deeper investigation to determine if it points to a systemic operational issue or a dangerous blind spot—especially when analyzing competitor vulnerabilities. While quantitative volume is a reliable baseline, individual outliers should still be evaluated.
MIKE B: Do you ever aggregate reviews from multiple top competitors into a single master data set to build a more robust text corpus, or do you analyze each competitor individually?
CELESTE: I aggregate them all into a single master working sheet. I include a dedicated column for the competitor's brand name so we can filter down to isolate specific brands for deeper research or view the market's aggregate data as a single theme.
GREG: Aggregating data makes perfect sense, especially if you want to isolate a specific competitor you are consistently losing bids to. You mentioned defining a client's true competitive set earlier, which is a great topic. Many business owners have a skewed perception of who their actual competitors are. Walk us through how you define a true competitive set and what factors you consider.
CELESTE: It requires balancing known competitive data with fresh market research. You start with the client's historical knowledge—the competitors they have traditionally run against and any existing competitive analyses. Next, you look at real-time local search visibility.
If you consistently search for core service terms on Google Maps, you can see exactly which businesses own the top local visibility. From there, you look at their review dynamics. With Google's AI-powered review summaries now live, you can check what specific attributes are being highlighted for those competitors. If a competitor is earning prominent AI call-outs for premium features your client wants to be known for, that competitor immediately goes into our active data set.
Finally, you must explicitly ask the client who they are losing jobs to in the field. A competitor might not rank on the first page of a local search when you audit their terms, but if a local service provider—like a plumber—tells you, "We lose three out of five calls to this specific company," that business belongs in the database. There is likely an operational or brand message hidden in their reviews that explains why they are winning conversions.
MIKE B: What industries or business sizes have you deployed this analysis framework for so far?
CELESTE: I have primarily run this for small-to-medium businesses, with a heavy focus on home services verticals. Now, at Last Mile Retail, I am deploying this framework for large enterprise-level clients. We are still compiling that data, but I look forward to sharing those insights once they are ready for publication.
MIKE B: It seems like running this at the enterprise level would be incredibly efficient. Having access to massive datasets across hundreds of locations makes it much easier to isolate broad market trends or filter down to specific regions to see why one location struggles at a 3.9 rating while another maintains a 4.7.
CELESTE: Exactly. The scale of enterprise data makes pattern recognition much faster.
GREG: Once you have extracted these insights from your master sheet, how do you translate them into concrete tactical actions? You've mentioned website copy modifications and social media content ideas, but give us a clearer picture of the tangible updates you execute based on this data.
MIKE B: And to add to Greg's question: Did your clients actually implement your recommendations?
GREG: Exactly. It's a common issue in our industry—we frequently deliver detailed strategic recommendations to clients, and they end up sitting on a shelf.
CELESTE: That is the ultimate battle, and it depends heavily on the strength of your client relationship, their operational agility, and how much direct influence you have over their broader marketing channels.
To answer your question about where to go next: when dealing with small-to-medium businesses, many unfortunately struggle to execute. For the ones who do listen, the most common discovery is that they are missing crucial content entirely. A client will say, "We want to be premium-tier for this specific service," or "Our top competitor is winning praise for this feature, and we want to match them."
When you look at their digital footprint, the content simply doesn't exist. There is no trace of that messaging on their website, landing pages, or social profiles. Step number one is building that missing content from scratch.
Step number two is mapping your strengths and weaknesses directly against your competitors. We isolate the specific operational weaknesses we can actually fix through communication. For example, if a client has an internal customer service issue, an SEO content update can't fix that. But if the reviews reveal that negative sentiment is driven by simple miscommunication or an unclear booking process, we can fix that instantly by publishing transparent, step-by-step FAQ blocks on the website to set accurate expectations.
MIKE B: For your most successful clients—the ones who provided clean data sets, listened to your advice, and executed your updates perfectly—what tangible business outcomes did you see?
CELESTE: Our two most successful case studies were a local plumbing business and a residential home builder. Both saw substantial increases in consumer engagement and on-site conversions after rolling out our updates.
We didn't just perform basic on-page SEO tweaks like changing an H2 tag; we introduced comprehensive FAQ blocks based directly on review anxieties. I closely track user behavior data to see if visitors are actually interacting with the new copy blocks, and then validate those trends against final conversion growth. To ensure accuracy, you always have to normalize the data against year-over-year seasonal benchmarks, but even after accounting for seasonality, both businesses achieved clear performance lifts.
GREG: Without revealing proprietary details or breaking client confidentiality, can you walk us through the specific insights you uncovered for those two businesses, how you implemented them, and what the final outcomes looked like?
MIKE B: You can share the secrets, Celeste—we're among friends here!
CELESTE: For the plumbing business, we started with the baseline reality that home services are generally viewed with a lack of consumer trust. Plumbing is often treated as a pure commodity category, and consumers frequently face stressful communication issues regarding pricing transparency. The absolute biggest anxiety for a consumer booking an emergency plumber is: "What is this going to cost me, can I afford it, and am I going to get hit with a massive surprise bill when the work is done?" The second most common complaint is punctuality—plumbers promising to arrive at 4:00 PM and not showing up until hours later.
To counter these widespread market frustrations, we completely overhauled their website copy to emphasize explicit pricing transparency and guaranteed arrival windows. We addressed those structural trust issues front and center where prospects look for reassurance.
For the home builder, the recurring market anxieties centered around chronic scheduling delays and a feeling of being left completely in the dark during construction. Clients would complain, "They told us our home would be ready by this date, and it took months longer without any updates."
While diving deep into their positive review data, we discovered something fascinating: multiple customers praised a proprietary mobile app the builder used to send automated, real-time status notifications for every step of the construction process.
When we spoke with the client, we realized they had never mentioned this app in any of their marketing materials or sales copy. They didn't view it as a marketing asset, yet their customers loved it. We immediately brought that feature to the forefront of their website copy, transforming a hidden operational tool into a core brand differentiator.
GREG: Were these updates confined entirely to website copy, or did you deploy them across other marketing channels?
CELESTE: Those specific updates were implemented as direct copy changes on their websites. I provided additional strategic recommendations to the plumbing client's internal brand and social media teams, advising them to use their social channels to lead with transparent operational messaging. However, I have since moved on from my previous agency, so I don't have visibility into how their social campaigns performed over the long term.
GREG: Your advice to their brand teams was spot-on: identify the primary market frustrations, be completely upfront with your information, and address those pain points directly. Were these projects focused purely on content creation, or did you introduce technical or functional upgrades as well?
CELESTE: They were entirely content-driven. Our review audits didn't surface any technical issues or structural site errors that needed tracking.
GREG: Mike and I were discussing in the green room before recording that we are launching a new series focused on local brand building, because it remains a highly misunderstood concept. Marketers talk about brand building constantly, but they rarely outline the practical steps required to differentiate a local business or improve its visibility without a massive, corporate budget.
At a local level, true differentiation is the primary engine of brand equity. How can marketers use review data to systematically discover those local branding and differentiation opportunities? Most service categories—like the chimney sweep and dryer vent cleaner I have coming to my house tomorrow—are highly commoditized. Consumers rarely have a specific brand in mind; they just look for the fastest availability or the best reviews at the lowest price point. Having a distinctive brand identity provides a massive competitive advantage. How does this data inform that branding framework?
MIKE B: Before Celeste answers, Greg, I have a quick question for you. Did you read our detailed investigative series on local duct cleaning scams? We published an extensive feature on the massive amount of consumer fraud running rampant through that specific category, and I want to make absolutely sure you aren't walking into a trap tomorrow.
GREG: Yes, I am acutely aware of the fraud in that space. If you want a quick diversion, I'll explain my exact vetting process for this job. I recently wrote an article about my experience buying a garden hose, which Celeste actually used as a blueprint for her own "path to purchase" study.
For this project, I used Yelp’s Request for Quote feature. When you submit an RFQ, Yelp automatically asks if you want to route your request to multiple local competitors, which immediately alerts a large subset of the local market.
As the quotes came in, I systematically vetted each business by cross-referencing their Yelp and Google review profiles. I instantly eliminated one provider because they took too long to reply. I disqualified another because their history revealed a pattern of aggressive on-site upselling—their technicians would show up under a low initial quote and then claim they found structural issues to inflate the final bill.
I checked both positive and negative reviews, asked targeted questions, and made sure I was dealing with a legitimate operation. Assuming I didn't fall victim to sophisticated review fraud, I think they are well-vetted.
MIKE B: We will check back in next week to see if Greg got scammed by the duct cleaner.
CELESTE: I will definitely be tuning in for that update!
To answer your question regarding brand differentiation: as you noted, many local business owners don't have a distinct brand concept, or they simply say, "I just want my phone to ring, I don't care about branding." That is where you have to guide the client relationship. You have to help them understand that clearly communicating their unique selling points online is non-negotiable.
With the rise of conversational AI search, every piece of public sentiment text is fair game for extraction. Seer Interactive recently published a study demonstrating that consumer review directories have become the second largest citation source for generative AI search results.
Even clients who aren't highly tech-savvy are starting to ask how these AI tools affect their businesses. It’s top-of-mind for them, which gives agencies an excellent opportunity to pivot the conversation toward long-term brand building. By running these audits, you can show a client exactly what the market is saying about them online and ask: "Is this the image you want to project?" If there is a massive gap between how they want to be perceived and what their reviews actually say, that creates the exact roadmap for your content strategy.
GREG: So while review audits don't always generate direct brand-level taglines out of the box, they initiate a critical diagnostic conversation with the client. It allows you to see the functional gaps between market perception and operational reality so you can steer the brand in a more profitable direction.
MIKE B: And Celeste noted an excellent example of this with the plumbing case study. In that world, standard marketing relies heavily on cliché taglines like "On Time and On Budget." If a local business can actually deliver on that baseline operational promise 98% of the time, they automatically establish a differentiated brand in a category notorious for poor service.
GREG: Exactly. The bar is remarkably low because the baseline standard in many commodity categories is plagued by poor communication.
Celeste, in our final couple of minutes, we've established that systematic review analysis is incredibly valuable for website copywriting and content development. But have you discovered any unexpected applications for this data—such as feeding insights directly into a client’s sales process or operational training?
CELESTE: I wouldn't say any of our findings were completely surprising, because qualitative feedback naturally impacts every channel of a business. If an agency runs a high-performing paid ad campaign, but a prospect checks the business's reviews and sees unaddressed complaints, that ad will not convert. You can produce a beautiful, high-budget commercial, but if your public review history highlights unverified pricing or poor service, your marketing spend is wasted. Review optimization is entirely about establishing foundational trust across every consumer touchpoint, meaning it directly impacts every layer of your marketing ecosystem.
GREG: You have to maintain absolute consistency across every marketing channel and operational touchpoint. In the early days of local SEO, the golden rule was structural NAP consistency—ensuring your Name, Address, and Phone Number matched perfectly across the web.
Today, largely driven by AI discovery engines, the standard has evolved into contextual brand consistency. The identical brand promises, values, and operational details must be mirrored seamlessly across your social media profiles, your main website, and your local search listings. Once you extract these core insights from your review data, you must ensure they are communicated across every digital surface where your brand appears.
MIKE B: It is critical to extend that consistency into your physical operations as well. Your operational reality must match your marketing messaging perfectly; otherwise, your digital strategies will fail to drive real business growth.
GREG: That focus on consumer trust is paramount. Definitive brand building requires making clear promises and reliably fulfilling them—showing up on time, sticking to your quotes, and executing clean work.
Review text analysis is a critical piece of a much larger operational strategy that modern local businesses must embrace. For years, call-tracking platforms like CallRail have talked about the massive amounts of conversational data sitting completely untapped inside recorded customer calls. Companies captured thousands of hours of real customer interactions containing deep insights into customer pain points and expectations, but nobody ever audited them because doing so manually was too labor-intensive.
Now, with modern AI tools, you can mine massive audio and text datasets instantly. Business owners should aggressively dive into their qualitative customer data, especially given the current search algorithm emphasis on authentic, non-commoditized content.
CELESTE: Public Google and Yelp reviews represent the easiest entry point for this type of research. Because it is entirely public data, you can feed it into AI analytics tools without running into data privacy issues or exposing proprietary information. Once you master that workflow, you can begin expanding your audits into internal call logs, customer satisfaction surveys, and direct feedback channels to uncover deeper trends.
GREG: I believe Google is currently building a native version of competitive sentiment analysis directly into its business suites—whether they brand it as Gemini for Google Business Profile or Gemini for Workspace. Users will soon be able to connect their business properties to Gemini to run automated competitor reviews and local market analyses. That will be an incredibly interesting paradigm shift, but it is a topic for a completely separate episode.
Celeste Gonzalez, it has been wonderful having you back on the show. Thank you for sharing your frameworks and insights with our audience. Mike, any final thoughts before we close?
MIKE B: I just want to encourage all of our listeners to tune in next week for the conclusion of Greg's duct cleaning adventure!
GREG: Yes, stay tuned for the exciting details of my home maintenance saga! It starts with the duct and chimney sweep tomorrow, then we move on to casement window repairs, and then I have to deal with a local locksmith. I'm systematically running through every local home services category on my own house, so we will have plenty of real-world local content for foreseeable future.
Celeste, thank you again for joining us. To everyone listening: please like, subscribe, and tell your friends about the show. Thank you for tuning in, and we will see you all next week!
CELESTE: Thank you!