Online Review Management [TRANSCRIPT]
B.L. Ochman: Hi, and welcome to Beyond Social Media Show, the podcast for advertising, marketing and digital communications professionals. I am here today with Curtis Boyd, who is the founder of objection.co. He describes himself as a husband, a father and a tech founder. He studied data science at MIT. And he says he loves solving complex data puzzles. You'll find his site at objection.co.
How Many Fake Reviews Are Online?
B.L. Ochman: Your website says that you want to help people find illegitimate reviews and it states, your mission in the following way "to provide affordable and quality software that can identify and dispute illegitimate content. We want to empower business owners and marketers to feel confident that they have protection from consumers who feel they can write whatever they want, to help them comply with terms and conditions to the websites they write reviews on. We want to help businesses create responses that generate consumer confidence. So even bad reviews can help them win new customers. So before we get into exactly what you do to make that happen, because to me, that's fascinating. Let's talk about how many fake reviews there actually are online.
Google’s Transparency Blog Cites Fake Reviews
Curtis Boyd: Holy moly. So let's put some context to it. Last year, Google released a blog. They call it a transparency blog. And they tell you how many fake reviews they removed from the Google Maps network. If you think of Google Maps, in 2020, they removed 55 million fake reviews, which is 160,000 fake reviews a day. And in 2019 they removed 75 million fake reviews. So in 2019 and 2020, it was 130 million fake reviews.
B.L. Ochman: And that's just on maps? That's just on Google Maps. Yep. What about TripAdvisor?
Whack-A-Mole With Reviews
Curtis Boyd: Sure, TripAdvisor's numbers are much lower, much lower. They reported a few million. Off the top of my head, I want to say it was about 2 million, maybe 3 million. They do a lot better job moderating their content. In my opinion, their back end is more powerful to recognize fake reviews, but they still struggle with it. I mean, millions of fake reviews are getting through. So Amazon has been under fire lately. Terrible. Yeah, they disclosed in 2020 that they removed over 200 million fake reviews. And just one year. So yeah, if you think of the last two years of Amazon, last two years at Google, you're looking at over 500 million fake reviews identified and removed. And in my opinion, they're playing whack a mole. When these moles are just multiplying like crazy. There are so many fake reviews that are still there that they haven't touched and they haven't addressed. And it's such a major, major problem.
Curtis’ Amazing Story
B.L. Ochman: So I was wondering, Was there some incident that happened that caused you to want to engage in this company to start it?
Curtis Boyd: Yeah, yeah. So a little over nine years ago, I was a nursing student finishing up my bachelor's in nursing science. And I was precepting in a hospital and I made my way to the ER where I was working. You know, I was following around a nurse that day. And a doctor came into the unit. He is a plastic surgeon. He was doing a consult with a patient who had, you know, issues. And the doctor was in a bad mood. The doctor, after he did his consult, was telling me how a bad review was hurting his private practice. He said it was fake. And it was illegitimate. I was kind of in a bad mood, though, too, because I had a ton of student loans. So I was like, No, I hear you, Doctor, I owe about $32,000 in student loans, like my day is not going great either. And he kind of sarcastically said, well, Curtis, if you can figure out how to remove this review, I'll pay off your student loans. I was like, What do you mean, he's like, well, I probably lost $150,000 worth of surgeries just this week, I probably had 10 to 12 cancellations that I'm aware of since this bad review came out
B.L. Ochman: One bad review?!
Persistence Paid Off, Bigly
Curtis Boyd: That's right. Wow. And so I was a student nurse. I didn't know about contracts, financials, how businesses work. I was you know, 22 years old. I was just just a student nurse trying to trying to listen to a new opportunity. And so I went home and I told my mom. My mom worked at the hospital. She knew the doctor. She said yeah, this doctor is legitimate. He probably means it if he said it. I called his office to verify with the front office staff. I was like, okay. The doctor said he'd pay off my student. I was like, was he serious? Because I'll try. I'll try my hardest. And she was the receptionist. She was so nice. She's like, yeah, we've hired lawyers. We've hired other people., Really? Yeah. And they're like, no one can do anything about it. And the doctor promised me it was fake. He's like, Curtis, I promise you this, this review is fake. So that's when I'm like, Okay, I'm on it doctor, I accept. And I'm emailing. I'm calling. I'm doing everything I know how to do. And I'm getting nowhere.
B.L. Ochman: I've been there. It's really hard.
Curtis Boyd: So what I ended up doing is I had about $800. In my bank account, I bought a plane ticket to San Francisco, where the headquarters of this company is, and I went into the building, and I started talking to people. As they walked in and out of that building. I said, Excuse me, do you know how to remove a fake review? They're like, what do you mean? Like, I'm sorry, I have a doctor in LA, they have a fake review. I need help removing it. Can you help? And they'd be like, do you want money? I'm like, No, I'm not homeless. I'm sorry. If I sound crazy, but I have a problem, Do you know how to get a fake review down? And I could tell like, I probably approached at least 200 people in and out who walked in and out of that building. I could tell a few of them knew exactly what to do, but they're just busy, or they just didn't want to help me or maybe I scared them. I don't know. But yeah, maybe a little bit. I was just a 22 year old on a mission. And I didn't get anywhere that day. But I had a lot of hope. Because I could tell again, in a few people's eyes, like they knew how to help me. I knew they did. And I came back the next day, same thing, I approached people in and out all day long. No one would talk to me. I stayed another day. And finally, finally someone said, Yeah, I can help you. And they, they, you know, they took me to a local coffee shop, they sat me down. And they showed me step by step what I needed to do to get this review, successfully disputed because I can't remove anything. I'm not an administrator. I'm a nobody really just trying to help advocate for this doctor. Anyway, she sat down and showed me step by step how to get this review properly disputed. And 48 hours later I had a check for $32,000.
B.L. Ochman: Get out! really what a fabulous story!
Suddenly In Business!
Curtis Boyd: Yeah, that particular doctor happened to be on the board of directors for the physician network of over 700 doctors at that Medical Center. He told me how much I should charge, a way to structure a business, got me introduced to a bookkeeper, a CPA to set up the business and everything. And by the time I graduated nursing school, I had over 500 physician customers paying me a monthly fee to manage their reputation.
B.L. Ochman: That's astounding. I love that story. That's fantastic. Yeah. How many fake reviews have you removed since then?
Defining Fake Reviews
Curtis Boyd: Gotcha, gotcha. So I define a fake review as someone who never had an experience with the company, right? Someone who never spent money in exchange for a product or a service, this might include an untruthful customer, an ex employee, or competitor, all of these things. So over time, 1000s of reviews.
B.L. Ochman: That's really remarkable. In a blog post, you said that negative customers, customer feedback is actually inevitable. What, what should companies do to prevent them?
Curtis Boyd: That's a great question. If, you know, I try to be really intentional with how I talk to new customers, because they'll come and they'll say, Hey, I got a bad review. Can you remove it? And I'm like, okay, is it legitimate? Is it real? You don't have a reputation problem, you have a customer experience problem. So if you want to prevent bad experiences, bad reviews from happening, you need to prevent bad experiences from happening. This has to do with setting expectations properly. This has to do with customer fulfillment, communication. I hate to say it but fundamental things, processes that you need to do better at so that you can set your customers up for success.
How To Respond To Legitimately Bad Reviews
Curtis Boyd: At that point, I like to do journey maps. We try and figure out each little step the customer takes throughout their business and identify what part of that machine is broken, what part of your business mission needs a little bit more attention and needs a little bit of grease. Normally, when you get a bad review, it's going to be legitimate, that feedback is actually extremely valuable because they're telling you what part you need to focus on what you need to do better on now. For doctors, sometimes You're gonna have a bad outcome. And that happens, like life happens, life isn't perfect. Things don't go the way they should, that's fine. What people read reviews for is to understand, do you care about my outcome? Like, if I hire you to do something, are you not only going to do your best, but if something goes wrong, are you going to try to fix it? Or are you going to try and take care of it? Because that's who I want to spend money with. I want to spend money with someone who appears to care about customer outcomes and looks like they're going to try to make it right if something bad happens,
People Want To Be Acknowledged
B.L. Ochman: You know, a long time ago, pre internet, I had a company that was called Rent-a-Kvetch. A kvetch is a Yiddish word for a complainer, and I would complain for other people. And what I learned was mostly what people want is someone to listen to them and say, Oh, I'm really sorry, that happened to you. And 99 out of 100 times, people don't say that. They sort of defend themselves, or they I don't know what they do. But they make people more angry than they started out to be by the way that they respond. So what you're talking about, really getting into the weeds of their process is so crucial, but also that they have to respond. Absolutely. That's really incredible. So is there something that your customers can do themselves through your software? Or do you set this up and do it for them?
Curtis Boyd: That's a great question. So when it comes to disputing reviews, that's all done by our system. Once you put your reviews into our dashboard, our software is going to tell you which ones it recommends disputing, and which ones it recommends responding to. Because if it's illegitimate, you should dispute it, right? It doesn't belong in there.
B.L. Ochman: So you train the AI of your software. That's right. So you left nursing school, you went to MIT,
Leaving Nursing School, Inspired By Elon Musk
Curtis Boyd: I left nursing school, I started my reputation practice. A few years later, I was getting really busy. One physician network turned into a few just working with doctors turned into working with doctors, lawyers, contractors, you name it, we started helping other industries. And my time just disappeared. Because I was on the computer all day reading reviews, monitoring reviews and disputing reviews. I asked myself, you know, is there a way that I don't have to do this anymore? Because what would happen is I would train someone to do this. And they would go off and start their own reputation company with what they learned. And that's okay. There's plenty of business out there. But you know, what I ended up hearing one day, I heard one of these TED talks with, you know, Elan Musk, talking about artificial intelligence and how we're all gonna lose our jobs. And then I was thinking, well, could I lose mine?
Curtis Boyd: Can I lose my job to AI? Could AI learn how to read and dispute online reviews? I was like, Wait a second, it totally could. So I went back to school for computer science. And I started learning how to code. I ended up building the software company objection.co,, which did what I used to do, it looks at reviews the way I do. It disputes the route the reviews the way that I do. So it's got years of experience and years of know-how within disputing reviews. And it's amazing because that way I get to focus on growing my business and the software gets to do all the deliverables, all the fulfillment, and the fulfillments a lot faster and more organized than than before.
B.L. Ochman: Which is really a classic smart entrepreneurs story. You know, it really is. So what's the reviewer pod analysis?
Analyzing & Scoring Reviewer Behavior
Curtis Boyd: Oh, reviewer pod analysis? Yeah. So that's a behavioral metric we look at when looking at a reviewer profile. So when we're trying to make a determination if a review is legitimate or not, we look at two things. We look at reviewer content. And we look at the reviewer behavior like the profile itself, We don't just look at the review that was written for our customer who hired us, we actually look at the all reviews written by the person to see who else they wrote reviews for. And then we look to see if there's a pod. A pod is kind of like, you know, dolphins and whales. Fake reviewers travel in pods too. They do because when someone goes, Hey, I want to hurt this person's reputation. I want you to post some one star reviews. They use the same profiles for their other customers. So what you see are the same profiles writing for the same businesses. You might have three bad reviews for your company, but those same three people likely wrote positive reviews for someone else. So we look for that automatically, which is something we check for something that might take you hours. It's done in a heartbeat.
B.L. Ochman: So you score the reviews in a certain way?
The Software Tracks Its Own Process With Video
Curtis Boyd: We do what's called the objection score. So zero means that the review just looks legit, everything about their behavior, everything about the content, just like squeaky clean. You got to respond to it. Now, the scores are zero to five. At the other scale, you have a review that scores a five, that's saying this review does not belong here, this review should totally be disputed. And all you have to do is press one button, it's the dispute button, you just push that dispute button, our server will do the rest. And the output is actually a video we call it a proof of work video, because our server records itself working, not only will it dispute it, but it will make a video of itself working so that you can see the work being done, you can see the communication to and from the administrators and everything's just really well documented.
How To Remove A Bad Review From Yelp
B.L. Ochman: That's absolutely brilliant. Now someone told me recently that they had a client who had like five year old negative reviews, that the company had since reorganized, that there was a new person running it, and that they were trying to get these reviews removed from Yelp. And she said no matter what you did, there was no way that Yelp would be responsive to that. But I think you have a way.
Curtis Boyd: Yeah, I mean, we don't have any special relationship with them, other than them probably not liking us very much. But that being said, there's a few things you can do with Yelp. Like I said, you need to look at both the reviewer content, like what they're saying, Look at the content guidelines, look at the terms of service. And then you need to look at the profile behavior, try to understand what the relationship is to this business. Is it as an untruthful customer? Is it an ex-employee? Is it a competitor, like you got to figure out like, Who are they and what's their relationship to the business, because if it's anything other than a customer who paid money, you need to dispute it, but you don't dispute it the same way. On Yelp, you can just flag the profile. So a lot of people will flag the review, when it's a behavioral thing when it looks off. And you actually need to flag the profile. It's a different type of dispute.
B.L. Ochman: That makes so much sense.
Curtis Boyd: Yeah. And then if they say no, Yelp has a follow up procedure that you can use. If you Google the phrase Yelp questionable content, it will be the first result that comes up, you enter the case number that Yelp gave you the first time, that way you can document it, and you say this is the case where you denied me previously, I would like to re explain my case, because I think that this is wrong. And I think there's been a mistake, I'd like you to reevaluate this review and then compare it to the content guidelines. Section 4A that talks about, review about review relevancy. And you need to be really specific about which part of the terms of service that you believe this review is violating or which community guideline that this profile is, is a big deal.
B.L. Ochman: Let's face it, the average person's not going to be able to do that. That's why we need you.
Curtis Boyd: Yeah, we've got a little under 800 reasons why reviews can qualify for removal. And really,
B.L. Ochman: Yes, I was gonna ask you what kind of review can and what kind of review can't be removed.
Some Reviews Can’t Be Removed
Curtis Boyd: So legitimate reviews can't be removed. For the most part, illegitimate reviews can be removed. It's always going to be up to Yelp or Google administrator whatever site review site this is. We do this for over 20 different websites. Yeah, so Glassdoor, trustpilot, TripAdvisor all the major ones,
B.L. Ochman: Any place where there are reviews, basically,
Curtis Boyd: The world's largest review sites we'll work on Absolutely.
How Objection.co Uses Neural Networks
B.L. Ochman: And so what's a tensor flow convolutional neural network?
Curtis Boyd: Yeah, let's break one word up at a time because each one opens up a huge world of explanation. But TensorFlow is an open source library. It's essentially a tensor is a format of data. And TensorFlow is what allows us to dissect this exorbitant amount of data using machine learning, you know, TensorFlow and Keras k-e-r-a-s are two large open source libraries that you can use to pick apart information now. There's different types of information, right? There's pixels, like the video, there's sound waves, there's sound bites, and then there's text Now within TensorFlow, you can dissect words you can dissect, images, you can dissect sound. You just have to be able to pull it apart. Think of it like atom by atom, and then let the computers understand the relationship between each atom, and then create its own analysis of it. Essentially tell it what you're looking for, and allow it to, to explain the trends to allow it to cut the machine learning to come out with an output that you're looking for. So that's in a nutshell what TensorFlow is.
Partnering With Google
B.L. Ochman: It's such a complicated thing. But so you mentioned on your website that you partner with Google, how do you do that?
Curtis Boyd: Yeah, so Google provides the NLP that we use in order to break down the language. So Google offers
B.L. Ochman: Natural linguistic programming, right?
Curtis Boyd: No, neuro linguistic programming, is like mantras, where if you start saying something enough, you're gonna program yourself to kind of live it. That's at least my understanding, a limited understanding of neuro linguistic programming, but natural language processing, okay, is the artificial intelligence version of NLP. And that's where we partnered with Google for them to help us to break down each word by word to relate. So if you have a sentence or a paragraph, it it looks at each individual word, and understands the relationship between each word and creates formulas, where we can try and create outputs that we're looking for, to predict if a review is written by a consumer, or if a review is written by someone else. Think of it this way, if you're a business with 100, reviews, there should be 100 unique authors, right? Because that's what a real review should look like. Unfortunately, when people purchase fake reviews, it'll be one author with 20 or 30. profiles, right? Because they've written all this content for these other profiles. So our NLP can identify unique authorship, so that we can say, hey, look, there's the same author for 30 profiles. Let's, let's go ahead and address that
B.L. Ochman: it is such a valuable service for companies. I mean, I've seen personally what happens to clients when, you know, I worked with a dentist, and one bad review was really a horrible thing for her. And, you know, I mean, after that, you really wouldn't want to go there. And I said, you have to respond. And you know, I ultimately talked her into responding. But that review was sitting there for a long time. So your technology is available on all the big sites, right?
Curtis Boyd: Our technology can collect reviews from all of these sites and make predictions on their legitimacy. Absolutely.
Alexa: ”Is this review real or fake?”
B.L. Ochman: You mentioned Alexa, are there voice reviews now?
Curtis Boyd: Oh, no. So we have an Alexa app. So if you look up our Alexa app, you can talk to Alexa and learn if a review is real or fake? Really? Yeah, absolutely. You just have to enable the skill. So it's an Amazon skill called objection.co and it will help you understand if a review is real or fake.
B.L. Ochman: Wow, that is really quite remarkable. So what's Win Back? That's when you get the review reversed if it was bad?
Curtis Boyd: That's right. So, you know, a lot of business owners will say, hey, Curtis, I got a bad review. I want to remove it. And we'll say, Wait a second. Is it legitimate? They're like, kind of like, you know, yes. Right. So it's like, well, let's let's pump the brakes on removal for a second. And let's try to win this customer back. Win them back What do you mean, I mean, by providing a secondary experience that inspires them to update that original one star review. So how do you do that?
Owning Up To Mistakes
Curtis Boyd: How do we do it? We need to strategize. So we talk about owning up to what happened, like what really happened inside of our dashboard, you can invite other employees to share their story about what their side of the story is, like what happened from their experience, but you can really get a multi dimensional view, not only from the words of the customer, but from your employees. From there, you can say, Okay, how can we prevent this from happening again? Like, what are we going to do? What processes are we going to implement to make sure that we're being prophylactically like creating prophylaxis, so that this doesn't occur again for future customers?
Curtis Boyd: The third thing the win back campaign asked for is okay, what are we going to do to make this right? What it gonna take to broker their satisfaction because at the end of the day, this is about your customer satisfaction and you as a business owner, making good on your promise to deliver good Customer satisfaction. How far out of your way are you going to go for them? How much of that red carpet are you going to roll out for them? So you need to come up with a plan on what you're willing to do to broker their satisfaction.
Using Private Messages For Ice-Breakers
Curtis Boyd: And then the fourth thing you need to do is prepare a private message to them as the icebreaker, right? Acknowledging that they are not happy, acknowledging that you as a business owner are committed to their success; that you are committed to making this right. So we help each of those processes so that they can win the customer back and hopefully win their business for life.
Review Scanner Tool, Free To Try
B.L. Ochman: So wonderful service, but wow! Are you really involved in your clients' businesses! And that is, you know, that's a huge value. So you also have on your site, a free review scanner tool? What's that?
Curtis Boyd: Yeah, so it's a place where you can literally just copy and paste the the review into the tool, and it will grade it'll, it'll score it. You don't have to create a free trial, you don't have to get into our dashboard. It's just a way to stay on the front end and kind of toy with it, take a look at our NLP, see if you like the technology, see if it aligns with you.
Who Should We Trust? There’s An App For That!
B.L. Ochman: And so I guess my question, after all of this is how do we know what reviews to trust?
Curtis Boyd: Oh, man, well, we're coming out with an application for that. It's called the transparency app. So pretty soon we'll have a solution for it. It'll be available September 28. Where we're going to provide reporting on which reviews you can trust. And we'll have a nice little badge for businesses who have real reviews who have earned their reputation versus businesses who have paid for the reputation. So that is coming out. There'll be a mobile app as well as a Chrome extension, which we're really excited about. And yeah, consumers can use it completely free. It's totally free for consumers.
Leaving A Legacy On The Planet
B.L. Ochman: You're a long way from nursing school.
Curtis Boyd: That's right. This isn't Kansas anymore. Absolutely.
B.L. Ochman: Your business is fascinating. It really is. It sounds expensive, is it?
Curtis Boyd: Oh, you know, running a business is expensive, but it's also a joy. Being able to, you know, employ such talented people being able to spend time with them daily and watch their growth as individuals is really fascinating to me. So yeah, absolutely. But, but it's worth it. And I know the long term impacts that we're going to have on the lives of authentic and genuine business owners and the impacts on all the consumers we're gonna have, hopefully will help me leave a legacy on this planet.
B.L. Ochman: I think you will, actually. And I learned a lot just listening to you. And I'm sure that our listeners and readers are going to agree as well. So I just want to say thank you so much. And I will include in the posts of the blog, and also in the notes on the video, all the ways to get in touch with you. And I will look forward to learning about the new application. That's extremely cool.
Curtis Boyd: Great. Thank you so much. Yeah, I'm always happy to chat with people on LinkedIn if anyone ever wants to chat reviews.
B.L. Ochman: Sounds great. Thank you.