Tag Archives: chatgpt

An Example of AI Gone Wrong

Sales Blog, The Navigator News, , , ,

AI is one hell of a tool. I use it a lot myself. But I’ve also talked a lot about avoiding Chat Crap in your work, because AI can sound inauthentic, and that’s where “Good AI” becomes “AI Gone Wrong.” As Amazon stated:  “Roughly 57% of all web-based text has been AI generated or translated through an AI algorithm, according to a separate study from a team of Amazon Web Services researchers published in June.” I’m not sure I completely buy that 57% number – it seems like an awfully high number, considering how short a time AI has been in play – but even if the number is half that, it’s huge.

I also think that AI can be handy for prospect research, as well as content generation.  But you have to be VERY careful. The linked article below is one great example of AI gone wrong. Here’s the opening paragraph:

Tony’s KC:  The Heart of Kansas City’s Barbecue Scene

“Tony’s Kansas City has a rich history dating back several decades in Kansas City. Tony, a BBQ aficionado with a love for crafting the ideal smoked meats, opened the restaurant and soon won over many devoted patrons. The early going was modest, with a limited menu that concentrated on honing the fundamentals. Tony’s KC grew in popularity and offers throughout time, earning a slot on the local and tourist calendar that is not to be missed.”

Sounds like a great place to eat, right? There’s only one problem. Tony’s Kansas City is NOT a restaurant. It’s a local news blog that keeps people up to date on what’s really happening in my fair city, and only every now and then does he even talk about barbecue.

Now, think about this. This piece of terrible, lazy “writing” is on the Washington Post Magazine‘s website. That’s a national outlet. With one quick fact-check – a visit to Tony’s Kansas City blog – whoever prompted the AI could have figured out that the article was junk. Nobody did.

This happens because people think that AI is magic and infallible.  They’re using AI the wrong way because they don’t understand it, and they don’t know how to treat it.  You should treat AI like the best intern you’ve ever had:  It has a higher IQ than you, it has 20 Ph.D’s, and it has absolutely no street smarts whatsoever.  My favorite platform at the moment is Claude; I did a back to back video comparison on Claude vs. ChatGPT and put it on YouTube to explain why.

The “street smarts” have to come from you.  Your job is to guide AI, edit it, and back-check it.  Apparently, the Washington Post doesn’t understand that.  Here’s what happened.  Someone – most likely a low-paid person and possibly an actual intern – was assigned to write an article. I haven’t a clue what platform they used, or how they prompted it, but I do know that they ended up with an article that was a complete fabrication with precious little relationship to the truth.  And, once they had it, they didn’t even bother clicking through to Tony’s website (his name really is Tony Botello – that’s one thing that the article got right) in order to verify that they were in the same solar system as the truth.  As of this writing, the article has been up for a week with no changes or corrections.

So, how should you use AI platforms?  Here are a few of the techniques that I use:

  1. Prompts should be several sentences in length and be fully explanatory; one-sentence prompts usually get you junk for results. What we call “AI” is also known as large language models, meaning that the real feature is the platform’s ability to understand sentences, paragraphs, and context.  The more you put in your prompt, the better your results will be, even if you’re doing prospect research.
  2. Don’t be afraid to ask your AI platform to try again. Every popular platform has a button to do so, but I recommend suggesting your own refinements.  For instance, often I’ll say something like, “That wasn’t bad, but can you try again without the buzzwords, using plainer language?”
  3. In that vein, you should always specify a tone or style in which you want the result, as well as a length (when you’re going for a written document). If you don’t specify word count, AI tends to go very long.
  4. If your work depends on the accuracy of AI’s conclusions, take a quick moment to fact-check at least the biggest issues in the result.
  5. Once you have something you’re happy with (again, for a written document), you should always refine it a bit, to add your own language and your own touches to it. I created a video on how I use AI to create written work.  For the record, I did NOT use AI in this one.

Now, here’s what is really scary.  The Post article now becomes part of the “knowledge base” on the Internet, and will at some point be referred to as source material for other articles, whether human-written or AI-written.  Expect the amount of Chat Crap to expand exponentially.  This, friends, is why AI will not replace people in too many meaningful roles.  If you use AI platforms (and to be clear, I recommend that you use them and use them well), make sure to remember that YOU are the street smarts.  Keep that in mind, and you will do fine, and you will avoid “AI Gone Wrong.”

How to Easily Spot AI Blog Posts – And How to Use AI Properly

Sales Blog, The Navigator News, , ,

Hey there, savvy sales champs! Ever thought about having a secret weapon that boosts your sales game? Enter ChatGPT, your new best buddy in the sales world. It’s like having a Jedi master of conversation on your side, helping you charm clients, answer tricky questions, and close deals faster than you can say “commission.” Whether you’re a seasoned sales pro or just dipping your toes, this AI wizard is about to revolutionize how you hustle. Buckle up and get ready to unlock the magic of ChatGPT for your sales hustle!

Now, did that sound like me, Troy Harrison, the Sales Navigator?  Hell, no, it didn’t!  And if you use AI applications (whether ChatGPT or others) without any editing, it won’t sound like YOU either.  I purposely asked ChatGPT to write an intro for this article knowing that it would come up with something that was so transparently NOT me as to (hopefully) get your attention.  What we’re going to be talking about today is how to easily spot AI in blog posts, articles, and other written documentation – which, hopefully, will show you how to use AI correctly.

At current count, there are over 5,000 Generative AI applications competing for a growing AI user base.  “Generative AI” is the proper term for applications that accept a prompt from you and then generate some sort of written document as an output.  AI can write blog posts, articles, books, and even movie scripts (which is a big reason that the screen writers have been on strike – seems self-defeating to me, but I digress), and it’s easy to get started.  It can do this well or it can do it badly.  I hope we can agree that the opening paragraph of this article was “doing it badly,” but I bet you’ve seen a ton of posts on social media that looked pretty much like that.  AI is in its infancy, but at the moment, it has a number of verbal “tics” that are dead giveaways.  Let’s look at three of those now.

  1. Over-the-top language: Don’t get me wrong, I like enthusiasm. I once had a client describe me as a “human exclamation point!”  But too much AI output can sound like the person is on the biggest sugar high ever. My opening paragraph is an example.  “Hey there, savvy sales warriors” is a phrase unlikely to ever be typed by human fingers.  When you see a post that is virtually screaming at you, it’s probably AI.Your Hack:  I asked ChatGPT to use “casual language” in my prompt for the intro, and that was the result.  If you don’t specify your wording, you’ll get an abundance of business buzzwords and language that sounds like a term paper for an MBA program.  If neither one of those extremes mirrors how you speak and write – and they probably don’t – you need to be specific in the language you want for an output.  “Don’t use buzzwords” is one of my favorite prompts, and sometimes I’ll paste in a couple of paragraphs of my own work as a writing sample and tell ChatGPT to mirror my verbal cadence.  Until you get really good at your initial prompts, don’t accept the first result you get.
  2. Giveaway words and phrases: There are certain phrases that are dead giveaways.  “Say goodbye to X and say hello to Y” is a frequent one.  Or, “I hope this email finds you well.”  (That one sticks in my craw a bit, because I used to open a lot of emails with, “I hope you are doing well.”  AI took that one away from me.)  Words like “unleash,” “revolutionize,” and “unlock” are also common tics that give away a written product as AI-generated.Your Hack:  Sometimes, it’s enough to just accept a result with some buzzwords in it, and then rephrase it yourself into more direct and plain lingo.  Watch LinkedIn and Facebook ads, and you’ll see a ton of AI generated ads.  Read them and analyze them, and you’ll see other common verbal tics.  Then avoid them.
  3. Intros and closes: ChatGPT uses very familiar patterns of introductions and closes to its output.  Again – you probably already know what they are.  Avoid these, especially the intro.  The first paragraph or even sentence tells the reader if they want to continue, and if your first paragraph sounds AI, many readers will bail out.  Hmmmm….I’m hoping that most of you knew what I was doing and made it past that crappy first paragraph!Your Hack:  Always write your own opening and closing to a written document.  ChatGPT is actually pretty good at writing the body of an article with bullet points and analysis, but is weaker at getting into the article.  Your solution is to use your own ideas and thoughts to open your posts, blogs, etc., and then allow ChatGPT to do some of the heavy lifting in the middle.  Remember – make sure that the language flows and sounds like you.  If you come off as inauthentic, people won’t read you, and they won’t buy from you.

The bottom line is this:  Anytime you want to use a Generative AI tool to make your communications more efficient – and you should – don’t look at it as the complete solution.  As my friend Chase Aucoin likes to say, “ChatGPT is the best intern you ever had.  It’s smart, it has 20 PhD’s, and it has no street smarts whatsoever.”  The first paragraph of this article was an example of the lack of street smarts of ChatGPT.  It’s also the only piece of this article written with ChatGPT – but when I asked ChatGPT to write an article explaining how to easily spot AI in written documents, something happened that has not happened to me before.

I got a blank response. Zilch.  Zip. Nada.  Seriously – I learned that ChatGPT cannot explain to you what the dead giveaways of ChatGPT are and how to easily spot AI! After I stopped laughing, I wrote this article.

Generative AI is a great tool and will make you more efficient – but it won’t replace you.  So don’t ask it to, and you’ll be fine.

Why Doesn’t Cold Calling Work Anymore?

Sales Blog, The Navigator News, , , ,

I’ve been thinking a lot about the future of selling.  Last week, I defined the four trends that I believe will drive sales into the rest of the 21st century, and I’m even hosting a Webinar on those trends on July 6.  One of the dangers of thinking about the future, however, is that you lose sight of what worked in the past and can still work today.  I’m not going to do that.  That’s why this week’s Navigator is going to be about why, if you think cold calling doesn’t work, maybe the problem is you and your skills.

Cold calling is probably the oldest technique for generating new prospects.  It’s taken many forms over the years, but it’s still around.  Should it be?  Spoiler alert – YES.  But you have to be good at it.  Yesterday, I had a cold call that illustrated something I’ve been saying for years:  If you think cold calling doesn’t work anymore, you’re probably lousy at it.  I’m going to tell you about the cold call I received, where he went wrong, and how you can do it right (and why you still should).

Yesterday, I was cold-called by a guy who was selling recruitment services that matched military veterans to employers (a worthwhile cause).  On the face of it, this is probably a guy who should be talked to if you are hiring.  Here’s where he went wrong.

The first time he called, his phone connection was so bad that I couldn’t understand anything he was saying.  He said, “This happened on my last call, let me switch phones and call you back.”  Wait – if it happened on his last call, why didn’t he just stay on the other phone and call me from that one?

When he called back, he introduced himself with his name and company (good so far).  Then, he went into a long spiel about what his company did, and that this was definitely a sales call, that he’d be wanting to talk about getting my hiring business, and could he have 30 seconds?  This spiel took him more than 30 seconds.  I timed it.  So at this point, I just explained that I don’t do hiring because I don’t have a staff.  That’s not 100% true, as I just hired my very own sales executive (the business is growing), but I didn’t have any MORE need for hiring.

Then he asked, “Well, could you tell me who would do all the hiring at Salesforce?  It’s a big company.”  That’s when it dawned on me.  He thought he was calling an executive at Salesforce.  For those of you who haven’t been with me for the nearly 19-year ride of my business (most of you), I’ll give you the backstory.

When I started my business on September 1, 2004, I called it “Salesforce Solutions.”  The idea was not that I worked with the Salesforce CRM, it was that I solve companies’ problems with their own sales forces.  Seemed like a great idea, but it wasn’t, due to marketplace confusion (up to and including calls like this one).  That’s why I rebranded around my own name in 2014 – ten years ago.  This guy had somehow gotten ahold of ten year old data and thought that he was calling an exec with a large company.  I explained his mistake to him, politely, and wished him well.

He is probably thinking, “I don’t know why I’m cold calling – cold calling doesn’t work!”  No.  He’s just very, very bad at it.   Before I give you the four keys to being good at cold calling, however, I want to talk about the elephant that I put in the middle of the room.

If I’m talking about the future of selling, artificial intelligence, what ChatGPT can do, etc., then why am I talking about this old sales technique?  Simple.  First of all, cold call prospecting is STILL the most controllable way that salespeople can build their businesses and attract new customers.  Other strategies, such as networking events, social media, etc., are great and can work, but they are not numerically predictable and controllable.  And we need a method that is.  Cold-call prospecting should be part of your repertoire and routine.

Second – and now we can directly look the elephant in the eyes – cold calling is not (yet) something that AI can do well.  In the article I wrote a few weeks ago, I said that salespeople need to get better at those things that AI cannot do well.  Cold calling is one of those things. By using AI well in things that it’s good at, and then using the time saved to improve your skills at the things it does not do well, you can become the complete package.  With that said, here are the four keys to being good at cold call prospecting:

  1. Test your technology. This means that, if you’re using the phone, make sure that you get a good connection and that you are understandable on the other end of the line.  If you sound like you’re calling from the bottom of an oil well, find a different phone.  You MUST be understandable.  And while we’re on the topic, no damned auto-dialers!  As a call recipient, nothing kills your chances quite like the 2-3 seconds of dead phone line between the time that the call recipient answers and says, “Hello?  HELLO?” and you pick up and start talking.  Use your fingers and dial the phone.
  2. Use a good database. Even the best databases can have about a 10% obsolescence factor – but in this case, “obsolete data” means a year and a half, not ten years.  I think someone would have to be working hard to find data old enough for my past company name.  By the way, here’s a quick hack.  Most good library systems have subscriptions to free database services like ReferenceUSA, Data Axle, A to Z Databases, or the like.  These databases give all pertinent information including contact names.  You must have a NAME to ask for – not “the person who.”
  3. Be concise and powerful. From the time your prospect picks up the phone, you have 15 seconds – MAXIMUM – to give that person a reason to talk to you.  Crap like “How are you today?”, “is this a good time?”, or other babbling uses up that 15 seconds, as well as your prospect’s patience.  Introduce yourself and give ONE powerful sentence that explains how you benefit your customers, and how you could benefit them.  Yes, this requires thinking and refinement.  DO IT.  You could even try getting ChatGPT to help you write the sentence!
  4. Ask a strong question. After giving your one-sentence explanation, ask an open ended question about the prospect’s potential problems and needs to stimulate the discussion.  Then you can ask for an appointment.

If you’re thinking that cold calling doesn’t work, it’s time to get good at prospecting.  Don’t waste your prospects’ time – or yours.