Longshot AI Longform Writer - Walkthrough and Review

The general AI chatbots like ChatGPT and Bard or Gemini and Bing or Copilot , whatever they're calling themselves this week , are really good at writing , but they have two weaknesses , accuracy and predictability .

I never know when the chatbots like Chat GPT , whether they're gonna give me just one paragraph or they're gonna give me an entire dissertation or something somewhere in between .

And I've found that even when I'm really specific in my prompt about how long I want something to be , they don't always listen .

I also experience unpredictability issues on what the formatting will be in what it generates .

Sometimes , they don't seem to understand what I want no matter how hard I try to explain it .

A few months back , I went on a mission to try and find a long form AI writing assistant to help me write articles and blog posts .

I wanted something that would be easy to use , produce full length articles , and be as accurate as possible .

After all my research and comparing , I settled on LongShot AI , and that's what I've been using .

And right now , I'm gonna show you around LongShot .

So come on .

We're inside LongShot at the dashboard , and this shows us a group of pretty popular auto blog workflows .

We're gonna switch and say view all so we can see everything here .

Some of their latest work flows are at the top , the video script , insurance blog , and fantasy story .

I also have how to guides , listicles , pillar content , which is my favorite , and then quite a few more .

Product specific things , we have product or service use case , alternative to , comparison posts , and product reviews .

You can also create your own workflow if you want a combination of some things that are here .

You wanna create something totally different .

You can use the custom blog builder or the bring your own workflow style .

They do add workflows fairly regularly , and that's what you see at the bottom is what's coming along coming down the pipe .

Now each one of these workflows is designed to give you something a little bit bit different , mostly in the formatting .

Sometimes it has to do with the length .

For instance , the video script , that'll be laid out like a script .

Listicles , obviously , will be in a list numbered list format and so on .

Pillar content is my favorite .

That's the one we'll look at , and there's really 3 steps to this .

The first step is you tell it what you want .

It'll then generate some options for a title .

You look through those .

You pick the one you want .

Once you have a title , it'll generate an outline .

You'll look at the outline , make any changes you want , or regenerate the whole thing , and tell it let's go , and then you'll end up with your completed article .

We'll keep the language as English .

For topic today , I think we'll do Apple Vision Pro .

For keywords description , you can either use comma separated keywords , or you can describe what you want the article to be about .

I'll use Apple , Vision Pro , and virtual reality .

Tell it how formal you want it to be , either casual , formal , or right in between .

And then for tone , you can pick between personable , confident , empathetic , engaging , witty , or direct .

We'll say personable , and then before you come hit generate down here , go back up to the top on the right side , decide what model you want to use .

You have GPT 4 .

It's more creative but slower .

Or you have Claude 2 , which is more comprehensive , long , and accurate according to LongShot .

I've had mixed results with both of these , so I can't necessarily say one is better than the other .

And then your mode , you have normal which is gonna use the AI , the GPT's pretrained data .

You have SERP which is gonna look at what the results are from Google .

It's particularly helpful if you need something more recent than what the AI was trained on .

It's also pretty good for accuracy .

File , you can upload a file , any file you have , and it will use that as its source of knowledge .

You can also point it to a URL , a specific page on a website .

If you wanna add URLs , you can it was that you could add up to 5 of them , but they're in the process of changing that to 10 .

So the limit will be 10 URLs that you can provide .

Or if you wanted to look at an entire domain and not specific URLs , not specific pages but just an entire website , you can use the domain mode .

For this article , since it's fairly recent and I know there's been plenty of talk about it , we'll use the cert mode .

Brand voice , we're gonna leave at default .

Brand voice allows you to upload files or provide it with URLs , and it will then learn your style .

Then when you come in to create something like this under brand voice , you'd select that style you created and it'll try and emulate it .

Okay .

So with everything selected , let's go ahead and hit generate .

We have some headlines to pick from .

If we like one of these headlines , we just pick it .

And if we don't , we can just say regenerate , and it'll make a new one .

If we like most of what's in in a headline but want to make some tweaks we can do that too .

We can just select the one that we kind of like and then make whatever changes we want to make here in the edit headline .

We're happy with the headline , we click next and now it gives us an outline for this article .

And you'll notice it's using different HTML tags , so h two and then under that , h threes .

Nicely broken up into sections .

If there's something we don't want in our article , we can just click the trash can that's , runs alongside that item .

We can also rearrange things .

If we wanna move where something is , we can just click and drag .

And if we want something to be at a different level , click on the tag , and it'll move through from h 2 , h 3 , h 4 .

We're We're gonna leave everything alone there .

If we don't like this outline at all , if there's just too much about this that we don't like and we wanted to start over , Just click regenerate , and it'll regenerate the outline .

We're happy with this outline , so we're gonna say generate blog .

Long shot refers to an article or a blog post as a blog .

It's a little weird to me .

The blog is the entire section of the website that contains blog posts or articles .

The single post , the single piece of content , that's a blog post or article , not a blog , but it it calls them blogs .

Now it's created the entire post or article , whatever you prefer to call it , and it has dropped us here into the long shot editor where we can make any changes that we wanna make to it .

It also provides us some information .

It tells us what the word count is , what the reading score is based on the Flesch Kincaid , and it tells us the semantic SEO score .

It also shows us the SEO keywords .

And if you have an SEMrush account , you can integrate that here with LongShot AI .

Now overall , the blog post here looks pretty good , but I would never publish something here without reading it first .

If you know the thing it's talking about very well , read it yourself .

Make sure it's accurate .

If you don't know the details or there are details in here that you're not sure of , then you gotta do some fact checking and research to make sure you're not putting garbage out there .

LongShot does give you a little help with this .

You can switch over from blog performance to claims and fact check , and you can say detect claims .

And now it's completed that process , and it is showing us what it believes our statement of facts within this article .

So we can click one of these and say fact check .

We can also click a bunch of those at the same time and have it go fact check multiple claims that it found .

I just did one to make it go a little faster .

And what did it come back with ?

Okay .

So the claim is Applevision Pro launched recently .

Factuality is true .

The statement is accurate based on evidence provided .

Announced on June 5th 23 .

Preorder sergeant , it became available on February 2nd in the United States , and then it found some sources for that .

So that's good stuff .

And , no , it does not just come back and always say that whatever it wrote the article with is accurate .

I have had it come back and say , no .

This is not accurate because and tells me why .

Seems like the claim detection is , a little bit on the conservative side , which is probably a good thing .

You also have the AI assistant right down here .

You can click on that and tell it what you want it to do .

When you're all finished with your article , blog post , or whatever it is , you can download it , you can export it , either via a link that you can share or any custom connection , or you can save it within your LongShot documents .

In addition to the Autoblog workflows , again , the one we used here was the pillar content , Longshot AI offers fact gpt .

This is more chatbot style but it's focused on generating factually accurate content and it provides you some examples of prompts that you might wanna use in here .

Compose a tweet about the benefits of meditation for mental health .

Write a short LinkedIn update about overcoming common work place challenges .

We'll try this one , summarize the latest developments in electric vehicle technology .

It just moves that example down to the prompt box , which you could type whatever you wanted here .

We're going to check the web search box , which ensures it'll search the Internet and not just rely on the most recent AI training .

It generates the response with inline citations and then gives links to those citations below .

We can take this response if we wanna work with it some more and add it to the editor , switch over to the advanced editor .

If we wanted more and wanted this paragraph to be a little bit longer , we could grab this text here , come down to our little AI chat box .

You have options to pick from here .

I'm gonna say expand content .

The robots go to work .

We look at that .

If that's what we wanna add , click it , and now it's replaced the original paragraph that was there .

The content planner , pretty nifty feature in Long shot AI .

You can simply enter a topic here and hit the arrow button , or you can add more context .

So maybe our topics , inexpensive meals , our audiences , people with large families , and our additional context is healthy meal ideas for busy moms on a budget .

We hit the go button , let the robots start thinking , and then what it's come up with is about 10 different ideas .

Any of these that we like and wanna work with , we can just bookmark .

If there's something that we wanna get into right now , we can click on it , pick our workflow , and start creating an outline .

LongShot AI has some other tools like fact check .

You drop the text in here , tell it what the topic is , and to fact check .

I just dropped a little bit of crazy text in here that has a whole lot of things wrong with it .

You just click the fact check button , and it starts detecting the claims .

It did a great job of finding everything in here that was a claim .

We'll just select the ones we want it to verify , and it tells us claim by claim whether each statement is accurate or not .

And if it's not , it gives us a correction , reasoning , and links where we can go find the evidence provided .

So they could not find anywhere where Tim Cook said the iPhone is the worst smartphone ever made , and it says it is not supported by the evidence provided .

So they could not find anywhere where Tim Cook said that , and I didn't think they would .

Another tool is the SEO checker .

You can type in text or select a document , and it will do some analysis on your SEO score .

The repurposer lets you take any document that you've already got in LongShot AI or you can paste in text .

You can turn it into a Twitter thread , a tweet , LinkedIn , promotional email , lead generation email , q and a executive summary .

So that's pretty handy .

BotShot allows you to create a no code knowledge base chatbot that you can put on your website or content management system .

So you just supply it with the training information you want it to have , whip it together , and you've got one of those handy dandy little chat bots for your website .

They've recently added this assets feature which allows you to store files , documents , information that you want LongShot to pull from when it's generating content .

Integrations , they also integrate with SEMrush , Hub Spot , Ghost , Medium , WordPress , and , of course , the Chrome extension .

The Chrome extension , while you're surfing the web , if you find something of interest , you can quickly mark it and get it into LongShot AI to start working with it .

Brand voice , when we talked about this earlier , this is where you just supply your text URL or upload a file and start to build the training on what your style is .

And then when it's generating content in the future , you can specify a brand voice that you want it to style after .

Pricing for LongShot AI starts at $19 a month if you pay annually .

That's for 1 user , 2,000 monthly credits , which is about a 100000 words .

Of course , the advanced AI models , the 50 templates , all the factuals , integrations , and unlimited projects and documents .

And this is the plan that I have .

The team plan is the next step up .

It includes everything in pro plus access for 5 users , 10,000 monthly credits , which is about 500,000 words , collaboration , analytics , and priority support .

They do offer a no credit card free trial , and that , of course , is how I started out and took it for a test drive to make sure it's what I wanted .

I don't recall exactly how many credits you get in the free trial , but I know that it was not restrictive .

It gave me plenty of room to try multiple different generations and make sure it was I was happy with the results .

There's a link to Longshot in the description after clicking my link , I may receive a small commission at no additional cost to you .

After clicking my link , I may receive a small commission at no additional cost to you , and I sincerely appreciate it .