Hey everyone, my name is Mike and I'm the developer.
On Treendly, you can search a topic or keyword in any region and immediately get an understanding of its trend.
You can also set up alerts so that you can monitor a keyword/topic and be alerted when it begins trending.
Treendly also curates trends. Getting in early on a trend is key these days and within a few minutes, you can find a topic that is worth exploring.
Under the hood, I also try to predict how a trend will perform in the near future by using Facebook Prophet and other similar technologies. Right now we are predicting using the last 5 years of data on any topic.
Sales/marketing guy here. I like the idea. If you want to get more signups, be clearer about what I get if I signup. There is a huge yellow banner on the trend page, telling me it's better to signup. Use some of that space to tell me why it's better and what registered users get.
Yes, we both find rising trends but we differ in so many ways.
First and foremost, we give our users the ability to monitor a keyword/topic over time. As a user, you can receive daily, weekly and monthly email alerts.
Secondly, you can search a keyword/topic in any country and what you search stays private to you so that you can build your own private library of trends.
Thirdly, each trend has so many related topics that we find by querying Google, Amazon, YouTube, etc that you can dive into.
And finally, we have a beta feature where the tool forecasts how a trend will likely perform in the next 6 months.
> Do you plan to display accounts/pages ranking the best for these queries?
No, but I do have another tool in the making for Amazon self publishers. I don't know if you are in that target market.
The interest over time graph uses spline interpolation to connect the datapoints, while that might look prettier than straight lines, it is something that is strongly discouraged in a scientific / data analysis context, because the curving line does not carry meaningful information.
We both find rising trends but we differ in so many ways.
First and foremost, we are an actual tool you can use, not a simple newsletter.
We give our users the ability to monitor a keyword/topic over time. As a user, you can receive daily, weekly and monthly email alerts.
Secondly, you can search a keyword/topic in any country and what you search stays private to you so that you can build your own private library of trends.
Thirdly, each trend has so many related topics that we find by querying Google, Amazon, YouTube, etc that you can dive into.
And finally, we have a beta feature where the tool forecasts how a trend will likely perform in the next 6 months.
The way our system works (might be similar - but Treendly looks like integrations?) is looking solely at comments. Basically, all that’s sent to our API are messages (author, comment/message, and some unique identifier). From there we use topic modeling to determine related topics, track sentiment, trends, author expertise, etc
So this is a bit like https://explodingtopics.com but with a different angle to find the categories and trends? I see no "world/global" option, is that on purpose?
I do like how the data is presented in the details page. In exploding topics you have to leave the site to see the same on gtrends.
Yes, we both find rising trends but we differ in so many ways.
First and foremost, we give our users the ability to monitor a keyword/topic over time. As a user, you can receive daily, weekly and monthly email alerts.
Secondly, you can search a keyword/topic in any country and what you search stays private to you so that you can build your own private library of trends.
Thirdly, each trend has so many related topics that we find by querying Google, Amazon, YouTube, etc that you can dive into.
And finally, we have a beta feature where the tool forecasts how a trend will likely perform in the next 6 months.
Thanks for explaining. So Exploding topics as a Service and with additional features to make it a business. Can be interesting depending on how accurate your trend spotting are and how well the prediction beta feature works. Keeping track of new technology trends and competitors offerings is a healthy amount of work in any business. If a service could already take out the initial search part that could be value.
Exploding Topics has a one huge advantage. They don't require account to search and test. Maybe leave basic features for everyone to use without account?
You do require an account to search because I'd like to understand who is searching and protect me a little bit better against bots. Also, searching has a much higher computing cost for me under the hood.
Hope it explains why I did it this way.
Thanks for the comment!
Noah here from https://meetglimpse.com. It looks like we're tackling a similar problem of surfacing trends, just a bit differently. I'd be super curious to hear more about how you're filtering through the noise. Fun challenge for sure!
Yep, we are. Well, let me tell you - there's a lot of noise xD Most of the time it's about finding enough people talking about something, but not too much.
Really cool - I'm interested in this. Despite similar services - I like your opportunity indicator. Love the search functionality idea - but hit a bug so it took a couple of tries to get it to search.
This has potential and I hope you keep working on it. I would consider paying for it down the road if it's more mature. Also hit the free limits very quickly.
Thanks for the feedback. We have no limits on all tiers regarding searches: you can search all the keywords/topic you want, for free.
The only limit we have on the free plan is the fact that you can track only 1 trend, meaning that you can receive alerts only for one trend.
The other limit we have on the free plan is regarding curated trends, because these are trends that we personally curate and are rising and I feel like $9/m is not a lot of money to get access to those opportunities.
Are you referring to any of these limits in particular?
I tried it out and it doesn't seem to support unicode well.
I searched for "学校" (Japanese: gakkou, school) in Japan and the results page was headed with "%E5%AD%A6%E6%A0%A1".
Additionally, the actual results weren't great; it found a 2014 movie and a handful of Amazon/Youtube searches that aren't accurate. Some of the Youtube search results weren't even Japanese, but Chinese.
Just in case I'd weirdly picked a bad word (dubious since school closings etc. are all over the news), I also tried パンくん (Japanese: pan-kun, the name of a chimpanzee who was famous for being with Shimura Ken, a recently deceased comedian), which is currently trending on Twitter.
Kinda tangential: but what’s with startups naming their companies ending in vowels since 2010- especially ending in -ly? It kinda irks me but I can’t put my finger on it.
It started with registering .ly domains (think bit.ly). Then it stuck for others who weren’t using a .ly domain. Same trend occurred for .io domains. There was also that trend of removing vowels.
Good observation. I personally do not spend a lot of time finding a domain, or a logo, or setting up the perfect onboarding sequence, etc..
The truth is the business can work out with whatever domain. I have another SaaS (in the e-com space) that started with the domain angage.net and now we have usecart.com - maybe if we grow we'll get cart.com
But, when you start, you don't need 99% of the things you'd think you need - and a good domain is one of them.
When I'm starting a SaaS I focus on its core features and making it profitable.
Tried adding a monitor for a trend with two words, got lots of matches for just one of the words that is generic. Tried then again but putting the two words in quotes, expecting it'll match the entire query instead of each one of them individually, but then encountered bug that quotes turn into %22 and trend report has no data.
Otherwise, looks like a useful service! Thanks for sharing
A suggestion! While logged in, when you want to track a trend, it isn't clear how I can track it after I search for it.
Instead of having a dropdown list menu for choosing frequency, it would be nice to spread them out. Like a normal list, or 4 types of buttons. Or you can also add a plus button which might bring a popup asking the frequency type too.
Thanks man! Let me tell you, it's not always like that! It's a roller coaster But, why not be positive? I get to work on something valuable and people are giving me great feedback!
And, there are definitely people out there who are in a worse situation than me, especially now with the COVID emergency.
Little CSS bug in the nav bar. 770px to 1200px screen width causes a horizontal scrollbar to appear. The problematic rule is coming from bulma.css, so it might be easiest to just set div.nav-left { overflow: visible; }, that fixed it for me.
On the website you are showing 7 trends for free. So I created an account in the assumption that I can see more for free. But when I logged in I can only see 6 for free and then you are trying to upsell me. That doesn't make sense.
Hey Mike, this looks really interesting -- nice work! We're going to try it out to follow 'referral software' topics. Any tips/advice you can give us? I've already created a new Track for that keyword set.
I have few questions:
1/ How do you determine keywords/topics. Do you use any kind of clustering, or do you manually enter the labels?
2/ How do you deal with different languages?
3/ How do you deal with Spams.
2 - We are still in the process of dealing with special characters in our search. But, it's coming along! Right now you can search in pretty much all languages that do not require special chars (English, Italian, etc..).
The project looks very good. It would be nice a feature to modify the time range, for instance on the Covid-19 trends, Zoom is -50% over the last 30 days. It would be interesting to know the change over 2 or 3 months.
Thanks! Yes, range modifier is coming, but also please take note that -50% is the change regarding the last 30 days compared to the previous 30 days.
This year the month over month growth in the US was 3.52% which is very good. And, the average interest was only 4.85% which tells me there's still a lot of margin for growth.
It seems to be incorrect in regards to zoom. Zoom interest rose 30ish percent from March 3 until today. During that time there was a 20% increase in uaers. The interest level should match levels somewhat.
Thanks for the comment.
Again, here we are discussing two different things.
Interest for Zoom in the last 30 days has definitely grew and we reflect that. Take a look at https://treendly.com/trend/zoom under the "This week & month" tab.
The minus 50% is referring to the interest level of the last 30 days compared to the interest level in the previous 30 days.
Very cool. I guess as you grow, you could get more partners and the data will keep getting more and more accurate. I wish big guys like Amazon released aggregated sales data.
How do you get access to Amazon searches, google searches etc? do you have to pay for those?
I've always wondered about building a career with scraping, but never got around to it. One reason is the legality of scraping - big sites like Craigslist have gone after (and shut down) smaller projects. I believe priceonomics (YC company) built a big business with scraping.
According to the " Interest over time by search volume " graph over the last 5 years, interest/what-your-measuring seems to have decreased.
I'm my experience, but I'm probably in a bubble, more and more people are talking, interested in and using JavaScript, in the last 5 years, but again, probably biased as I'm a web developer.
Is the graph adjusted to the number of people on the planet or some other metric? Or why do you think it shows the data it's showing?
Hello, at this point this question is hard to answer.
I've identified a lot of real customer profiles (as I call them) that can use the product and receive value.
Some of those are: e-commerces, marketers, affiliates, investors, SEO experts. If you are curious, you can read more about each of them on the homepage, in the "use cases" section.
I don't know which one of these customer profiles will benefit the most at this point. It's too soon, but time will tell.
Nice. A lot of people have said that tools like Glimpse and ExplodingTopics are cool and fascinating but they find it hard to find a concrete use case for them. once you find that target audience, I recommend really nailing down that value proposition. Rather than rely on people trying to find a job to be done for it.
Our data relies on Google and it is based on what consumers are searching for. Specifically, it relies on Google Trends, Google News, Google searches, Amazon searches, Twitter searches and YouTube searches. We also have a partner for e-commerce data (usecart.com)
I'm not too worried. Everything that is publicly available and printed to a screen, I can take. So unless they completely shut down the service, I should be good =)
Hey there, URL is optional. You are not the first one with this problem but all I'm doing is using a HTML5 input with its own validation (type=url) so not sure what's happening there.
First and foremost, we give our users the ability to monitor a keyword/topic over time. As a user, you can receive daily, weekly and monthly email alerts.
Secondly, you can search a keyword/topic in any country and what you search stays private to you so that you can build your own private library of trends.
Thirdly, each trend has so many related topics that we find by querying Google, Amazon, YouTube, etc that you can dive into.
And finally, we have a beta feature where the tool forecasts how a trend will likely perform in the next 6 months.
Btw, I'm pretty sure that I originally launched this tool before them. I'm just revamping it now =)
On Treendly, you can search a topic or keyword in any region and immediately get an understanding of its trend.
You can also set up alerts so that you can monitor a keyword/topic and be alerted when it begins trending.
Treendly also curates trends. Getting in early on a trend is key these days and within a few minutes, you can find a topic that is worth exploring.
Under the hood, I also try to predict how a trend will perform in the near future by using Facebook Prophet and other similar technologies. Right now we are predicting using the last 5 years of data on any topic.
Hope you like it!