Subscribe
Sign in
Magic Spreadsheet
p/magic-spreadsheetSuperpowers for every Google Sheets user!
Start new thread
Garry Tan

Magic Spreadsheet - Human workforce and AI plugged into your Google Spreadsheets

by

Magic Spreadsheet is the best way to automatically gather information from the web into a google sheet, using a real-time human workforce boosted by AI engines

Add a comment

Replies

Best
Sagiv Malihi
Maker
@garrytan - thanks for hunting us! Hello producthunt! Sagiv here. Co-founder of HEAT, along with @yig, We built the Magic Spreadsheet to make HEAT’s hybrid workforce of humans and AIs accessible to anyone with a spreadsheet. The possibilities are truly endless - your spreadsheet literally turns into a dashboard where you can send tasks to thousands of workers, in parallel, to get you any piece of information that you want (as long as it’s publicly available out there…) In private beta, we saw all sorts of crazy ways to use the spreadsheet that we didn’t even think about - like editing images, and making phone calls! We’d love to hear your feedback, so we’ll be hanging around here today! Cheers, — Sagiv
Ernest Semerda
@sagivmalihi @garrytan @yig video looks super cool! Nice work guys.. I have questions ;-) is Magic Spreadsheet just 1 application of HEAT? How do you guarantee quality of the output? ie. how can i trust it if a turk was involved or if it's not publicly accessible data? Is there any regression models to predict say what an email address could be based on historical data? And do you have enough data to make it statistically significant? Finally, what does HEAT stand for? Cheers!
Sagiv Malihi
Maker
@ernestsemerda wow, lots of interest! I'll try to answer briefly: Yes, Magic Spreadsheet is currently the easiest way to use HEAT. Another way is to use the HEAT API directly. We're working hard to make it as easy as possible to use us! Quality is guaranteed by a variety of methods, but the short answer is that all workers are constantly measured (accuracy & response time) - and they have an incentive to keep the scores VERY high. Non-publicly accessible data - I'm not sure what you mean. HEAT is great at doing anything a human can do, but if a piece of information is not available to human workers, it's a problem ;) (unless it can be guessed easily with very very high confidence). Our automated models (usually regarded to as 'AIs') currently 'learn' from our human workforce. So for data-collection tasks (such as ones that are popular on Magic Spreadsheet) - they just 'learn' where to get information (according to specific inputs).
Ernest Semerda
@sagivmalihi Cheers for a detailed response :-) So it sounds like it's still more human than machine and you have KPIs in place to make sure the humans do an outstanding job? The part I'm trying to understand is how HEAT will learn unstructured data or something that it doesn't know where the source is. Are you using backpropagation in the DNN to achieve any of it? And what does HEAT stand for? :-) It's an awesome acronym that has me intrigued :D Happy to chat offline if it's easier.
Pascal Andy
@sagivmalihi @garrytan @yig what is the site for heat ?
Garry Tan
Hunter
I met Yig recently and was impressed by this demo — it's a browser plugin that sprinkles itself into Google Sheets and enables a large set of crowdworkers to do your bidding directly inside a spreadsheet. It's useful for filling in all sorts of knowledge work quickly, e.g. finding addresses, websites, email addresses, or even jobs like describing people or objects. It's one of the coolest demos I've seen this year.
Menachem Pritzker
this is so cool. I assume you're using mturk to power this? btw, the pricing page does not work. I'm very curious to understand how you're pricing this
Yigael Berger
Maker
@mdavep Oh no, it's not mturk, this can never work with an external crowd. It all relies on HEAT (heatintelligence.com), the platform we've been building for a year+
Yigael Berger
Maker
@mdavep pricing is 15 cents per cell. That's the basic pricing plan, that should cover standard type of info tasks.
Ernest Semerda
@yig @mdavep so heat intelligence is basically supervised ML with turks? Can you please clear up where the Research piece (video example) fits into ML? Where do the regression models come into play here? or classification problems? Is the Magic Spreadsheet just one application of HEAT?
Menachem Pritzker
@yig how many humans work on the platform? Do you give them any training? What makes them more qualified than mTurkers?
Salil Sethi
This looks like a great tool for lead generation. What is the pricing to use this. The pricing tab did not work on the top.
Yigael Berger
Maker
@salil_sethi pricing is 15 cents per cell. That's the basic pricing plan, that should cover standard type of info tasks. For "heavier" cells that take longer time to fill than the norm, we will create more pricing plans that fit those cases better.
Kumar Thangudu
Very curious about the API of human workers you've built. Would you say it's better than ScaleAPI.com specifically for data entry tasks into google sheets?
Yigael Berger
Maker
@datarade Scale is a competitor we respect (@alexandrw I'd be happy to meet one day). Our API is great for numerous things not just data collection/entry. Customers use us for analyzing articles, photoshoping images on the fly, tier 1 support, and lots more. The two points we especially excel at are 1. The reliable, high quality, fast responding human workforce we run 2. Seemless transition to AI
Kumar Thangudu
@yig @alexandrw when you say "AI" what do you mean? I have an operations research background, and when web developers say AI, I assume they're just using basic prob/stat and calling it AI or is there more to it? Is it learning modalities of entered data and making suggestions?
Yigael Berger
Maker
@datarade There is a long roadmap to that, part of which I can share in exchange for a cup of joe :)
Pablo Fernández
I love the idea. So many times I created google spreadsheets, then divided them up, sent them to various people, got the results, merged it. Such a pain! If this works well it might become an indispensable tool for the small business.
Nick Julia
Looks cool! But I'm not sure I understand how the pricing translates... In the finding Linkedin example - ballpark how much would it be per LI profile? Or let's say I want to do email prospecting. Is that a potential use case? And again how would that translate to pricing - Especially if a human has to get involved?
Sagiv Malihi
Maker
@nickajulia thanks! pricing for standard information retrieval tasks is 15 cents per cell (yes, even if a human has to get involved!)
Nick Julia
@sagivmalihi 👍 I'll try it out soon
Yigael Berger
Maker
@nickajulia The basic pricing is set fixed at 15 cents per spreadsheet cell. This should cover tasks like find LI profile and similar. Email prospecting is definitely a use case, and as more people use this, the faster and better the results will be because email prospecting is not just searching rocketreach or hunter or guessing name patterns. So when you will run use cases that will require more than 1 minute to complete (per cell) you will have more pricing plans to choose from.
Steve Brigden
Loved the idea, and had an immediate need - so I dived straight in and gave it a go. Sadly, it hasn't delivered. It thought my requirement was quite simple - I gave it a list of 25 names and the company in which they worked, and asked for a search for the country they live in, and their job title. Both things that are easily searched by hand on LinkedIn. It's just that I have 1,500 names to get through, so if this worked it would have been awesome. But it didn't, so it isn't..... Entered my request (25 names requiring two cells - so 50 requests) at 09:14hrs UK (timed it as I was interested to see and report back on how quickly it worked...). I got two requests back by 09:50hrs - and one of them was wrong (Job Title in the Country field!). Since then, nothing. So here I am at 14:19hrs UK and I still only have the same two results. Maybe all of the human people searching behind this are West Coast and so are still in bed? Oh yes, it cost me $0.48 for the two cells. If my maths is correct, that's more than $0.15 per cell.... I'm hoping that @yig or @sagivmalihi have got some wise words, as I'd still really love this to actually work. Now it looks like I'm going to have to go through my 1,500 contacts by hand....... :-(
Steve Brigden
@yig @sagivmalihi Great news! Thanks to @Yig everything is now working perfectly. Not sure what magic he did behind the scenes, but all of my data was updated within a couple of minutes of his intervention. Quality of data is pretty good too, so I am now fully confident to try this with the 1,500 contacts that need sorting out. Oh, and he sorted the pricing discrepancy out too. If there's one thing I like better than a great product, it's a great product backed up by great customer service from the founding team. Thank @Yig!
Dror Bren
This is amazing! Good luck guys
Misbah Ashraf
Looks very interesting @yig. I'd love to talk in detail sometime.
Yigael Berger
Maker
@misbahspeaks Anytime. My email is my name at heatintelligence.com :)
Nissim Lehyani
Magic spreadsheet is just awesome! You got to try it to understand
Ouriel Ohayon
sounds like a reversed engineered Clearbit?
Sagiv Malihi
Maker
@ourielohayon - I like to think about it as a more general thing. You collect any information that is publicly available (not necessarily for the purpose of enriching user information). On the other side - my guess is Clearbit have their own 'secret data sources' (but that's just a guess)
Niv Dror
Super Cool!
Rajath DM
Very interesting. This can disrupt a lot of lead generation tools in like Hunter.io, Aeroleads etc. in the future.
David West
2 words: How? & Wow! This looks incredible guys. I'm totally going to be using this soon! 🤓
Sameer Sontakey
This looks super 🔥. Definitely first thought of lead-gen as an application for this.
Sagiv Malihi
Maker
@sameer_sontakey thanks! lead-gen / lead-enrichment is definitely a classic use-case for this
Nick
I wonder if you can do something similar for Sketch.
Sagiv Malihi
Maker
@hurleyharper - interesting concept! we're already toying with the option of getting workers with very specific skills (e.g. - photoshop) and have them perform short & simple tasks seamlessly
David Carpe
how does this compare to mechanical turk HITs for similar cell-based pricing (data points like contact profiles)? seems a thousand times easier to use...
Sagiv Malihi
Maker
@passingnotes - indeed, MTurk is a great concept, but hard to use (a lot of overhead managing the process). Our mission in HEAT is to make this as easy as possible to tap intelligence (either human or artificial).
Ari Shpanya
@sagivmalihi Pretty cool. are you thinking of adding a layer that would 'automate' LInkedIn connects and inmails ? It would require a sales team using : 1) CRM with LinkedIn Profile URLs for their prospects 2) Outreach tool which uses tasks that can activate triggers to push the task of 'connect' or 'inmail' to outsourced workforce on the day the task is due 3) LinkedIn profile access for the user (lastpass) 4) Measurement on our end to show connect rate/ROI of using the solution
Sagiv Malihi
Maker
@arie_shpanya - this is a very interesting concept. We're definitely thinking about targeting CRMs (such as salesforce) for the next Magic plugin. I would love to discuss these ideas offline with you - shoot me an email at (my name) @heatintelligence.com
Ari Shpanya
@sagivmalihi @heatintelligence send, also check scaleapi
Avi Parshan
I was waiting for soemthing like this