Hey guys, I'm building a Superhuman for Excel, pretty much a premium version of Excel for power users. To that end, I need to ensure that I understand what people's pain-points around Excel are, and I'm hoping to hear from you about what yours are!
One thing that I want to point out is that Excel is used for two main things: calculating, and storing (with a bunch of subsets in reach of these two), and a Superhuman for Excel is unlikely to do something like airtable.com, who focused only on the storage side of things!
Thanks in advance :)
Hi Namraj! I am an equity research analyst covering multiple sectors so excel is close to water and air to me as I have to compile large data sets on it, 24/7. I think one of the reasons that people complain about excel is that they are not familiar with the functionalities it offers...as an avid user, I think what I expect from a superhuman excel are:
1) Easy-to-deploy macros for shortcut creations
2) Faster speed to load large financial models
TEMPLATES is a big one for me. I spent a decent portion of my time recycling existing sheets.
Its funny that everything i ever do in excel probably millions of people have done it before and have done it better .
Like imagine i want to generate a discounted cash flow model, i hit "/" then discounted cash flow then using a few extra questions the superhuman excel generates the skeleton of the shit i need
Strong keyboard shortcut focus, and a CMD+K style command palette search tool to find formulas, chart options, data sources, and more.
Spreadsheets are actually fairly hard to navigate with only the keyboard; off the top of my head, I wouldn't know how to switch to a different sheet's tab in Google Sheets without clicking on the tab. And discovering formulas can often be difficult if you don't know the formula name to type it in autocomplete. Same with charts: Typically I want a specific chart where I know its name, but don't want to click through menus to find it visually. A command palette could speed up discovery.
And while Google Sheets already has some deeper web data integrations with say being able to call Google Finance and Google Translate via functions (with more via add-ons from 3rd parties), could be interesting to see more data sources, again searchable via a command palette.
Very exciting, can't wait to try out your app once it's ready for testing!
Quick powerful filters in 1-2 mouse clicks like Airtable!
I agree with @kimpma126 and @maguay: there are three fundamental features you need to nail:
1) easy to use keyboard shortcuts. This should accomplish the top ~10 most common tasks (navigating, creating simple formulas, etc)
2) a cmd+k style action bar with all the options to keep your hands on the keyboard.
3) Insanely great and fast to use macros. I think of these like "snippets" in the Superhuman world.
Function parity with Excel (and probably G Sheets as well).
When I moved from Win to Mac some 10 years ago, I lost all my shortcuts and was unproductive for a week. I'd love for you to make good use of the shortcut menu SH was introducing. It's popping up in a multitude of applications (saw it in Pitch a Power Point replacement tool last week). The ability to not use the mouse except for in rare cases would be the thing for me.
That said, the difficult thing will probably not be figuring the new UI, but rather building the spreadsheet engine and making sure the outcome of functions from YOUR tool match those that Excel delivers.
There are probably some things in the methods available in Pandas dataframes that would be useful via keyboard shortcuts within spreadsheets.
describe and value_counts come to mind as examples that are single steps in python that require multiple clicks to answer in Google Sheets (I haven’t used Excel on the desktop in several years so this might be easier there).
Another thought: One thing I regularly do in Sheets is a quick validation/reconciliation workflow that looks like this:
- import source of truth data into a tab
- import new thing (that I want to validate) into another tab
- count the number of objects in source of truth that are present in validation tab
- then filter to show missing records as a launching point for troubleshooting the process that generates the second set of data
“Easy-to-deploy macros for shortcut creation” as @kimpmal126 mentions would be an excellent tool — this and similarly repetitive workflows are just different enough that writing/managing scripts/macros is a little too cumbersome, but lightweight shortcut creation would be helpful.
A community where people contribute macros, and a command palette where you could search through macros and use them in a keystroke ala Sublime Text Package Control would be super useful.
Would people pay for a Superhuman for Google Sheets, vs a Superhuman for Excel?
People are already paying for add-ons for both tools, but I think that it is more interesting to make a Superhuman for Sheets, as the barrier to entry to use the base tool is inexistent.
Moreover, Google sheets is still growing and probably will continue to do so as the need for collaboration increases, and sheets is way better at that than Excel.
I would pay for a tool that i can use on my machine with all superpowers it gives me but is live time syncing to a google sheet. so that others that dont feel like learning new tool and i need to collaborate with would not be a barrier for me. Can even smth like "created using X" automatically written in first row cell
Oh you're right, if it worked with Google Sheets so collaborators could still just use Google Sheets if they didn't want to use the new product, that'd make it instantly easier to adopt.
Something that made Google Sheets work better on desktop as a native (or native-seeming) app would be a huge advantage for existing Sheets users, much like the main value add of many email clients for Gmail being that they make email reliable to use offline.
Would love to see what you put together here. I'm a daily user of excel in the financial services space.
Here are some things I'd like excel to do better:
-better guide to show how new visualization and BI tools can be used natively in Excel. Power BI is something not nearly enough excel users are onboarded with and likely don't even know where to begin with it.
-better integration between separate excel sheets, or at least easier to trace/track, especially if files are moved around
-easier way to format text and numbers. It's not straightforward when doing custom formatting (with all the spaces, semi colons, dashes etc)
Hi Namra! I'm a Product Manager and what you're building sounds really cool. I think a Superhuman for anything really needs to optimize for the end goal of that user. So thinking about it, the end goal for a spreadsheet experience could be: saving time, building error-less spreadsheets, building a spreadsheet with transfers in mind etc.
Some of the things I would love to see are:
1) An easy way for documentation, something like Roam's graph method to see the interlinking between all your spreadsheets
2) Some cmd+k type of short cuts that really ease the discovery to the different things the sheet allows and speeds up tasks
3) Superhuman snippets for spreadsheets would be cool too
I also think a nice modeling module would be super cool as well, so like really easy to use models and visualizations.
This is a really cool idea, I'd love to try it out once it's ready!
I'm going to be a contrarian and say I think Superhuman for Excel seems like an attempt to cash in on their success from a marketing standpoint but is poorly-fitted as a design mantra. Apart from my ongoing study of design, I am old enough to have seem multiple iterations of pre-GUI command-key apps and what worked and didn't. I've lived through several generations of the same idea being re-invented, not always as an improved version.
Superhuman addresses the pain point of needing to churn through email fast. They deliver a relatively small set of features through commands and save people time being being fast.
You can say you're doing a better Excel but I don't think there's an equivalent pain point of speed that can be addressed by simply providing keyboard shortcuts - Excel already does that for power users.
Mis-identifying AirTable as focus only on the storage side of things indicates you either don't understand why it's been successful or are just trying to throw out some slander to make your product look better. I suggest anyone interested in why AirTable has gone so well read https://usefyi.com/airtable-history/ and consider the many ways it provides for people to visualise information.
Seriously, dive deep into that article. Also consider that their key inspiration point of how people use Excel (lists of lists) is exactly the same one behind Trello, due to Joel Spolsky's experience on the Excel team.
BTW I'm now a massive fan of https://www.causal.app/ for modeling because it makes it trivially easy to play with ranges of input values and generate models.
Despite my negativity above, I'd encourage you to go for it but be clear about what you're doing. There should at least be a couple of good articles in discussing the pain points you uncover.
If I can see Superhuman for X as a red flag, I'm sure many a VC does the same thing!
I'm super curious about what you said about "list of lists". I'm not sure I understand what you mean by that. Would you mind elaborating?
Btw, the FYI article on Airtable is great indeed!
Read the original, I'm not going to do bettter ;-)
I love that article. "Most people just used Excel to make lists" is such a great insight, the core idea behind the modern startup advice that every unique spreadsheet use case could be the basis for a new dedicated SaaS app. And yet, that's the case with almost any horizontal app—including Trello now, where one could easily imagine "Trello for X" being a viable if small product.
(I'm super busy) but sorry, realised I may need to add more context for people. Long before he started StackOverflow and dreamed up Trello, Joel wrote this amazing column "Joel on Software".
It is full of wisdom on software development and business. He's still mostly right. You could have an entire discussion forum just going back and picking out nuggets of wisdom.
Thanks so much for sharing Andy, super interesting and had not come across this article in the past.
"The great horizontal killer applications are actually just fancy data structures." Made me think of Roam which main innovation is in the data structure as well.
We have 15k newsletter subscribers, and have around ~2k of them in a Slack group. We're starting to encounter issues in terms of community management - specifically, it's hard to pin content like c...
Google lets you subscribe to a calendar using a URL - although when using an Outlook 365 Calendar link, events are copied over once, and then the syncing stops. This seems to be a relatively new is...
Three major considerations I have been using to evaluate the plethora of options available: 1. Effortless/non-intrusive: It shouldn't feel like a video call 2. Price: As this app would be complime...
I like this idea