Tildes exclusive: the Ethoscore Project
Hi Tilderiños,
This post is long enough for a TLDR: I made this thing, it's a platform for conscious consumerism. I'm launching the beta on Tildes. It's far from complete, expect imperfect scores in many cases. Scoring is a work in progress that's part of the platform but not the sole focus. It's invite only, and on Tildes only, see my top level reply below or DM me for an invite.
To start, I'm a long time member who posts enough that at least some of you would recognize my username. I made a new account for this project because I don't want to think about being a representative on my main account.
Early on during development I decided Tildes would be the perfect place to launch a private beta because Tildoes are some of the best people on the internet in my experience. And also fairly diverse by internet standards. Aside from a handful of friends and family you'll be the first to see it.
I should give you a little backstory. A couple of years ago I lost my income. Fortunately we were in a place where we could mostly afford to live without it for a while, so I had some time to figure out what I wanted to do next. For context I have over 30 years of experience in software development. Similar for marketing, brand strategy, business development, mostly but not exclusively in the context of the internet. I'm not quite as old as that sounds, I was still a minor when I starting building things on the internet, back when almost no one really knew what the internet was or what to do with it. In short, I'm comfortably fluent in all the languages of the internet.
I've built all sorts of different types of applications and businesses, mostly commercial. This time I wanted to build something that would have a positive impact and I wanted that to be the point, rather than a side effect. I spent most of a year researching the feasibility of a number of ideas before I landed on this one.
I've been on the internet long enough to see a lot of well meaning ideas fall flat. At this point I have no idea if this is another one of those. I know the concept is solid, but that doesn't mean it will inspire people. Or more specifically in this case, be a platform people can use to inspire each other. This is complicated by my intention to be as uncynical as possible about the project. I know all sorts of tricks for capturing attention and creating hype, I know how to monetize an app and sell products. But all of that feels out of place to me if the goal in uncynical. No doubt Tilders could have a long conversation here about the balance between good intentions and practical reality. I'd have lot to contribute.
The Ethoscore Project is an attempt to make the entirely unsexy concept of conscious consumerism easy and accessible. I genuinely believe it's a viable way to make the world a little better. It includes things like company scores and community lists (including boycott lists) and accepts community contributions and corrections for every company data point including scoring. I won't make a wall of text about all the functionality and features here, you can find out more on the site.
The project is currently self funded, which will hopefully be able to last long enough to see if the idea has potential. For now, funding is a problem for later.
The roadmap is long and wide. There are so many things I want to add, at least some of which aren't terrible ideas, but at this point development has been going on for over a year and I think it's time to see what strangers think. The project's 50k lines of code took time but the larger part of development has been company research, there are currently over 15,000 brands and companies covered, which is really just a start.
You might imagine that researching the international conglomerates that own most of the brands we see would be depressing in a capitalist dystopia sort of way and you wouldn't be wrong. But there's also more good news that I expected. There are big, successful companies doing the right thing when they don't have to. More than you'd expect. And there's evidence that their practices are inspiring other companies to do the same. I don't know if it's outpacing the general capitalist race to the bottom but it's not nothing! I see reasons for optimism.
I'm hoping some of you will be kind enough to take a look. I welcome all questions, contributions and feedback. The company database and scoring are far from complete, my goal for the beta was to have enough coverage that scanning products at a big box grocery store would result in more hits than misses. I think scoring is necessary but, to me, community lists are more interesting and have more potential to be useful. At this point there are going to be a lot of bugs, imperfections and companies with scores that seem off. I want to hear about all the issues you encounter. I'm more excited about your impressions and ideas though. How does it feel to use it? Does it make sense? Is it low friction? As a user is there something important missing that you'd expect to be there? What works well? Would you do anything differently?
There aren't smartphone apps yet, I want the codebase closer to complete before I start on the joys of maintaining two more codebases and dealing with the app stores. You can install it as a PWA (web app) on your phone if you like, which mostly just makes it quicker to access. I definitely recommend using your phone if you want to play with the barcode scanner.
Throughout the beta the app will be invite only and available only to Tildes users and people I personally invite. See top level comment below for an invite.
I snagged one of the 3 invites and my initial impression that I am skeptical of the methodology used to generate the scores and feel it fails to capture company impact adequately.
The example I first noticed was Microsoft scoring a 76 despite constantly being in trouble for openly agreeing to lend the use of their data centers to regimes commiting genocide and having subsidiaries who poor track records for treating workers well I.e Activision-Blizzard, among many other issues.
I understand that a project like this is a difficult to scale but I can't help but feel the site is misleading in certain cases.
That being said I admire the idea behind this project and want to see it grow. The feature I would be most interested in is being able to drill down into specific products. I want to be able to see the ethics of using a Google Pixel phone but not have YouTube affect the score for example. Large companies are effectively composed of subsections that would ideally be scored separately.
Kind of reminds me of The Good Place where nobody has entered heaven since the Industrial Revolution made every action slightly immoral.
Just added this to my "to watch" short list, based entirely on this description. Thanx.
Just forget you read his comment before taking the plunge. It's a phenomenal show but that comment is a somewhat large spoiler.
What comment? ( ;-P )
I've watched it twice through and found it phenomenal both times. I think you're in for a treat!
You're right that Microsoft's score is off. It doesn't have full scoring impacts yet, which is true of many companies in the database right now. In those cases the score defaults to scoring data aggregated from various sources. The app accepts community contributed score impacts! I understand if you don't have the time or inclination but you brought up some good candidates for new impacts.
It would be interesting to be able to separate Pixel from YouTube. That's something that could be possible with community support, but it would need to be a fairly large community.
My thinking at this point is that, using your example, if Google is doing shady things at YouTube, it should ideally impact their profits across their whole ecosystem, Pixel included.
Thanks for the feedback!
That's a fair point. Allow me to elaborate on my reasoning. In order to exert market pressure on a company through coordinated purchasing decisions, potential alternative products and services must be economically viable for both producers and consumers.
By looking at a company as a whole this becomes nearly impossible to achieve (good luck trying to replace every single way in you interact with Google). However if you can subdivide the problem it becomes more manageable. The YouTube and Google Pixel example was arbitrary. I wanted to show a place where viable alternatives exist (Pixel phone) and where they do not exist (YouTube).
In a scenario where every individual vertical and brand of a company is scored independently, doesn't that allow a big company to silo bad behavior in certain areas without paying for it in their overall reputation?
I suppose you could do both, score everything in the ultimate parent and also score all individual brands. It would be amazing to get to the point where there's a big and engaged enough community to make that possible.
As far as alternatives go, if there's no alternative to YouTube (not sure I agree there) then there's no consequence for anything YouTube does unless people decide not to buy, or interact with, other Google products in response. To me that makes a strong case for overall scoring.
By the way, I think for a lot of people de-googling is totally possible, even if it's not a perfect 100% break.
I think having both approaches makes sense is possible, especially if we start looking into the enormous conglomerates of FMCGs like Unilever and Nestlé.
For example, maybe there’s a fairtrade, all green energy, responsibly sourced chocolate or coffee or something, but if they’re also willing to privatise and cut off the water supply of a third world region, any individual might have their own way of weighting those two sides of consideration.
The way it works now is that brand and subsidiary scores map to the ultimate parent. So both do get scored at the parent level, there's just no individual score on a brand.
Indeed! Don't let all the scoring conversation give you the wrong idea, that's a big part of the goal.
Your example is perfect, though maybe a little unlikely since fair trade certification includes water management! Sorry for being pedantic, I've been spending a lot of time reading about certifications and standards.
The way the system would work for your example:
Overall score is broken into 3 metrics: Environment, Human Rights and Ethics. Your hypothetical company would probably have high a environment score and a low human rights score, which would lead you to look at their human rights impacts and read about the water rights issue. If you didn't like that you might add them to your Avoid list. Thereafter that company and every brand they own would show a warning badge when they came up in searches or from barcode scans.
Along with that, there are tags (fair trade already exists, green energy doesn't yet) you can use to filter based on broader behaviors. So you could find all the Fair Trade Certified companies and then figure out which ones try to pillage as minimally as possible while sourcing their chocolate. Then you could put them on a list and share it so other people can buy pillage free chocolate too.
Assuming someone hadn't already created that list, in which case you could skip the research and just follow it.
Think of a component starting an eco-friendly/socially responsible brand. Sure, you can call it green washing, but isn't it still useful to be able to tell a company "I won't buy your products...but I will if they're responsibly made"?
I would say it's probably best to focus on the company as a whole first instead of dividing your focus when you don't have a product that you consider particularly accurate. That's where a roadmap is encouraging to users.
Not sure I understand. Yes, being able to tell a company, via spending, that "we won't buy your products unless they're responsibly made" is a big part of the point.
I'm not saying I don't consider the platform accurate. Accuracy isn't really the point, instead it's about utility. It could have no scoring at all and tags and lists would still make it useful for the intended purpose.
Update
Many thanks to everyone who's taken a look so far. I've already learned some things from the comments and DMs.
The first thing I learned is that people expect scores to be complete out of the gate. That was never the goal for the beta but I can understand the expectation. Scoring will keep improving, I've added full scoring impacts for a half dozen companies since my post yesterday, but complete scoring is going to need community involvement. It's one of the hard problems the project exists to solve.
The reality is that all of the existing attempts to score companies based on behavior, including general metrics like ESG and SDG, are either inaccurate, disagree with one another, or only cover a limited subsection of companies. Despite that I do pull in scoring from a bunch of 3rd party sources and then run it through an algorithm to weight scores based on the source quality and type and then average them into something useful. However, "useful" is relative, in a lot of cases it's more valuable for comparison between companies than it is for accurately capturing the practices of a company.
From my research, the only metric systems that actually claim large scale company coverage are AI slop. They look ok at first glance but fall apart when you look close. If anyone has any leads on good scoring data sources (general or in specific areas) I'm excited to hear about them.
So that's where custom scoring impacts come in. A perfect, objective measurement system isn't likely to exist anywhere, any time soon, but with community participation I think it's possible to build a system that works well enough.
As I mentioned in my OP, I think lists (or should they be called campaigns?) are an even more interesting way to solve the problem of determining company values than scoring. Creating and sharing lists is also a lot easier for community members than researching score impacts.
Something else I've learned is that I should have completed scoring impacts for the Villains lists before beta! I'd only finished the Champions lists. That's a work in progress which should be done in a few days.
Thanks to @CannibalisticApple for pointing out that there was no way to see all lists a company shows up on from the company profile page. I just deployed an update that fixes that. It will be sparse until users start creating their own lists.
Thanks again to everyone who's contributed their time so far, it means a lot. Keep it coming!
Made an account and am browsing. Seems fairly detailed so far, and the interface seems nice and clean.
My first point of input from a cursory glance: maybe it's because I'm on mobile, but I'm not seeing details on which "avoid lists" companies are on. Just the fact they're on a list. Will the icons only apply for lists made by you (Custodian), or for all lists made by anyone? Is there a way to see what lists they're on?
On that note, interesting to see half the "Ethics Villain" list is green. When I checked some of them, their ethics also seemed to be well-rated, and it didn't give a breakdown of why they're on that list. The green companies on the list include United Healthcare, Altria Group (Marlboro) and Johnson & Johnson, all of which showed positive metrics on the site. The only score impact listed for all three had to do with the environment score. I agree with their inclusion on the list, but could you explain the criteria used for the list?
Thanks for the detailed notes!
In order:
I love that the interface feels clean.
That's a good point, and it's already on my list to add something to the UI to show which lists a company shows up on. I'll move it to the top.
The icons/badges show a count of all lists a company shows up on. The above UI change would add a List detail section to the company pages that shows specific lists they're on globally. I may eventually add a setting to control that behavior (badges show for all lists or just owned/followed lists)
You're right about the Villain lists, some of the companies don't yet have full scoring impacts. I've been working on that this week. The aggregate data sources (where the base scores come from) are... imperfect. The champion list companies are fully scored, and most of the environment villains list. Once the impacts are finished it should be more clear why they're included.
Besides the obvious USA focus based on my 10-15 min visit, I don't think I would use this.
It seems to me (again: 10min!) that there is no way for me to decide what ethical values I value or what points I would like to appoint for different things, so I just don't think I'm the intended audience.
If I'm going to try to do an ethical choice for my consumption I would want to use my ethics rather than yours. (But you know: "there is no ethical consumption under capitalism", and all that!)
For me to even consider using this I would like to have more data on the economic and company-size side of things.
I understand these are early days, and it is fantastically fine to not take me into consideration :)
No worries if you're not the intended audience. But if you don't mind I have questions!
How would you want to be able to do that? Is creating your own lists unappealing? I might not understand what you mean here.
The goal is for the company descriptions, scores, impacts and tags to be as neutral as possible. Did it feel like that's not the case? If so I'd love details about what made it feel that way.
Whenever the data's available, company profile pages show revenue, profit and asset numbers. Employee numbers too. What kind of economic and size metrics would you like to see?
Thanks for taking the time to look at it and post!
I'm assuming there are X metrics that you intend to track. Each one can get a score of 0-100 and then you use that average to present a total score.
I would like to get a default list based on my own weights.
Let's say you track:
A) use fossil oil products
B) use of animal products
C) use of mined products
D) profit usage
E) countries that the company operates in
F) use of LLMs
G) reparability
For some consumers all of these are equally important, but for others they are not.
Being able to assign my own weights to these parameters/variables would make the project vastly more useful for me.
Ethics are by definition not neutral. You will have to pick one of those.
I don't know really... It just feels weird to me that a company like Unilever can score this high... Maybe country/ies of production, country/ies where they sell to consumers, country/ies from where they source their raw materials, policies around distribution of profits, wage structure, policies around sub contractors.
But I really don't know, I just hate capitalism :)
You're wolcome, I do enjoy picking apart others work, so thanks for posting :)
Ah I see, that makes sense. And it's an interesting idea to let users determine how to weight different factors on a super granular level.
I learned early on that scoring is a hard problem and that perfect is the enemy of good. That's why I added multiple ways to choose/rank/compare companies.
So for example, users can choose to make lists based on any metric, or combination of metrics, that they want. You could only look at environment scores for example, or the forced labor tag, or really anything. Or they can follow lists made by someone they know shares their values.
That doesn't give you the ability to dive into the minute details like you're talking about but it does mean that people aren't obligated to use the scores in any particular way. They can even ignore scores entirely and use their own metrics, then use the platform to keep track of their results, and make buying decisions, via lists, barcode scanning, etc.. The platform will still trace a product to an ultimate parent for you, no matter how you choose to rank companies.
It would be amazing to have the kind of universal scoring detail that you'd need to accomplish what you're talking about, but also incredibly expensive and time consuming to create. For some companies the data to do most of it is out there, but for the vast majority it's not published either by the company or 3rd parties.
So step one would be investigative deep dives into all of the companies. In most cases even then you'd only end up with partial metrics because big private companies don't tend to share, or even track, a lot of the data you'd want. Public companies are easier in some ways because they're required to share certain data, but unfortunately they aren't required to share or track much of what you'd need.
What you'd end up with would be difficult to use because you wouldn't consistently have the same data points for all companies, making it hard to compare.
I do think, though, that with a big enough community, more is possible, including better scoring. The trick is getting to that point!
Note that any user can submit sourced scoring impacts (modifications) to any metric in a company score. It's the same system I use for scoring companies on the backend when 3rd party data sources produce scores that aren't right, which is most of the time.
Which brings us to Unilever! I agree, it makes no sense for them to be scored that high. I haven't done a researched scoring pass on Unilever yet, which means they get their scores from aggregated 3rd party data where available. As it happens, Unilever looks really good on paper compared to their conglomerate peers from the perspective of most existing scoring systems.
Unilever has been making me cringe as I've seen people looking at them over the last few days. It's on my list to do a full scoring pass. Sounds like I should move it to the top of the list! Once the scoring pass is done, there will be impacts showing detail about the factors used to modify the scores, along with source links. You can see that in action on a bunch of companies already.
The key point, though, is that you can put unilever on an avoid or boycott list and never buy from any of their brands again no matter what their score says. Or you can fix their score via revisions.
As I'm learning, having scores at all seems to nudge people in the direction of expecting them to be perfect and complete. When really that's not a realisitic early goal. It's something that no organization of any size has yet managed to accomplish.
I'm wondering if it might make sense to rethink the UI to de-emphasize scoring for the time being.
A system that is less numerical will absolutely keep people from assuming precision that doesn't exist. Colors or the Angry / Sad / Neutral / Happy / Delighted faces/emoji range will give people the vibe they need without thinking that a specific number should mean something you think it means.
This is a great idea, I'll give it some thought
I don't think I managed to explain what I meant, I will try again...
This is not what I meant (although I'm all in favour of granularity down the line).
What I was trying to say was: I assume you use some metrics to evaluate the ethics of the companies [0].
You then combine these metrics into an overall score (regardless of if you decide to show it as a number, smiley or not at all!)
This total is currently based on how you, personally, value the different metrics in relation to each other.
I, and every other user, will value these metrics differently from you and each other.
I "just" want to be able to put in my weights on the current metrics, that is all.
For me, it would be interesting to be able to read up on the metrics and your method of evaluating the companies, that would also be a nice way to drill into the data.
I've been thinking about this project a bit and I think I don't really understand how you see it being used as a consumer or contributor.
[0] I scream a little inside when typing this out. So adding a little reminder here for myself and whoever needs it: companies can not have ethics because companies are not persons. Only living creatures can have ethics. People can decide to be ethical at work, but that doesn't make the company ethical.
Sorry for misunderstanding.
It will probably help to differentiate between different types of metrics here. There are the 3 metrics the app uses: Environment, Human Rights and Ethics [0]. Those are averaged to get the overall score with no weighting. If these are the metrics you mean, user customization is indeed an interesting idea.
And then there are 3rd party metrics that are basically used as a placeholder for companies that don't yet have scoring impacts. Those do get weighted during the process of translating them into the Enviro, Rights and Ethics metrics. For example, policy metrics get less weight than practice metrics and well sourced metrics get more weight than less rigorous metrics. That system is a bit complex, and still evolving, so at this point it would be challenging, and likely confusing, to try to expose levers for people to customize weights.
Agreed, I plan to add more details about scoring when creating score impact submission guidelines. Long term, score impacts are the method for evaluating companies and they're entirely transparent, each impact is detailed and available on company pages. Possibly you just didn't happen to look at a company that had full score impacts.
What this says to me is that I need to do a better job of communicating what it's for and so does the app user experience.
There are a lot of potential use cases but I'll just pick one quick example.
Prerequisite: Someone needs to care about what their spending is supporting. Without that the platform is entirely useless to them.
Say someone is trying to decide which vacuum cleaner to buy. They're looking at 3 options based on the features they want. They could use Ethoscore to compare the companies behind those 3 brands either by searching for the companies or by scanning the barcodes on the packaging.
How they do the comparison is up to them, they could use overall score, or focus on the Environment metric, or look at specific tags that are important to them. Or they could dive in and read all of the scoring impacts for all three companies.
Alternatively, and to me this is probably the most useful for the majority of people... if they've already created or followed some lists, the work might already be done. Maybe they already follow an Ethical Home Appliances 'support' list and one of the brands is on it. Or they follow an Evil Conglomerate 'boycott' list and two of the brands are on that. In which case the decision is straightforward.
I'm with you there. I struggled with that when naming the Ethics metric. I needed a category for everything that was left after Environment and Human Rights that wasn't clunky or unintuitive (like Governance). I look at it as a way of capturing company behaviors that people would generally consider unethical (lying, cheating, stealing).
Apologies in advance if I’m way too corporate-brained at the moment, but maybe “integrity” or “honesty” or something? Just a suggestion.
Thanks for the suggestion. I kinda like Integrity. It has the same issue as ethics though "corporations can't have integrity!"
Just because I want to pick a boring fight...a lack of neutrality is a defining feature of morals and generally not a defining feature of ethics ;)
Honestly, I never figured those out... I just wing their usage.
Would you mind explaining them again and include why I was wrong to say ethics can't be neutral while it's morals that can't?
In order for this to be useful to me it needs to be more visible where each company is based, ideally listed on the card or at the very top of the company details view.
Also, you should add USA as country of origin, not just the city and state for American companies.
For some more general feedback, I think this is such a great idea. However, the way you introduced the project gave me pause. The only way I could see a project like this succeeding is following the Wikipedia model and remaining nonprofit and asking for donations. The minute investors become involved the validity of the ethics scores could come into question, not to mention leveraging community labour to flesh out the scores before capitalizing on it for yourself (and whoever else stands to gain financially) would be quite unethical itself in my opinion.
Just some food for thought.
I like the idea of giving location first class citizenship. I'd say it's reasonably accessible on company profiles as is, but it sounds like it would be useful to a lot of people to be able to filter or search based on location. Is that sort of what you're getting at? Or is it just that you hate having to scroll for it on company profiles?
We're on the same page, I'm curious what gave you the impression investor money was the goal? Was it the use of the word funding?
I actually had a bit about this in the OP, but removed it because it was already too long. My goal is for the platform to remain independent, that's part of the idea of doing it as uncynically as possible. That means a benefit corp at first and eventually either 501c3 status or foundation ownership.
It really doesn't make sense to me any other way, as you say, taking for-profit funding of any kind makes the entire thing suspect. It would also be hypocritical, the highest scored companies on the platform are going to consistently be those that bake the mission into their structure so that markets can't corrupt them.
1: I had read your blurb about how you've been out of work but you have experience monetizing applications in the past.
2: You said you had no plans for funding at the moment but that part would come later.
3: I read this at at like 530am just after I woke up.
I'm glad to hear my fears were unwarranted, at least as far as what you plan at the moment.
As far as wanting company location to be more accessible, it's just because when browsing the pages on mobile I had to scroll way down to find it and that was annoying.
Thanks for sharing this, it's clear that you've put a lot of throught and effort in to the project already!
After exploring the site for an hour or so, I've got a lot of thoughts. Please excuse the wall of text, I'll try to divide by function/area to keep things neat.
Score Impacts: I really appreciate the Score Impacts pages and found these to be the most useful for my purposes. While the overall score is important at a glance and can give a decent overview, being able to review the qualatitive information that justifies the score and verify via the sources is what I value most. That said, I feel like this information is a little hidden away and could be better integrated in each company's main page. Instead of clicking on a 'view score impacts' hyperlink, could these be expanded/hidden by the user? Alternatively, clicking on any of the Metric Scores currently takes the user to the 'how scoring works' page; could clicking on the Metric Scores instead expand a list of relevant score impacts with a seperate link for how scoring works (either when expanding, or always visible to the right of the metrics?)
Metrics/Politics Regarding the metrics themselves, I love the categories you've chosen but think Politics needs a bit of work. As you've outlined, the Politics metric is unique in that it is the only one that doesn't alter the overall score. That's great and I agree wholeheartedly with that decision, however I feel like the UI doesn't really reflect this. At a glance, the Politics metric has a number score with a colour gradient background. The only noticible difference is the gradient is red-blue rather than red-green (and that isn't immediately obvious as a distinguishing factor if the business is conservative when comparing to other metrics).
Another user has pointed out that the site is quite US-centric, and this is reflected in the colours: I'm assuming you've gone with the associations of the Democrats and Republicans with Blue and Red respectively, whereas in Australia our leading conservative party's primary colour is Blue and our more Progressive party is Red (Also our conservative party is literally called the Australian Liberal Party... Insert upside-down joke here).
My reccomendation would be, if at all possible, to reconfigure this metric to present differently to the others to distinguish it further and clarify it's purpose. Perhaps chevrons could be used instead of a score/colour? (e.g. conservative > and progressive < with extra chevrons depending on the score i.e >> more conservative learning? ) Or could this be a visual scale with a value sitting further to the left or right as appropriate? If you did stick with the value-in-a-circle approach, I think at least offsetting Politics or having some sort of visual seperator would help distinguish it from the other metrics, i.e:
(E) (E) (H) | (P)
for Ethics, Environment, Human Rights and Politics.
User-specific issues and causes: @fnulare has given you some great feedback in their comment here which I think highlights the need for users to quickly compare to their own preferences. I don't entirely agree with their suggestion of having a user-weighted score for each company (I personally like the idea of having a universal score that I can then adjust myself in my head after reading the score impacts), but I agree that the site in its current form makes it a little difficult for users to identify issues they care about on a more granular level than the broad metrics you've provided.
This is a wild tangent, but if you'll endulge me for a moment... Have you ever played any of The Sims games? In each entry, players take control of characters and need to monitor their wants and needs. In The Sims and The Sims 2, this is communicated to the user by several bars/scales for each need (i.e Hunger, Social, Hygeine, Fun, etc, with 0/empty = low and 100/full = high). In The Sims 3, 'Moodlets' were introuced to the UI. These were essentially small UI elements with an icon reflecting different impacts (or potential impacts) to a Sim's needs. For example, a 'Well Rested' moodlet appears when a Sim has had a good rest in a quality bed in addition to the Sleep bar being full. This helps the player track what variables are impacting on the need values at a glance instead of just looking at the bars and working backwards from there.
You're already using tags for another purpose, so for a lack of a better term... Could you introduce 'moodlets' for different causes and issues? e.g. I care about plastic packaging waste and learned from Ethoscore that Coca Cola ditched many of their recyclable packaging goals in 2024. If Score Impacts were tied to 'moodlets', that could display a negative 'Plastic Waste' moodlet next to the specific score impact, and perhaps all 'moodlets' could be listed on the company's main page. These could be binary (i.e positive/negative) or have a value-score tied to them (though I'm assuming the former would be easier to implement) and would allow users to quickly check how companies are acting specific to issues they care about. Another example: I am progressive-leaning myself, but if I found that a company had been bribing an elected official I would be less inclined to buy their goods regardless of their political stance. A 'political interference' moodlet on the company's main page could communicate this quickly and, depending on how it is implemented, quickly direct me to the relevant Score Impact so I could read specifically about that issue.
... Of course that is a suggestion from a user with no appreciation or understanding of how difficult that would be to implement or if it would even work. But I thought it might be worth sharing to demonstrate some of the challenges and opportunities available to you.
RSS? - Lastly, would it be possible to push your blog updates to an RSS feed? I would definitely subscribe to keep up with updates to the site/project. As a stretch... I would also love it if Score Impacts had RSS feeds so I could subscribe for companies I use and track how they're acting over time (especially in our ever-enshittified world, I'm conscious good companies don't always remain good and would like to ditch as appropriate).
Keep up the great work!
Edit: Fixed the directions of the chevrons in my example.
First off, this is outstanding feedback! 🤌I'm grateful you took the time to browse the site and write it up.
I'm glad someone found the impacts! Which makes your point, they're not prominent enough. I've noticed this from analytics over the last couple days, a lot of visitors never open them. And yes, linking from metrics to impacts is definitely the way to go. When I added the scoring and metrics, impacts didn't exist yet and I never got back to it.
Entirely valid. As it happens I've been working on updates to the scoring UI and I've been trying to figure out what to do about politics.
Starting off somewhat US-centric was intentional for simplicity, it gets more international all the time though. But your comment makes me realize the Democrat/Republican colors is terrible UI. Which introduces the question of what replaces it?
Your chevrons suggestion is a possibility... but it's maybe a little busy? I'll need to think about it. I like the idea of a visual separator.
[...]
I'm not sure tags actually are being used for another purpose. Plastic Waste (Noun) makes perfect sense as a tag, and tags already have built in polarity. They're meant to serve as something like a 'moodlet', but primarily for filtering at the moment. And then Lists cover the rest of the concept. If someone creates or follows a "Worst plastic polluters" Avoid List, they'll see a badge every time they encounter a company on the list after that. All it's missing is an easier way to see which specific lists are involved in the badges when there's more than one in the same category (support, avoid, boycott).
Assuming I'm understanding what you mean by moodlets, I think the functionality is mostly already there.
But if so, here's my real question: Why do you think it didn't seem that way when you were browsing? Are Lists and Tags not prominent enough? Is their utility unclear? Are they insufficiently explained in the docs? There's definitely a problem somewhere if it's not obvious that "hey I can use Lists to personalize things based on my values".
Absolutely, currently RSS is filed under "nice to have" but I'm happy to prioritize it. No promises on when, and no loss yet, there isn't much to read at the moment.
An RSS feed for impact/score/etc. updates is a really interesting idea. I have email subscriptions on the list but I hadn't though about using RSS for that. My only concern there is scrapers, I've taken great pains to make it hard for scrapers, partly because of AI training scraping exploding lately. But there's no reason it couldn't send update abstracts instead of complete records.
I think the in app notification system is also a good place for "this company on your support list just annexed a small country and is planning to sacrifice their children to Mammon" type updates.
This is great stuff, thanks!
Thanks so much for the kind words and detailed response!
I hope the US-centric feedback wasn't too harsh. It's important to work with what you know, and it's impossible for any website to be completely geoculturaly -agnostic. Completely understandable.
As mentioned in my edit, I managed to confuse the directions of the chevrons in my own example so they're probably not the best idea. I do think that the UI element should visualise the left-right political spectrum somehow, though. I know it's not perfect, but it's a shorthand that I think users would appreciate.
... I genuinely have no idea how I managed to so completely misinterpret tags. I can assume the why would have something to do with scrolling through your site at 4am due to a restless toddler waking frequently. It's well documented, and the functionality is clear when interacting with tags on each company's main page. You've essentially already implemented exactly what I was suggesting. From memory, I may have first noticed tags when exploring the lists functionality and assumed only lists were taggable? Unfortunately, this was just a good example of the Dunning-Kruger effect. I was confidently incorrect.
In spite of myself, I'm going to extrapolate from there and explore how you could make the concept sleep deprived parent proof. I think the main reason why I wouldn't have noticed tags straight away is that they're separated from the Metric Scores by the Brands and Subsidiaries section. Ideally, it would be great if the UI elements for tags themselves had icons rather than acronyms, but I'm guessing either A) the acronyms are placeholders or B) icons would be a lot of work. Otherwise, I love that I can hover over them to see a brief description. Good stuff! Other than that, maybe a few tags for businesses could be present in the search list? (perhaps limited to 3 with a '+2 more' descriptor if there were five, for example), but I get that would interfere with the icons you currently have to demonstrate when a business has been included in a list.
Honestly, please feel free to chalk this up to my ineptitude and not reflective of any poor UI decisions!
Oh that's absolutely fair enough. I hadn't considered that, and I now understand why so many of the feeds I used to subscribe to have gone dark after 2023. I know I would use them, but I'm probably an edge-case and well outnumbered by bad-actors. An in-app notification would serve this purpose well for the vast majority of users I think.
Tangentially related, but depending on how you want to use your blog/mail list it might be worth checking out Open Street Map's Weekly OSM - Rather than drip-feeding every single update via RSS they aggregate major updates, news pieces and stories in to a weekly summary. I'm guessing something similar would be administratively burdensome at the moment but even just a highlights reel of notable new Score Impacts on a regular basis would be neat.
Cheers!
Agreed, some sort of left/right indicator makes the most sense for politics. That's intuitive for at least most countries.
Not at all, one of the main points of the beta is to polish the rough edges. I'm happy to hear about things that don't make sense or feel off.
Thanks for the details about Tags and Lists. Restless toddlers are a good excuse for almost anything! It's not just you though, there have been a few comments in the personalization vein. Somehow much of the functionality beyond just scores isn't obvious for everyone. Just need to figure out how to make it more clear in the UI. Maybe an opt in tutorial.
About the tag acronyms... I expect there to eventually be enough tags that finding visually intuitive icons for all of them would be challenging. Also, eventually tags will be community created, where icons would be even more difficult to manage simply. I agree though, the letters are a lot harder to parse at a glance. I think giving users the ability to highlight certain tags they care about, which would change how they display, will help with visual differentiation.
Adding tags to search makes sense, that'll go in when I add an advanced search UI.
About RSS... yeah it's mostly dead, but that doesn't mean we can't keep supporting it. It's early internet tech, long live Web 1.0! At the very least it makes sense for new post updates.
RSS or not, an updates feed for tracking company developments is a great idea.
Invites
Reply to this comment for an invite.
Here are a few to start things off:
[snip]
Fresh invite link:
[snip]
Turns out bots love the invite links, should have anticipated that. I want to minimize the bot noise for now.
Please reply or DM for an invite instead.
Hello, I'm interested in an invite!
Sent, thanks for taking a look!
I'm pressed for time for a few weeks so I may take a while to check this out and comment. But I love the idea and would like an invite please!
Sent, and best of luck with whatever's taking all your time!
I would be interested in taking a look.
Sent
Hey I'm curious about this.
Sent
Would love to give it a look. Thanks!
Sent
Would be happy to take a look! Hope it goes well.
Thanks! Sent
I'd like a look
Sent
I'd like to check this out
Sent
I'd like to take a look!
sent, thanks!
I'm all for cool projects and such but I'm wondering, what makes this different than something like Buycott? I haven't used any of the invites or anything yet, but I'm curious.
Also as general constructive criticism, I'd maybe move the section on your backstory till after you describe what the project actually is :)
I hadn't heard of Buycott, but at a glance it looks like Buycott is a barcode database that links products to user created campaigns. It doesn't appear to have any info on the companies and brands themselves.
When I did a search for "Apple" from their main page it took about 15 seconds before returning "no results". When I chose the "brand" search and tried again it took another 10+ seconds before returning a page of results that didn't include Apple. Coca-Cola was the top result. After a few other searches it looks like the brand search always returns the same list of results for any search term.
ETA: I got a minute to look into Buycott a bit more. Appears it hasn't been updated in over 10 years, it's essentially abandoned.
Thanks for sharing this! I certainly don't envy this task of attempting to rate companies as it gets very complex very quickly.
My first point of feedback is the colorscale. The white text on the light green background is more difficult to read than it should be. I'd recommend using a different shade of green or something to outline the shape of the text, like a dark grey or black.
Some type of breakdown to see why a score is where it is I think is very important. You have this kind of with "impacts" but this seems arbitrary. For example, why does Tesla have a score of 21 for human rights (calling out their union busting) but Geely, which is not transparent over its material sourcing and very likely benefits from Uighur slave labor, has a much higher score of 39? These ratings should be justifiable and transparent, like the EV environmental and human rights tracking website Lead the Charge does with their 77 page published methodology and yearly dataset publishing.
Thanks for the feedback!
I'm not sure it has to get too complex, but I'm definitely hearing that's where a lot of people's minds go. As I mentioned in a different comment: perfect is the enemy of good.
Regarding the text contrast, that's useful thanks. Looks like on the green end of the spectrum at the worst case the contrast ratio is about 2.5:1, which is definitely too low. As it happens I'm working on the score UI right now so I'll fix that soon.
What you're seeing with the scores is the result of incompleteness, neither Tesla or Geely have had a full scoring impact run yet. Thanks for being specific, that makes it easier to talk about. I'm probably about to give you way more information than you wanted!
The impacts aren't arbitrary, if you click through to the impacts page there are details about each impact, along with source links. However in the case of those two companies the only impacts are tag related, which cover a lot of companies with a given tag at once.
On a company that's had a full scoring pass there are many more impacts, each with reasoning and details. Which, I think, makes it pretty clear why they're scored the way they are.
Prior to a full scoring run companies get their scores from aggregated 3rd party sources (assuming the company is covered by them). To give you an example, the most popular and widespread scoring system is Environmental, Social, and Governance (ESG). It's used by investors, regulators and internally at some companies. There are a variety of problems with it including: No two ESG data sources agree on scores, to the degree that it calls the whole system into question. Scoring is mostly focused on publicly traded companies, which gives private equity a pass. Scoring is weighted towards policy and compliance, as opposed to practice and outcomes. The end result is that ESG scores kinda suck. But at least they exist! This is a problem society is still trying to solve.
The same is true of other large scale scoring systems. It's great that they exist but they don't cover anywhere near all of the companies you'd want them to and they don't capture the full picture. Part of the reason for that is, despite what many people might imagine, there just aren't reliable or comprehensive sources for the data. Companies aren't required to track or report as much data as you'd hope and there's no 3rd party organization with anywhere near the resources to gather and publish the data on anything other than a limited subset of companies. Many kudos to the orgs that try.
On the other side you have highly targeted scoring like the Lead the Charge org you linked. It's granular because they only have to cover 18 companies. And even then I suspect a lot of people would disagree that Tesla should be ranked number 1. Mostly because: Musk. In the case of Ethoscore his impacts would likely show up in Ethics. So there you have a project where they don't have to deal with a huge list of companies and even then scoring is incomplete. Not to say they're doing anything wrong, it looks like they kept it narrow intentionally. It's a targeted awareness campaign and for that purpose it looks great.
Scoring is hard, it's an unsolved problem that I think is worth trying to solve and I think community contribution can be part of the solution. Someone could submit new scoring impacts for Geely or Tesla that would modify their scores and make them less wrong. And less wrong is a lot better than nothing. Eventually scores could even become mostly right! Though of course people will never completely agree on what right looks like.
My view, at least right now, is that keeping scoring relatively simple is a better solution than trying to give it too much granularity and intimidating people. Contributing should feel straightforward enough that anyone can do it. If it involved, say, 500 data points, no one would even consider it.
The goal is to give people a tool to make spending intentionally easier. Scoring has to be a part of that, but it's just one piece.
More information is good! :) I definitely agree scoring is hard. How much do you plan to crowd source, and what steps are you considering implementing to prevent groupthink?
I do agree less wrong is better than nothing, "don't let the perfect be the enemy of the good" style ideal, but I don't think scoring being simple and unexplained is better than having too many points and being intimidating. If I see scores that I don't expect or agree with, with no firm explanation or justification on why they are that way, I'm going to lose trust in those scores and wouldn't bother using the website, because it would feel like a biased take from the creator. Transparency in how scoring works, what contributes to it specifically, and why that score is the value it is, is extremely important to breaking through perception barriers for the types of people who would seek out and use a website like this to begin with. Does that make sense?
My plan is to increase the amount of community involvement as far as possible. For now community contributions are admin reviewed but at some point I hope the community will be able to handle reviewing changes and submissions. It's a problem for when the community actually exists though. For now it's enough for the tools to exist, and I plan to add more community contribution tools before the end of the beta.
Perfect sense. We might have different ideas about who those people are though. In my mind the main demographic is people who want to be intentional about their spending but also have busy lives so the less bandwidth required the better. But I think there's plenty of opportunity to dig deeper for people that want to via impacts and tags.
Quoting from my previous post:
I hear you about transparency but again I think you're extrapolating its lack from incomplete scoring. The goal is definitely for it to be easy for users to understand why a particular score is the way it is. At the same time the scores really aren't as important as they currently seem, which is a problem I've been working on for the last few days. Posts like yours have helped me realize that the way scoring is presented distracts from arguably more useful ways to determine whether or not to buy from a company like Lists. I think I mentioned this in a different comment thread but: With scoring completely removed the platform would still be useful. Maybe more useful, I've considered just removing scoring.
By the way, the Tesla scoring pass is done. It's actually worse than I expected. It's rare for a company to have so much recent evidence of bad practices on the public record. Along with so few good practices. Which likely means their bad practices that are unproven and/or undiscovered are epic in scope.
Thanks for the reply, your perspective is really helpful. At this point nothing about how the app works is written in stone.
Another short comment regarding the impacts. Ideally I'd love to see a tag cloud next to each company listing that shows at a glance what mainly impacted the overall score. Things like 'planned obsolescence', 'child labour', 'greenwashing', 'vendor lock-in', 'extractive pricing', 'inhumane work conditions' etc. would lower the initial friction of using the site and pull people in to read more about an issue that they especially care about.
I know it's probably a nightmare to implement!
This exists, though not in cloud form. There's a 'tags' stub that lists all tags attached to the company on company profile pages.
It's not comprehensive at this point, managing a large list of tags and assigning them to all the companies is going to take community contribution. In the meantime I'm working on it slowly (14 tags and counting, including Child Labor and Greenwashing from your list). The best way to find companies with tags at the moment is using the tags filter dropdown on the /companies page.
Thanks, I responded in the other comment and then edited it. I'll get back to this later! Sorry for the rushed comments :)
Do you enable custom weights, or do you make the ethical judgments for your users? I'm sure there's a lot of variance between what different people want.
For instance, in the current world situation, I would like to find all companies that work with Russia in any way and ban all of them from my personal consideration. As for China, I'm not so strict, but would prefer not to get anything produced there. Other countries are mostly fine by me.
Another would be fossilized fuels, anyone working on those would be a big no no.
3rd thing: political opinions of company leaders and owners. That'd be an interesting angle possibly.
If I'm understanding you, I think what you're asking in answered in this comment thread. I'd add that people can make their own decisions on how to interpret the data, the goal is absolutely not to make the decisions for them.
About Russia... this is a fantastic idea for a tag, which would then make it easy to avoid companies that haven't exited Russia. That's going on the list!
China is more challenging because you'd have to go multiple levels deep into supply chains to determine if a company has Chinese suppliers. That's complicated by the fact that almost no one publishes that information. You could pull together some of it from investigative journalism sources, certain NGOs and regulatory actions, but right now it would be essentially impossible for it to be comprehensive. Still, anyone could make the attempt via the Lists system.
ETA: The Forced Labor tag accomplishes the China goal to some degree because in so many cases the source of the forced labor is Uyghurs. Coverage is limited at the moment though, the supply chain tracing issue again.
Fossil fuels... 100%. There's actually a start to that in the form of a big research project I did on the industry shortly before the beta launch. It's captured in the score impacts of a bunch of companies and also via the Climate Policy Obstruction and Climate Deception tags. The reason it's captured in those tags is that, almost without exception, every fossil fuel company, along with most automakers and some investment firms, have been involved in fighting against climate regulation. I have a post planned covering the research for when time allows, but the scoring impacts and tags are already applied.
Another good solution for the fossil fuel industry would be a list that includes all of them, and all related companies.
I've been hoping and waiting for something like this to emerge by a trusted and unaffiliated person/group and I'm so happy you decided to go for it!
I tried out the request form by submitting a company that I know well and love. Not sure if I used the right option in the menu ("company request" / "new company"). I chose the first option because I don't have time to do the longer one at the moment - I hope that was okay.
One feature I'd love to see is whether a company is listed or not. I try to favour unlisted companies because I believe they aren't as profit-hungry and short sighted as those that have to please shareholders (generally speaking).
I would also like to easily see what type of company each one is, and the different company types scored on their suitability for ethical practices. And I mean every company structure form every country all over the world! I know that some new company types have emerged recently that try to make it clear the company doesn't only exist to maximise financial profit to its owners. I'd love to learn what those are in each country and which companies have chosen to function under those particular structures. This should also affect their scoring.
Hello target audience member! Good to see you :)
Your request is queued, thanks for using the feature.
The 'listed/unlisted' feature does exist in the form of company type (on company profile pages). Currently it includes Public (as in publicly traded), Private (not the former), Government and Coop (cooperative). Do you feel like that should be more visible?
Not so fun fact: There's a new trend. These days some of the worst offenders aren't publicly traded shareholder pleasing companies, they're private equity conglomerates.
More on company types: There is currently a B Corp tag, which is one of the new structures you referenced. B Corp certification has requirements like baking environmental, social and ethical considerations into their corporate structure (whether or not they're publicly traded), transparency and so on. It's one of the most popular ways for companies to make impact a part of their mission and prove they're serious about it. More on the B Corp organization if you're curious.
This is a great idea for a blog post, I'll add it to the list. It's also a good idea for a curated List, which anyone can create.
Damn it, I guess I have to educate myself about that.
The company types you list are American. As soon as you can, I'd love to see this site become internationally focused and include country-specific information. I know it will be a huge undertaking with how many different countries and company structures there are in the EU alone, but it's really necessary for credibility and adoption. If you manage to pull it off, you will entice more companies to reconsider how they are structured! At the moment, company type gets very little public attention (at least in my field) while all sorts of certificates and other paid credentials things get all the attention (and in my field, some of the certificates are more about greenwashing than anything real).
My apologies, I didn't even see the tags.. That's kind of the point of my other comment: I'd love to see them without having to use a drop-down - directly on the table that lists the companies. One column for company type, another for the other tags, or perhaps a column each for each separate metric (showing the tags that influenced that metric).
I am stoked! <3
Yes, making it more international is definitely a goal. All feedback on that is welcome.
I'll look into broader terms for company types. I know from company research that the number of designations for company types globally is big enough that trying to include them all would likely be overwhelming and confusing for users. But more generic designations covering all of the important structures might work.
The /companies table can definitely include more columns, and/or filtering options. I was waiting for user feedback to help determine what to include, so thanks for yours! The trick is balancing UI simplicity with having the options most people will want. That and fitting it all into mobile screen sizes.
Point taken about tag visibility, I've been giving a lot of thought to how to surface the various features to casual users without adding clutter so that's useful feedback.