Thomas and I found our way to this project via our background in fintech. Prior to Bend, I was a co-founder of Abacus (YC W14), a spend management company. At Abacus, we noticed that finance teams are increasingly paying attention to the climate impact of their purchase decisions, from travel policies, to cloud hosting, and beyond.
We believe this 'spend based' approach is the key to unlocking scalable carbon accounting. Today, most carbon accounting is manual, conducted once a year, and takes weeks or months to complete. It's like doing your taxes.
Fortunately, in the last couple years, we’ve started to reach a critical mass of good merchant data. It used to be that only a handful of companies tracked and disclosed their emissions. Today, 70%+ of Fortune 500 companies disclose their emissions in annual sustainability reports, and 1/3 of the entire global economy is now covered by a Science Based Target (and this coverage and quality is accelerating). The spend-based approach is fully automated and starts working the moment you connect your bank account.
That’s the good news. The bad news is that this emissions data is trapped in PDFs and blog posts, scattered across the internet. We aggregate and normalize this data by hand today, and plan to automate the process in the future.
Here’s how we measure the tCO2e (metric tons of carbon dioxide equivalents) of your transactions: imagine that you get a $1,000 bill from Atlassian and want to know the carbon impact of that purchase. We know that for every $1,000 spent with Atlassian, there’s a 30.2 kg footprint — we multiply your bill total * (Atlassian’s annual emissions / Atlassian’s annual revenue) and return the tCO2e of that transaction.
Now imagine a similar calculation for each of the thousands of purchases your company makes every month / quarter / year. For merchants that don’t yet publish their emissions data, we fall back to category averages (e.g. for a Starbucks, we use the specific Starbucks carbon intensity factor, but for a mom-and-pop coffee shop that doesn’t disclose their greenhouse gas info, we use a generic benchmark ‘coffee shop’ factor).
To get a feel for the data we track, click on some of the corporate logos on https://bend.green/ — these aren’t customers; they’re examples of Bend’s merchant data. We have a climate scientist PhD on the team named Marion — we’d be happy to answer questions about our methodology!
Measuring and reducing your company’s emissions is of course good for the planet, but it also prepares your company for upcoming regulations and investor requirements. We help you create a 'climate profile' that you can use to close sales as the sustainable alternative to your competitors (you can share your info with prospects, customers, employees, investors, etc. — think of it like the climate equivalent of becoming SOC 2 compliant). And we just rolled out the ability to purchase carbon removal credits, powered by Patch, to offset some or all of your remaining emissions (optionally opt-in to automatic monthly purchasing).
Our pricing is $100 / month per company, and your company can try Bend for free for 14 days: https://app.bend.green/sign-up
Bonus points: if you’re building a fintech app, Bend data is also available via API (email us for API keys and docs). And if you work at a large / public company that already measures emissions, we encourage you to claim your company profile on Bend (for free!), and ask your vendors to track their emissions (after all, your vendors’ emissions become your emissions).
We’d love to hear your feedback and we’re excited to answer any questions!
If you pay google 100 bucks, and See how google spends that money: it saves some, pays off its employees, it spends on infra and then finally on actual energy. Saved money is never accessed if parked. What it gives to other companies and people will spawn the same split we created when we paid google ad infinitum. Money spent on infra can have an outsized environmental impact (plastic, concrete) but I’m hoping we can ignore it, and the Final expenditure is pure energy.
I feel like if you keep breaking down every Avenue parts of your 100 bucks goes into, it’ll eventually go to creating energy more or less. Even if you choose google because they use green energy, the employees who are paid by google will use amazon powered Netflix so it might not matter any more than what’s the fraction of green energy in the total economy.
Thus the only solution to reduce your carbon footprint is to pretty much just spend less money! But no one wants that do they? They just want to have their cake and also eat it so we are just doomed.
Been outlining this for an article for a while, comments , prior research and criticism appreciated! https://www.ramrajv.com/blog/how-to-truly-reduce-your-carbon...
That calculation is not dissimilar from what we do on a merchant-level, and I think it's a good, macro-comparison! For your three variables, we just substitute 1) transaction amount, 2) vendor revenue, and 3) vendor emissions, if that makes sense.
There are many reasons why it's important to do this at the merchant level, but one of the most exciting (from an economics perspective) is to amplify market pressure on high-emitters. In the future, you can imagine a green marketplace powered by Bend, where you can instantly comparison shop for providers with lower "carbon intensities" than those you're currently using.
But I think the point you guys are making, is that actually everything is Scope 1 emissions plus some number of hops. So if you buy gas, that's 0 hops. If you buy energy from your utility, and they use natural gas, that's 1 hop. If you buy an ice cream that was made in a factory powered by natural gas, that's 2 hops. And if you buy that ice cream with a cone, purchased by the ice cream shop, that was manufactured in a factory ... etc.
The problem is, consumers are often the ones who apply the pressure to address climate (don't wait for the fossil fuel companies to take this on themselves). And so we need to trace back all those upstream emissions, all the way back to those Scope 1 emissions, to really size the impact, and align incentives to decarbonize.
Historically, climate programs only focused on Scope 1 and Scope 2 emissions. But that only addresses maybe 20% of emissions for most companies. This is why it's critical to consider the impact of all the goods and services your company purchases.
On average, it is true by definition (at least, if one holds that all activity in some way supports personal consumption) that carbon footprint from consumption is (global carbon output) × (personal consumption) / (global consumption).
It sounds like your proposed hypotheses essentially rephrases this.
> Thus the only solution to reduce your carbon footprint is to pretty much just spend less money!
In the short-term, sure. What spending a premium on things that reduce carbon footprint in normal narrow analysis (but which force the reductions in the short term to be someone else's increases who aren't paying the premium) does isn't actually reducing the near term footprint of the whole economy, it increases the economic incentive for production models which reduce the footprint per unit output.
What you spend your money on matters, even more than voting.
10% of companies advertising that they use 100% green energy, while the other 90% of society uses 0% is no better than the government forcing 10% of all energy to be zero carbon.
Economist agree that carbon taxes work, but they work at every level not just the first time it's spent which amplifies the effect.
However, I think I disagree on two important things:
a) fixing climate change is the cheaper option. We're "doomed" only to the extent that system incentives drive us to do the wrong, more expensive thing
b) more info can only help that case, even at the margin.
Voting is the most powerful way for most people to act. Voting with your wallet helps too.
For example methane and nitrous oxide, two extremely potent greenhouse gases, are mostly by-product of agriculture/cattle rearing: methane from cows and rice fields, nitrous oxide from the usage of fertilizers (from memory, I might be wrong).
Another example: the chemical process behind concrete production (CaCO3 -> CaO + CO2) emits more CO2 (about 4% of global emissions) than the world airplane fleet (about 2% of global emissions). That's without accounting for the energy necessary to trigger the chemical process, usually coming from fossil fuel...
I do. I've never been more frugal in my life (almost an hermit now) and yet I'm earning >€200K/y and have over $4M in $TSLA at ~35 y-o. I don't spend because I know any spending would end up adding pollution more than satisfy me.
I could give all the assets now though but I'm not sure which organization would best use it to speed up the transition to sustainable energy and prevent biodiversity collapse (to many players in clean energy don't care about the latter and even would work to worsen the situation).
If enough people did that, every supplier would eventually reduce CO2.
Energy can be produced with very little CO2 emissions, like solar, wind, hydro, and nuclear.
Without letting perfect be the enemy of good here... I have some questions.
It seems nice to have a backstop for what GHGP calls "Corporate-level data" which is the lowest level and least specific for purchased goods and services in GHGP's Scope 3 calculation guidance.
Do you have any concerns that your users might shortcut the work to produce high quality estimates of their Scope 3 emissions when you have made it so easy to get lower quality estimates?
As a second angle on this, do you have any accuracy or uncertainty estimates for the CO2e values a user receives? Let's say my company goes to a supermarket one month and buys $1000 worth of beef, and the next month buys $1000 worth of lentils from the same store. The GHG impacts of these purchases are (in reality) entirely different, but your API would tell me the emissions are the same. I know this line-item accounting is a massive challenge, but it seems there is an equal risk of green-washing as brown-washing here. This might be acceptable at the aggregate level but potentially harmful and very inaccurate at the individual level. Is there sufficient information returned from the API for your users to communicate data quality in a way aligned with the GHGP reporting standard?
Second line of questioning - GHGP guidance says companies must re-account historic emissions when data becomes available that significantly change estimated emissions. Would your API be compatible with this requirement? The Docs are locked off and the example on the homepage doesn't show any time component (presumably Uber's data would be different for 2022 than 2020).
Some disclosure: I work at the World Resources Institute with many colleagues who co-authored the GHGP guidance, though I have very low association with that project. I am acting on my own here.
Re: do you have any concerns that your users might shortcut the work to produce high quality estimates of their Scope 3 emissions when you have made it so easy to get lower quality estimates?
> I think the high order bit here is that 99%+ of companies don't measure their emissions at all. This is for a good reason — measuring your emissions historically has been quite labor-intensive. Even for large companies, there is always a 'long tail' of 'Scope 3 goods and services' transactions that are hard to measure. Our goal is to create a scalable solution so that a much larger share of companies are able to participate.
Re: your grocery store example —
> This is a fair point. Our main belief is that realtime, actionable data trumps perfectly attributed data, if perfectly attributed data requires a bottoms-up manual model. The advantage of the spend-based approach is that (1) it's realtime, and (2) it aligns incentives at the company level. The holy grail, however, would be itemized spend data (level 3 data), where you could factor in the emissions of your specific line-items. Unfortunately, that data is nearly impossible to get (yet). Maybe that's Bend 2.0 :)
Re: re-accounting historical emissions —
> Yes! We use the emissions 'factor' that most closely matches the transaction date. So for example, if you bought a Starbucks coffee in 2020, we would use the 2020 Starbucks factor, and if you bought a Starbucks in 2021, we would use the 2021 Starbucks factor. If Starbucks is late to publish their 2022 report, we would recalculate the emissions when the info is updated. For our category fallbacks, we also take currency / region into consideration.
Happy to chat more, either with you or the WRI folks! Thanks for the questions.
On your belief that actionable data _trumps_ perfectly attributed data, I'm not entirely convinced. I think actionable data _complements_ attributable data. But you need the attributable data to accurately measure the impact and learn what specific actions caused that impact.
I think all companies want to make actions that are well-informed, 'the right choice', and have the potential to demonstrate it was 'the right choice'. My concern is that when someone takes a 'good action' such as replacing high intensity animal protein with low intensity plant protein there is no evidence from your side that it made any difference. You are divorcing the actual choice that was made from what is perceived as the outcome.
The dollars-to-emissions relationship is just not as simple as is being represented, and for those who are not specialists there might be a false sense of progress.
"We cut our daily emissions from transport by having employees purchase Uber rides in off-peak times."
"We cut our emissions for business travel by setting up a policy that flights must be purchased at least 2 months in advance."
"We cut our emissions from our regular food purchases by looking at the local newspaper for coupons and signing up for a customer loyalty account."
"Maybe we all should fly to Las Vegas (tickets are cheap) rather than have Linda fly to NYC (an expensive ticket)."
Each of these might be smart business choices, but they have absolutely no real world effect on emissions that should be attributable to a company, but that isn't what the company is being told.
As a first pass to estimate sense of scale and where to look into unsustainable practices and prioritize better data collection, Bend seems to be valuable.
There's been some discussion about the accuracy of environmental footprint calculators. In particular, there are concnerns that such calculations don't take into account the consequences of choices made to reduce one's footprint, which may end up increasing, rather than decreasing, emissions.
For a brief introduction see:
Environmental footprint calculators have one big flaw we need to talk about
https://theconversation.com/environmental-footprint-calculat...
And in particular some research cited in the article, like:
Using Attributional Life Cycle Assessment to Estimate Climate-Change Mitigation Benefits Misleads Policy Makers
https://onlinelibrary.wiley.com/doi/10.1111/jiec.12074
Is your company aware of this discussion? If so, what are your plans to address those concerns?
At Bend, we are trying to really apply pressure on the market to be real about going "net-zero." Our offsets package is priced at $100 per ton, which is still surprisingly affordable for a lot of small companies. We blend that package across reforestation projects (low-cost) and true carbon removal project (high-cost) to offset a company's emissions, in order to funnel investment towards those more expensive operations. This is important so that companies like Charm Industrial and Climeworks can increase their efficiency and bring the price-point of carbon removal down.
And while $100/tCo2e might sound expensive, our thesis is that it's still remarkably affordable, especially for small companies who factor it in early on. This program is also opt-in, so you can start out with Bend just to measure your emissions and then decide about offsetting later... but for us, for example, our "Carbon Bill" hovers between just $25 to a $100 per month. Barely doubles the cost of the base subscription!
What's the easy, accurate, and fruitful way? Toting up the amount of fuel that comes out of refineries and coal mines multiplied by its carbon content.
Then, tax it, which will provide a disincentive for all uses that burn that fuel.
I actually wrote a blog post on the subject, if you're curious: https://bend.green/blog/the-path-to-a-carbon-tax
As I mentioned, tracking and measuring emissions is hopelessly complicated. It's far easier to tax the carbon content of the fuel where the fuel is produced.
As for taxing, we already tax gasoline (rather heavily). I don't really understand why taxing the carbon content is resisted so strongly. Taxing the carbon content would also do things like make natural gas more attractive than coal, because ng has twice the energy in it per carbon atom.
As for the amount of the tax, one starts out with a small carbon tax, then gradually increase it until the CO2 emissions go down. This also gives the market time to adapt.
It is impossible to determine the carbon footprint of a mere pencil, let alone some large business operation:
Actually, you don't "know" that, and the tCO2e for every single step in the supply chain is highly dependent on suppliers and situations. You can find an average for manufacturing a single glass bottle in Stockholm, and its shipping to Germany. But what if you used a small, low-carbon, shipping company instead of the regular transit options ? You would not know the proper tCO2e for that unless you already have specific data about this small shipping supplier.
It's completely recursive and mostly information-incomplete.
"Look at this lead pencil. There’s not a single person in the world who could make this pencil. Remarkable statement? Not at all. The wood from which it is made, for all I know, comes from a tree that was cut down in the state of Washington. To cut down that tree, it took a saw. To make the saw, it took steel. To make steel, it took iron ore. This black center—we call it lead but it’s really graphite, compressed graphite—I’m not sure where it comes from, but I think it comes from some mines in South America. This red top up here, this eraser, a bit of rubber, probably comes from Malaya, where the rubber tree isn’t even native! It was imported from South America by some businessmen with the help of the British government. This brass ferrule? [Self-effacing laughter.] I haven’t the slightest idea where it came from. Or the yellow paint! Or the paint that made the black lines. Or the glue that holds it together. Literally thousands of people co-operated to make this pencil. People who don’t speak the same language, who practice different religions, who might hate one another if they ever met! When you go down to the store and buy this pencil, you are in effect trading a few minutes of your time for a few seconds of the time of all those thousands of people. What brought them together and induced them to cooperate to make this pencil? There was no commissar sending … out orders from some central office. It was the magic of the price system: the impersonal operation of prices that brought them together and got them to cooperate, to make this pencil, so you could have it for a trifling sum."
https://thenewinquiry.com/milton-friedmans-pencil/
It's a hopeless task. You're never going to get an accurate accounting of carbon footprints by looking at the end result. Not a chance.
Presumably you can also measure the carbon footprint of buying carbon removal credits using Bend, and pay to remove that carbon, and measure the carbon footprint of buying those carbon removal credits, and so on forever...
It would be neat if carbon footprint could be translated to some more humanized unit. Like trees or global warming degrees or idk.
Also, it seems like some products at companies would be more carbon friendly. Would be nice if you could categorize the spending within a bank line item. Naive example… I bought a $10 reusable cup from my local coffee shop so that every time I shop there I am not using a styrofoam cup so let me tweak how that impacts my carbon footprint. That way I can still get coffee and feel good about doing so in a conscientious way. I’m sure there are similar big company examples, too ¯\_(ツ)_/¯
And yes, line-item and other "level 3" data is very exciting to us techies—unfortunately it's also really difficult to access! One of the reasons we like the Bend model is that it simplifies/abstracts away from that by looking at your total contribution to a company's bottom-line as the source of truth for emissions, which prevents a kind of "line item greenwashing."
But yes, in the future it would be nifty to sync in level 3 data from receipts or something to factor that into our calculations!
And that kids is how you not only not solve a problem, but introduce a bureaucratic feeld good layer that actually accelerates the problems.
Easier solution would be to just compute industrial product logetivity, aka how long it lasts as a usefull entity and tax accordingly. That would change things up, to reward a car lasting 40 years.
Another note — most cloud emissions only factor in the energy footprint ('scope 2' in technical greenhouse gas inventory terms). We believe this significantly undercounts emissions, because it ignores the capital expenditure of building the facility, buying all the machines, etc. The great thing about the spend-based approach is all this overhead is factored in. (BTW, Google Cloud Platform just started to layer in some of this 'scope 3' operational overhead data, but I believe AWS still ignores it, significantly undercounting emissions).
I'm from Seattle, but I've actually wanted to visit Bend for a while. Seems like a nice river town
That being said, the Greenhouse Gas Protocol reporting has been pretty inconsistent in the past. Companies cherry-pick and leave out important info, or define their 'reporting boundary' in inconsistent ways. One of our goals is to help 'debug' these inconsistencies.
Lucas Joppa from Microsoft laid this out quite well here: https://www.ted.com/talks/lucas_joppa_how_to_fix_the_bugs_in...
In contrast, if your company paid for your metro card or reimbursed for mileage or gas on your commute, we would estimate the carbon intensity for those expenses.
As startups, you have an unfair advantage — it’s much easier to build good carbon habits early, vs. retooling your business after having already invested large amounts of capital in carbon-intensive practices and depreciating assets.
We do offer enterprise pricing for API access for larger customers.
https://www.youtube.com/watch?v=1J9LOqiXdpE
Carbon offsets are a scam:
https://www.youtube.com/watch?v=EIezuL_doYw
Hearing people talk about these give me the same vibes as techbros talking about stocks and crypto.
If we don't start changing the standard for what's acceptable, who will, ya know?
Companies would be better investing in encouraging people to behave better on compus if they were serious, and I mean by actually investing to make using eco friendly things pleasent, not using brown paper to give printouts to clients or having some guy who shouts about the wrong paper going in the wrong bin all the time.
Companies should buy things that people want that produce less waste (yes this then drives cost up and profits down), consolidating physical services (it, printing, office space etc.) and using recycled products by preference (again good recycled paper is not brown toilet paper and can even be given to clients). And when companies start doing this rather than individuals it has a knock on effect on the market and affordability.
And finally, I'm sorry, I fail to understand why anyone needs to pay to be told this.
Yes, and the goal of carbon accounting is basically to give actual data about which of these things offer the better trade off between the goal of reducing emissions and keeping the lights on in your business.
The exercise is only as useful as the actions it enables. Short of absolutes ("just close shop ! The greenest form of transportation is, after all, the hearse") or "one size fits all" solutions, there is some level of accounting that's needed.
The biggest risk, of course, is that the customer hides all information about its emission and just wants you to package a shiny report to fuel the marketing department's greenwashing op.
I'll soon be working for one of bend competitors, it seems, so I'll get to see how it goes...
I've been lamenting for a while the sole metric that is being used for environmental impact is "carbon footprint", just so that bureaucrats and companies can have something to measure while appearing to do something.
PFAS and microplastics might be damaging our genetic material, but their very existence if meaningless to the CO2 metric.
A bad metric is worse than no metric at times.
