Towards carbon negativity

Humans have been pumping greenhouse gases into Earth’s atmosphere at an unsustainable rate. It’s on us to reverse course as quickly as possible to stay below the tipping point of 1.5℃ global warming. Without action, the future is beyond bleak.

At Basecamp, we’re committing to becoming carbon negative for our cumulative history and moving forward.

The first step of that commitment is understanding the size of our carbon footprint. We’ve just completed our first carbon footprinting exercise as a company and are publishing our results. By sharing our approach, we hope to make it easier for other companies to also join the collective mission to carbon negativity. 

In 2019, Basecamp’s estimated carbon footprint was ~820 tCO2e. Here’s how that breaks down:

🏘  Remote offices: ~5 tCO2e
Basecamp is a remote company which means even in the pre-COVID era, we mostly work from home. The power drawn to run our work computers and monitors are attributable to Basecamp. We roughly estimated power drawn for 50 Basecampers whose homes are not powered by renewable energy, a laptop + monitor/person for 40 hours/week/Basecamper, 47 work weeks/year, and a CO2e/kWH emissions factor from the US electricity supply average.

🛌  Event lodging and meals: ~20 tCO2e
As a company, we typically hold two in-person gatherings a year. We also occasionally travel to attend conferences for professional development or train new team members. The majority of the non-transportation emissions from these gatherings are tied to hotel stays, using emission factors from the Cornell Hotel Sustainability Benchmarking Index.

💻  Hardware manufacturing + end-of-life: ~25 tCO2e
There are emissions associated with the production and disposal of each laptop, monitor, on-prem server, etc. we use as a company. Those emissions can be prorated by the hardware’s expected lifespan for each year. Several manufacturers helpfully publish their own carbon footprint estimates for their devices so we used those specs to calculate the emissions associated with 56 laptops and monitors and 55 on-prem databases.

✈️  Transportation: ~75 tCO2e
Basecampers flew nearly 400k miles (644k km) and traveled roughly 35k miles (56k km) over ground combined in 2019 for meetups, conferences, and trainings. Those travels translate into roughly 63 tCO2e for flights and 12 tCO2e for ground transport based on EPA emissions data.

🗄 On-prem hosting: ~80 tCO2e
We run several of our applications in on-prem data centers. It takes energy to host the apps, to the tune of roughly 330,000 kWH in 2019. One of our on-prem data centers is powered via an emission-free energy supply contract. After excluding the power from that data center, we translated the power consumption from our remaining on-prem data servers using emissions factors from the US Energy Information Administration and global warming potentials from the US Environmental Protection Agency.

🏢  Company office: ~120 tCO2e
For the last ten years, Basecamp has had a headquarters office in Chicago. It’s a beautiful space that was the gathering grounds for our company meetups. Turns out, that office is also heated by electricity and we used a whopping 261,130 kWH in 2019. In Illinois, 60% of power is generated from fossil fuels so that consumption translates into more emissions than all of our company transportation in 2019. This realization was one of the most shocking findings from our carbon footprint. We’re saying goodbye to the office in July

📚  Printed books: ~135 tCO2e
Basecamp’s founders have written several books including REMOTEREWORK, and It Doesn’t Have to be Crazy at Work. Approximately 45k copies sold in 2019 combined. Using a benchmark of 3 kg CO2e/book puts the footprint of these books in the same ballpark as the company office emissions.

☁️ Cloud hosting (including data transfers and storage): ~285 tCO2e
We also use Amazon Web Services to run our apps, with a few supporting infrastructure on Google Cloud Platform. Cloud data centers use a lot of electricity, though there are efficiencies gained by scale. However, not all cloud services are equal from a climate change perspective. Since 2017, GCP has added enough additional renewable energysources to the grids its data centers are on to be considered carbon neutral. AWS can’t say the same thing

It’s tough to estimate the carbon footprint of AWS service usage as they do not share power consumption as a metric. So to estimate, we broke down data we do have — server hours used, GB data transferred, and GB stored — and converted to energy consumed and carbon emissions from there. We assumed a 30% discount in power usage for AWS’ servers, based on their own commissioned report on how much savings typically comes from facility efficiency and renewable energy usage. We then used an extrapolated research estimate of kWh/GB used to transfer data over the Internet. Overall, this method led to an estimated 580,000 kWH for cloud compute, 633,000 kWH for data transfer, and 140,000 kWH for data storage in regions that aren’t fully carbon offset in 2019.

There is a lot of uncertainty in this estimate, in both directions:

  • On one hand, AWS could merit a higher discount on emissions. Their current stated goal is to get to 100% energy offset by 2030 and a significant portion of the renewable energy projects they have invested in appear to be going into the grids that power our apps. We just don’t know for sure and so erred on the environmentally conservative side.
  • On the other hand, the extrapolated energy intensity we used for data transfer emissions could have been too generous. If there are any informed experts on the carbon footprint of data transfer out there, we’d love to learn from your work.

🐛 Wiggle room: 75 tCO2e
Since this is our first time doing a carbon footprinting, we assume we’ve missed something decently substantial. For instance, we have not gone through to footprint the upstream indirect emissions from other software vendors, many of which are also hosted on AWS. In an attempt to account for these unknowns, this wiggle room category is an additional 10% of associated emissions.

So in total, this puts us at ~820 tCO2e for 2019. We’ve also projected a footprint of ~770 tCO2e for 2020 and estimated a very rough historical cumulative carbon footprint of 6,000 tCO2e for 1999-2018.

But there’s one more unaccounted for unknown worth mentioning:

💰  Banking: ??? tCO2e
Money makes the world run and we’re cognizant that our banking partner could easily be the largest source of indirect emissions. In the course of our research, we found a 2008 report from the Rainforest Action Network that was very eye-opening on the potential footprint. We’re looking for more recent research or data that could help us factor in the indirect emissions from our deposit accounts.


In the next few weeks, Basecamp will purchase enough certified carbon credits to offset our estimated cumulative carbon footprint and make us carbon neutral. To get to carbon negative, we further invest at least a further $100,000 each year starting this year (2020) to sponsor new carbon offsetting projects. 

We will also place even more emphasis internally on the reduction of our own emissions: through our choice of vendors, our company events, and our software design. The environmental benefit of not emitting is more certain and immediate than the eventual carbon sequestration from reforestation projects and emerging technologies. We can emit less and we’re going to do the work to do so.

43 thoughts on “Towards carbon negativity

  1. Have you considered the trees that needed to be cut to print all your books? This is not directly related to carbon, but it hurts the environment nevertheless, and should be compensated for. Same for other kinds of paper used at the company, if any.

    1. Yes, the benchmark carbon footprint for the books included “fiber sourcing”, aka fresh trees and recycled paper. The book printing is the main source of paper at Basecamp: as a remote company, we simply don’t use that much physical paper for day-to-day work. In terms of compensation, we are looking at reforestation projects as a primary means of carbon offsetting currently.

    2. You silly, most (80%) of the trees are not grown for paper. Instead, they are grown for furniture and constructions. And to make room for… corn and soy beans used for vegetable oil production to enhance the gasoline and diesel.
      Trees grown for paper have a short lifespan, they are grown in an intensive way, the fibers are less dense and are not suitable for constructions and furniture.

      On the same note, if you are such a climate conscious, how often are you replacing your iphones and macs? :)) :))
      How many trees have you planted in your life?

  2. I’m curious which carbon credits you plan to purchase. Will you be sharing that?

    Have you considered investing in reforestation projects, natural farming, and other methods of direct carbon sequestration?

    1. Great question. I’d also be interested in this, as carbon offsets tend to be a bit of a… questionable business…

      I ended up buying personal carbon offsets through offset.earth, as they were Gold Standard certified and publicly release all of their financial information, but I would definitely like to know what you all end up settling on, too.

    2. Yes, we’ll be sharing a follow-up post after we’ve finalized the projects we’ll be sponsoring and where we’re buying the carbon credits from. Vetting is super important and that’s what we’re doing right now.

  3. Great initiative.

    Some remarks :

    – should you take into account the use of your products by your customers ? I don’t know what is the norm (if any) on this.

    – since anthropogenic emissions far exceed the amount of “offsets” available worldwide, purchasing carbon credits cannot be considered a viable solution on a large scale. So the reduction of your (our) emissions is definitely the way to go.

    – 820 tCO2e / 50 basecampers = 16.4 tCO2 per employee seems like a pretty high carbon footprint per employee (Well, I’m not an expert, and haven’t looked for comparative data). That’s interesting as we could naively think that a business like basecamp would be a low carbon intensive business.

    Thanks for sharing the data

    1. Thanks, Thomas!

      On customer usage: part of our cloud hosting calculation is the data transfer in and out from/to customers as they use our products, which has associated emissions. We are not, however, accounting for indirect emissions that our customers may be responsible for because of what they make or product while using Basecamp. I don’t believe that the typical boundaries for indirect emission accounting would include the latter.

      On footprint per employee: it is a high footprint per employee, though that’s also because we have relatively few employees running so many products. A couple of other software companies’ have published footprint accounting such as AutoCAD, Adobe, and Shopify in case you’re interested in comparing approaches.

  4. Carbon offsets don’t work. Take, for example, offsetting flights. Trees that sequester carbon don’t make the carbon disappear – they simply store it. When those trees die (as all trees do) the carbon from those flights is released into the atmosphere.

    The only way to prevent carbon entering the atmosphere via travel is not to travel. Doable, in a remote company.

    1. Hey Jeremy – totally agree that mitigation is a core responsibility and that’s part of our approach. Thanks!

    2. It’s too blanket to say ‘carbon offsets don’t work’. They are mostly missing the point, which is that actual reductions are required, however there are offsets, like carbon seqestration in basalt below ground, which _do_ work, in that they actual remove carbon essentially permanently from the atmosphere/biosphere/hydrosphere. Those are of course some of the most expensive offsets. There are plenty of other offsets which don’t involve trees, of varying degrees of plausibility. Solar panels plus LED lamps replacing use of kerosene for lighting do reduce emissions by about 330Kg/yr per lamp for example, whilst the lamp+panel+battery are something like 5-8 Kg and last a few years. This is only avoidance not sequestration but it’s a genuine change away from fossil fuels which is what we all need to do, and thus probably reasonable value for a very cheap offset. All offsets must be considered on their merits, and are no substitute for actually reducing your own emissions directly.
      All offsets must be considered on their merits

  5. Great initiative! I’d love to hear more about any goals around reduction in addition to what you’ve spoken about regarding offsetting – as you probably found in your research, offsetting can be a minefield of challenges regarding trust, additionality, tracking, etc.

    Given that, reduction is always going to be the most reliable way to take climate action at a corporate level 🙂

    1. Absolutely – and that’s why mitigation is part of our approach as well.

  6. Obviously cloud hosting is a huge part of your business and therefore a huge part of the environmental impact of your business.

    GCP has a much better environment record than AWS. I’d love to know more about David’s remarks on Twitter that GCP nearly burnt Basecamp to the ground, twice. I have no affiliation, just interested.

    1. Absolutely. We’re taking a hard look at the options for our cloud and on-prem hosting. GCP is undoubtedly ahead of AWS on the environmental front.

      Re: the challenges we faced with GCP earlier, we experienced a series of outages in early 2019. You can read some of the details in earlier SvN posts, e.g. /svn3/basecamp-outage-when-it-rains-it-pours/

  7. I would like to leave you with the suggestion of rather than buying carbon credits, you could potentially buy land to expand existing protected or similarly managed natural reserves. Harder to account for, but you could probably do it in a similar way as for the carbon credits, and it may be a bit more permanent solution? (with added biodiversity benefits) This is something I have been thinking about for some time, and would actually be happy to chat through the pros and cons with anyone that might be interested!

    1. Thanks for the thought, Luisa. We’re considering some options that are similar to what you’re describing.

    1. Hi Nate – at the company level, we took into account a rough estimate of the meat and dairy we consume. Diet is definitely a huge factor for individuals but for us as a company, the bigger fish to fry is reducing emissions related to running our products.

  8. Plants grow up everywhere, very quick and on zero human effort.

    That said, do you realize that when you help spread this fallacy of global warming you are actually condemning developing countries into a perpetual state of under development?

    Your nation grew and became rich doing one thing: consuming natural resources and polluting as a result of that, for centuries.

    Now, developed nations don’t want any competition by prohibiting, through political correctness, developing nations to develop themselves.

    What do you prefer? Condemn developing nations families into poverty or a pristine natural view to appreciate when you travel using the money generated by your heavily polluting activities?

    1. That’s a false choice to think the only options are wealth and destroying the environment or poverty and protecting the environment. Just to give one example, many developing countries are in parts of the world that receive a lot of sunlight. They can easily and cheaply generate most of their power from the sun. It’s win-win: more people get access to electricity and there are no carbon emissions.

  9. Thank you for making such a strong commitment to dealing with our climate crisis! Carbon offsets certainly aren’t the complete solution — but this will make a difference.

    1. Thanks, Adam! And agree: the most important part of the solution will be companies and people mitigating their own emissions.

  10. I would add home heating/cooling into the home office calculations. If people are going to an office then you won’t be heating/cooling the house as much in winter/summer. If you’re at home then you need the house at the right temperature 24/7. I found that when I worked from home our electricity bill was 1.5-2x normal in winter than when not WFH. Obviously this will vary greatly based on employee location but heating was a far greater expense than running a computer and monitor.

    1. Thanks, Glen – that’s a good point. We’ll work in that estimate moving forward.

  11. I would like to suggest that this money in carbon credits be destinated to help families in the developing world to survive the gigantic economic gap that you are helping to increase.

    If you want to save the world, start saving the people that lives in it. We dont want to migrate your country, we want equal opportunities to produce and use plenty of energy to develop ourselves.

    The nature will be here far over the humanity end.

    1. Hi Guilherme,

      I completely agree that the burden of mitigating the effects of climate change should fall primarily on the countries that have profited so much from emissions already. That’s true both for reduction of emissions and responsibility for managing offsets. For that reason, I’m particularly interested in vetting offsetting efforts that are based in the US and Europe.

      The enormous and widening economic inequality in the world is atrocious. I personally see these issues of economic inequality and mitigating the effects of climate change on humans as “and”, not “or” in terms of Basecamp’s responsibility to do more.

      Thanks for sharing your concerns,
      Jane

  12. Very interesting, thanks for sharing. As others have already raised, offsets are not really a scalable solution if we are talking about decarbonising the whole of society, but I’m glad to see you’re thinking about it and considering ways to reduce emissions too.

    Have you considered embodied emissions in your cloud hosting calculation? I.e. the emissions that it took to manufacture the hardware that is used to provide the service? I don’t even know where you’d begin with that but it seems like a relevant detail to me.

    Also, where providers (e.g. GCP) are claiming to provide green electricity, I suspect this doesn’t include the embodied emissions of the generation infrastructure. Once you’ve made a solar panel there are no emissions, but it takes quite a bit of carbon to manufacture the panel in the first place, and it won’t last forever.

    1. Hi Jon – on embodied emissions for cloud hosting, we haven’t but we should. We took into account embodied emissions for our on-prem servers, computers, and monitors. Thanks for pointing that out!

    2. All generation has an embodied footprint. For PV it is between 25 and 50g CO2e/kWh, depending on how a sunny the place you put the panels in is. For wind it is 6-12 (6 is recent value for latest huge 14MW offshore machines). For nuclear it is about 10. Hydro is quite variable 6-50, and can be massive (>1000) if the reservoir generates a lot of methane.

      All of these are ‘very good’ in comparison to gas: 470gCO2e/kWh and coal 800-900. Biomass is tricky. It’s about 1000 so even worse than coal, but nominally renewable not fossil so it very much depends how it is done.
      But yes, none are zero. https://en.wikipedia.org/wiki/Life-cycle_greenhouse-gas_emissions_of_energy_sources (Some slightly different numbers there – mine come from NREL plus some newer sources).

      In general any low-carbon source is a factor of ‘lots’ better than any fossil source. Wind and nuclear are best(beaten by some hydro projects). PV is not far behind (and both wind and solar are still improving every year on these measures).

  13. Would you consider adding your endorsement to Citizens’ Climate Lobby carbon fee and dividend approach? Carbon Fee on upstream GHG sources, thus downstream carbon energy-based commodities increase in price while non-carbon-energy commodities do not – a market based approach to developing more solar, wind, etc. as these become relatively cheaper. Fee is then rebated to every household in form of a monthly dividend check. That softens the increase in commodity prices, esp for low income. Border adjustments protect American companies and incentivize carbon fee to spread internationally. Right now HR 763, EICDA, has 85 co-sponsors in the House and it would reduce carbon emissions by 40% in 10 years – what we need to do per IPCC Oct 2018 report. CCL is also bipartisan and has lots of support from both ideological points of view (from env justice to conservatives). We need powerful legislation asap if we are going to have a livable/healthy world in 2050 onwards… and it has to be legislation that has widespread public support, ie that economically benefits low income/middle class especially. check it out. Would be great to see Basecamp endorse HR 763??

  14. It’s great to hear that you’re moving towards a reduction in carbon emissions from your side. The report is shocking, but it is even more shocking to think of all the other companies that have an even higher carbon footprint, of which there are many. My colleagues from http://www.worktime.com are all still working from home and I feel that thanks to remote work, our carbon footprint has dramatically decreased as we no longer need to commute. Whatsmore, I can stay at my family home in another state instead of visiting them every other weekend, which also affects carbon emissions. Did you include information about the energy your employees use when they work from home on their computers, WiFi, or coffee machines? Basically, all the energy that was used in one corporate office is now used in many households. So although your carbon footprint as a company is going down, your individual footprint may get bigger.

    1. Hi Mabel – yes, we did attempt to account for that in the “Remote Offices” section. It’s not perfect by any means and we’ll try to get more accurate in the accounting each year moving forward. Thanks!

    2. It seemed like you measured electricity at remote offices, but not heating/cooling. Perhaps this was a detail just missed from this article and you did actually include this. In winter in places where heating is needed it typically a factor of 4-6 more than electrical usage. Also cooling is the biggest chunk in places that need a lot of summer cooling.

      In theory heating one office (probably with the computers you were using anyway) is a lot more efficient than heating all the worker’s homes (unless they were already heated anyway). It’s quite complicated have these factors play out, and I’m not sure to what degree you have included this.

      Despite this quibble, I’m really pleased to see you publishing this data. _Every_ company should be doing this and working out what on earth they intend to do to reduce it to zero, or damn close, in a decade or two. Thanks for being open, and caring about our world.

  15. Hi Jane, it’s awesome that you guys have initiated the process towards a more sustainable/responsible business. I work for a SaaS company and we are about to embark in the carbon footprint adventure as well. First I have to say that you reporting on your results and the overall experience is SO VALUABLE to us. We have been struggling to get returns from SaaS companies that have done this. So Thank you!

    I have a few questions about the process:
    – What did you guys include in your Scope 1 ?
    – I am also very interested in your internal Action Plan to reduce your emissions ? Are you thinking about putting in place Sustainable Purchasing guidelines for your IT and electronic equipment ?
    – How did you get everyone onboard to evaluate your carbon footprint ? Were you regularly communicating on the topics of sustainability and CSR for a while prior to starting the process ?

    Thank you SO MUCH for your great insights and looking forward to hearing more about your sustainability initiatives at Basecamp !

    1. Hi Houda,

      Glad you’re also beginning your carbon footprint reduction journey! We need more companies to jump on board and we ourselves have been inspired by some of the earliest tech leaders in the movement like Stripe and Shopify.

      To answer your questions:

      1. We don’t really have Scope 1 (direct) emissions. Basecamp as a company doesn’t produce energy, we do not own buildings that combust fuel, and there are no company cars. The vast vast majority of our footprint comes from Scope 3 indirect emissions from along our supply chains.

      2. We’ll publish more on our internal mitigation actions in a later post. Right now we’re focusing on what we can do amongst the biggest drivers of emissions.

      3. Basecampers are a pretty conscientious crew and there have been many discussions internally about sustainability for a while. Several of us follow IPCC news and have also read the dire Uninhabitable Earth by David Wallace-Wells. Late last year, a couple Basecampers formed an informal group called “Greening Basecamp” and as the resident data analyst, I decided to take on the project of evaluating our footprint. Our founders, David and Jason, have been supportive throughout the process.

      Good luck with your efforts!

Comments are closed.