Does Sustainability Sell?
- Isin Acun
- 21 hours ago
- 6 min read
Sustainability is usually framed as a cost for firms and customers alike, but recent research tells a different story. An interview with Davide Proserpio written by Işın Acun, Ph.D. candidate at WU Vienna
For years, going green has been framed mostly as a cost, a virtuous thing to do that mainstream customers applaud but rarely pay for. A study built on Amazon’s data tells a different story. In The Impact of Sustainability Programs on Consumer Purchase Behavior: Evidence from Amazon, Davide Proserpio (USC Marshall) and co-authors Ali Goli, Tyler Mangini, Ken Lau, and Daniela Yu measure what happens when a product earns Amazon’s Climate Pledge Friendly (CPF) badge, the green leaf that has appeared in search results and on product pages since 2020, marking products with a recognized sustainability certification. The paper grew out of Davide’s two and a half years as an Amazon Scholar.
The question could not be simpler: when Amazon adds a product to the program, do its sales go up? They do. Using roughly 45,000 products, the team finds that joining CPF lifts gross merchandise sales by 13.3%, sales by 12.5%, and shipped units by 4.4% over the following twelve weeks. No experiment, Davide quickly notes; they just used the variation the program’s rollout gave them.
Investing in sustainability is not a waste of money, as maybe some could think. It’s actually something that can help you drive sales.
— Davide Proserpio
For a literature stymied by the attitude-behavior gap, meaning consumers say they care about the planet, but their wallets rarely follow through, this is an encouraging result. What makes this paper unusual is where it was written: this time, researchers could see everything sellers were doing behind the scenes. That access runs through everything that follows: the findings, the correction behind it, and the fact that the paper exists at all.
Why marketing activity can’t be ignored
Most academic researchers never get to see what sellers are doing. “As researchers we almost always work with imperfect data: review data, scraped sales data. You can’t see marketing activities, so you assume they’re the same across treatment and control. Which may or may not be true.” Inside Amazon, Davide and his team could observe, and control for, the marketing levers sellers were pulling alongside the badge: advertising spend, price changes, promotions and discounts. It turns out to matter a great deal. Sellers who join the program also step up their advertising and promotions, so the badge gets credit for sales the marketing efforts produced, inflating estimates by 20 to 30%.
Sellers are not naive. Because they invest to get into the program, they invest more in advertising too. If you don’t account for that, you’re conflating effects.
— Davide Proserpio
Also, they found out that not every product benefits equally from the CPF badge. Consumables experience the largest effects: beauty, personal care, food. “For products you put on your skin or eat, consumers already care about certifications. The sustainability label acts as another signal of quality, and that additional trust generates more sales.” He flags a caveat himself: they did not ask consumers directly, so the mechanism behind the observations is plausible rather than confirmed, raising an interesting follow-up question, he notes.
The conference dinner then took us somewhere I did not have on my bingo card: Bath Racecourse. Instead of a sit-down meal there were food and drink stands dotted around the track, and between them we watched a real horse race under reliably grey English skies. Later a very good DJ took over and worked his way from 60s and 70s classics all the way to the 2020s by the end of the night.
Getting a paper out of Amazon
The same access that makes the numerical measures credible is also what makes the paper rare. “Most people who leave academia to work for Amazon don’t expect to publish. Culturally, it’s very hard.” This paper began almost by accident: Davide who was working in advertising, got chatting with the Climate Pledge Friendly team, and a question surfaced. Can you take a look? He could. Preliminary results convinced the team to push, and after months of back-and-forth with leadership and lawyers, the green light came. The five authors each brought something different. Daniela Yu and Ken Lau wrangled the data, “finding the data is not super easy when you have data everywhere”.Tyler Mangini, CPF’s product manager, brought the question to Davide in the first place, and Ali Goli joined on a later stage and provided invaluable help. Industry colleagues kept asking why things had to be done the academic way, and answering them turned out to be useful practice.
Industry leaders care about whether the findings are true, why they’re true, and how to convey them.
— Davide Proserpio
To separate the badge’s effect from sellers’ own choices, the team exploited the program’s staggered rollout in a difference-in-differences design, comparing products before and after joining against those not yet in the program, while controlling for advertising, pricing, promotions, and reviews.
The review process at IJRM was rapid, too. Reviewers found the result interesting and largely believed the analysis, so their questions mainly sharpened the framing. The cross-country comparison, matching the same product across US and European marketplaces, appears as a supporting result because matching across countries yields a selected sample. From submission to publication took under a year, fast enough for Amazon to put a peer-reviewed finding to work in the program itself.
The takeaway is direct. For brands, certification is a demand lever, strongest in consumables. For retailers, clear sustainability labels turn stated preferences into actual purchases, especially for products consumers might otherwise overlook.
What comes next
Davide has since traded one inside view for another. He left Amazon in late 2025 for Airbnb, where he is now a housing scholar. This is a homecoming of sorts; one of the papers from his dissertation used Airbnb data that he collected in 2013. Now a new working paper, recently presented at EMAC in Bath this June, studies what happens when cities ban short-term rentals. The early results, he says, suggest the costs of a ban surface elsewhere in the local economy. Stay tuned for further insights from Davide.
If you ban Airbnb because of housing, something else may happen to a differen but connected part of the economy. Policymakers should carefully think about the trade-offs.
— Davide Proserpio
Read the paper: Curious how a sustainability label reshapes purchase behavior, and why controlling for marketing matters? Read the full paper here.
Want to cite it? Proserpio, D., Goli, A., Mangini, T., Lau, K., & Yu, D. (2025). The impact of sustainability programs on consumer purchase behavior: Evidence from Amazon. International Journal of Research in Marketing. https://doi.org/10.1016/j.ijresmar.2025.04.009
Meet Davide Proserpio
Associate Professor of Marketing, USC Marshall; Scholar at Airbnb
If you weren’t an academic, what would you be?
“I did many things in my life. I was a waiter, then an electrician, then a waiter again in Spain. Then I studied engineering, and during my PhD in computer science I pivoted toward marketing without knowing it. If I could do anything? Maybe spend a couple of years just biking around the world, doing races.” And he adds that students must find that path reassuring. “People always think you need a linear path. Students at USC come to my office; they’re 21, they tell me they don’t know what to do with their lives. I tell them: look, I still don’t know, so I think it’s okay.”
Best advice you’ve ever received?
“When I started my PhD, someone more senior told me: don’t go too fast, it’s a marathon, not a 5K. He was right; some people burn out very fast. My own advice is to find a hobby that takes real time and lets you disconnect. For me it was running, then triathlon (Ironman and then Ultraman), then mountain biking (I recently completed the Cape Epic in Cape Town, South Africa); but you don’t need to be that extreme.”
Lunch with anyone, dead or alive?
“Myself, twenty years younger. I’d ask him: so, what do you think? How do you think we did so far? I’d like to see how I would react.”
This article was written by
Işın Acun
Ph.D. candidate at the WU, Vienna
Comments