Is Uber even part of the sharing economy? Does Airbnb’s business model have a positive or negative impact on communities? Are on-demand startups like TaskRabbit leveraging exploitative business models? What is the right approach for local governments when it comes to regulating the sharing economy?

These are just a few of the persistent questions regarding the new range of businesses reshaping our business landscape. Some prefer to see the world as black and white: “Sharing will save the world by reducing our environmental consumption while solving income inequality” versus “sharing platforms are really just a share-washing platform version of capitalism for the 1%.” But in reality, the emerging sharing landscape is very complex.

To begin to get under the hood of the business models of sharing-economy players, my colleague, Pablo Muñoz and I analyzed hundreds of sources of data on 36 different sharing business startups representative of Jeremiah Owyang’s Honeycomb model, a graphical depiction of the different sectors where sharing startups have gained traction. While the Honeycomb model has been of great use in framing the diversity of sectors being impacted or disrupted by the sharing economy, it does not provide any insights on the underlying business models across the 12 different sharing-economy sectors that Owyang identifies. So we identified six key dimensions of sharing-economy business models, each of them with three distinct decisions that can be made by sharing startups. We converted this into what we hope is a useful tool, the Sharing Business Model Compass.

Four of these dimensions–transaction, business approach, governance model and platform type–offer business model decision choices roughly on a continuum from more market-based sharing (i.e. platform capitalism) towards commons-based sharing (i.e. platform cooperatives). While the other two dimensions–technology and shared resources–have decisions relatively agnostic to market or commons orientations.

If you’re thinking of starting a sharing-based business, we believe this tool is useful in demonstrating that there are a plethora of key decisions unique to the sharing economy that entrepreneurs must make along the way. If you’re thinking about the sharing economy in general–and how it interacts with people, government, and the world–this framework can be helpful in explaining what specific parts of the sharing economy you find optimal–or objectionable.

Within this dimension, there are three choices: tech-driven, tech-enabled, and low/no-tech. Tech-driven business models are ones leverage technology not just to connect users but also to complete the transaction without the need for offline interaction. Crowdfunding sites like Kickstarter and Indiegogo and online learning portals such as Udacity and Coursera fall in the tech-driven category. The majority of the startups we studied fall more in the tech-enabled category, which represents business models reliant on technology to facilitate the connections but require or are enhanced by offline interactions. Uber would fall into this category, as would Wallapop, a fast-growing hyper-local mobile classified business for P2P reselling of used goods. While most of what we think about in the sharing-economy space is at least enabled by technology, there are still many startups in the sharing space where technology is at most a supporting-tool but not critical (i.e. low/no-tech). Co-working spaces and shared commercial kitchens are good examples of the low/no-tech business model, while other models such as the Repair Café are hyper-local, nonprofit, no-tech sharing actors.

We observed three variations in transaction types: market, alternative, and free. Perhaps the most extreme version of market transaction is Uber, which does not only vary pricing based on demand, but also includes the highly controversial surge pricing. Most of the scaled, venture capital-backed sharing startups–such as Airbnb and Rent the Runway–opt for market pricing. The alternative option is jut starting to emerge. While there are perhaps fewer scaled examples so far, Bliive, a rapidly globalizing time bank out of Brazil is a good example. Instead of on-demand models that seek to facilitate economic transactions between a service provider and a home-owner, time banks allow service providers to earn “time dollars” to be exchanged later for other service provided at a later date from someone else in the community. Yerdle, a P2P service for exchanging used goods also falls into the alternative category. Instead of exchange for cash, users earn Yerdle dollars for acquiring other goods in the future. Peerby, another P2P used goods exchange, founded in Amsterdam in 2011, offers a completely free service. Of course, many bike-share services are free or just require an annual deposit. While you may think that free bike-sharing services have no business model, and perhaps are just subsidized public transit options, think again. Most of the scaled bike-sharing services in cities like Paris, Mexico City, London, and New York City, generate most of their revenue through sponsorships or advertising models.