2. Creating Manifests
Manifest is a is an investigative toolkit intended for anyone interested in visualizing, analyzing, and documenting supply chains, production lines, and trade networks. If you want to jump directly to creating Manifest documents, see: Creating Manifests in Google Sheets and this overview of the Manifest Document Format.
For many investigators, the idea of documenting the complete supply chain of a product or a comprehensive examination of a trade network represents an ideal. It is, unfortunately, a largely unatainable one. Supply chains are not only complex—enrolling hundreds of sites and thousands, if not tens of thousands, of people—they are in a constant state of flux. Parts are revised, suppliers come and ago, workers are moved from one station to the next. The idea of a "supply chain" is misleading. These are supply networks. Each node might connect to multiple pathways, making it difficult—if not impossible—to establish which company may have manufactured a particular part, or which distributor delivered it to the next factory. Indeed, same product is often made at by factories, in entirely different places—with both material differences in their constitution and wide-ranging variability in their impacts. But let's imagine you ignored all of these challenges, how would you go about finding the whole story of the supply chain? Even the companies that make these products and the carriers who move them might only know the first tier of the supply chain (their immediate suppliers) and the list of the fulfillment centers and distribution hubs that they do business with. In rare occasions they might know the second tier (their suppliers' suppliers, the hubs connected to their hubs), but what about the third, or the fourth? The reality is that most companies don't fully understand their supply chain. Few logistics providers have the entire scope of the operation. In large part, this is by design. Supply chains work by abstracting out these details, replacing knowledge of real people and places with partial interfaces designed to sort them only by standard and price.
Manifest is not a supply chain management platform, and its users are not necessarily supply chain managers. Still, it might be useful to know some brief terms that are used in and around industry (and which we sometimes use in Manifest's documentation)
-
Tiers - Most supply chains are imagined as a series of tiers, based on a company's closeness to the final product or to the client they are working with. The first tier (Tier 1) is the top of the supply chain—it is the final product (or at least the final product that company delivers). Tier 1 suppliers are those companies that are direct suppliers of the final product, Tier 2 suppliers are those companies' suppliers or subcontractors, and Tier 3 are those companies' suppliers or subcontractors, and so on down to tier 4 and more—however many steps exist between the final product and its raw materials. A company selling a cotton T-shirt would describe the companies producing T-shirts for them as Tier 1 suppliers, the cotton mills supplying the Tier 1 companies would be Tier 2 suppliers, and the farms growing cotton to be sent to the mills might be Tier 3, and so on.
-
Upstream and Downstream - Related to the idea of supplier tiers is the idea of upstream and downstream. If you imagine the supply chain as a a chain, you could talk about the direction of movement within that chain from raw materials to finished goods and consumer purchased. Upstream refers to movement toward the raw materials of the supply chain, and downstream refers to movement toward retail and consumption.
We think that Manifest is better suited to telling partial supply chain stories, creators interested in documenting full supply chains look might begin by scoping the number of tiers they will follow, and then meticulously document a product's parts, the suppliers responsible for those parts, and the down through the successive subassemblies (and ultimately raw materials) that make up the chain. For the rest of us, it makes sense to begin by reflecting on the goals for our project and how Manifest fits into those goals.
-
What is the goal of your Manifest? : It is helpful to identify the main goal and any related subgoals for your Manifest document. What supply chain story are you most interested in telling? Are you interested in understanding the impact of a company like Amazon on urban spaces? Then you are going to be focused on Amazon, rather than other logistics competitors like Walmart or Home Depot. You are going to be looking more at things like warehouses and distribution centers than factories for Amazon Essentials products or even their (comparatively small) retail footprint, and looking for measures like facility size, employee numbers, cost, and so on. If you are interested in the environmental impact of a mobile phone, on the other hand, you might not be tied to a specific company, or even a specific set of manufacturing facilities—you might be more interested in looking at geographies of resource extraction and measures like embodied energy costs.
-
What time period does your Manifest document? : We sometimes talk about supply chains like they are fixed things, but the supply chain is a momentary instance, the singular crystallization of all the people and places, machines and materials, that go into the assembly and delivery of a particular thing. To write about the supply chain is to write about thousands, millions of these moments. If your investigation is focused on contemporary supply chains, what years are "contemporary"? This is sometimes a factory of data availability (there is often a data-lag of a year (or five) for certain data sets. Historic investigations might find it valuable to bracket their investigation by significant events, either in a broader historical context or circumstances that would have had a significant impact on production and distribution. The US entry in the Second World War, for example, forced many companies to shift production toward war efforts, and radically altered consumer behavior. Smaller, localized events, like the spinoff of electrical production for Bell telephone manufacturer Western Electric, can mean similar divisions in supplier relationships between two time periods.
-
What is the scope of your Manifest? : Related to this, you should set the scope of your Manifest. You should decide if you are focused on a single company or a sector, if you are limiting yourself to a broad geography like the United States or Asia, or a specific one like a single state or province. If you are doing a contemporary investigation, how many years back will you look, or how current do you want your information to be? If you are doing a historic Manifest, what pivotal events help to constrain your period of investigation? We can't look at everything. Do you need to tell the manufacturing story of every part of an object, or would a single part be enough? Does it matter if it's a particular company? Does it matter if it's general or specific?
-
What are your sources of data, and what are the limitations of those sources? : A company like Apple publishes suppliers for all the products it makes, but it doesn't differentiate them by product. If this is the only source you are working with, telling the story of Apple's general supply chain becomes much easier than telling the story of the supply chain for the iPhone, and much easier than a specific model of iPhone. More historic investigations may be forced to contend with the limitations of their archives. While production and distribution became increasingly systemized over the course of the late 19th and early 20th century, for example, many of the operational records that would be most useful in reconstructing these processes are poorly represented in the archives.
-
How will you organize your data? : Manifest is intentionally open to a variety of uses. At its simplest form, a Manifest document is a list of nodes. But what do those nodes actually represent? They could represent suppliers, factories, or all kinds of locations. But what do they represent in your document? Making these organizational choices first can also help with scoping your Manifest. A Manifest documenting all the locations and geographies enrolled in the Manufacture of a product could include factories, but it also might include nearby communities, the homes of workers, the design offices of the company that led to the creation of the supply chain, and so on. Saying a Manifest is a map of "factory locations" where every node represents a clear site of production: a factory, workshop, or assembly site, makes clear the scope of that investigation. There isn't a single approach that works for every project. For some investigations, nodes as "buildings" makes sense. Other projects might want nodes that include things like individual actors (from CEOs and VPs to factory and warehouse workers), or nodes might be better thought of as materials, or even abstract ideas (where does the product's "modularity" come from, etc.).
Another common organizational issue is how you might deal with repeat entries. Many Manifest documents have nodes that are similar in name, description, and other details—sometimes varying only by location or measure. Do these nodes repeat information like a detailed description? Or perhaps the first node includes a detailed description, and the rest just reference this. For example, a Manifest that includes a large collection of Retail sites might provide a detailed description of the product's retail strategy for this channel in the first listed node, and a something like: "another retail site: for more information on retail see the entry for the company's flagship retail store."
One of the most common questions about building a Manifest is about the research process for supply chains. How do you go about finding the information about where materials come from? How do you know what the sites of production and distribution are? The answers to these questions vary considerably from project to project. For historic Manifests, these can be incredibly specific—with the geographies and nature of the Manifest document indicating relevant archives of interest. For more contemporary Manifests, investigators might benefit from documents detailing infrastructures of production and distribution, tear-downs that reduce product to their constituent components (with initial information on suppliers, part numbers, and similar identifying information), and descriptions of the underlying raw materials necessary for a product. In either case, the Supply Studies Research Network provides listings of potential data sources and tools for supply chain research.
Still, the idea of having a complete representation of a supply chain or trade network remains. While
This section details types of Manifests you might create. It is important to remember that this attempt at categorization is intended as a conceptual tool: to suggest some possible ways Manifest documents might look. They are not mutually exclusive (you could have a narrative Manifest that includes infrastructure and draws from supplier reports, a Manifest that is both heavily quantitative and richly narrative that includes commodities and components for a particular product, and so on.
-
Complete Manifests - In all but a few circumstances, a "complete" Manifest is largely an aspirational goal. Even for the most powerful companies on the planet, it is rarely possible to produce an accurate accounting of their supply chains. The supply chain is an object of unknowing, a tool of abstraction. Its entire purpose it to shield each successive tier from the complex exchanges required to complete one part or to conduct one process. Supply chains shift and churn constantly. Companies come and go, designs change in subtle, rippling ways, as the people who mine, assemble, or ship them come off the line, climb out of the mines, or turn off their trucks for the night.
-
Partial Manifests - Most Manifests are partial, and Manifest is designed, primarily, as a platform for telling partial supply chain stories. If your Manifest is only part of a much, much greater network, you are exactly the kind of investigator we imagine working with this platform.
-
Contemporary or Historic Manifests - All accounts of supply chains will, in a certain sense, become historic. But we generally distinguish between Manifests that use historic research methods like archival research, and Manifests that use research methods only available to contemporary researchers (such as ethnographic investigation, interviews, and similar approaches). Again, there is not always a clear distinction. Some Manifest might rely on "oral histories." Corporate records and supplier reports available on websites might be from this year or ten years ago. One method of distinction might be that, for Contemporary Manifests, new data is still be creating—new reports, new employees being hired, etc. For Historic Manifests, new data might be discovered, but it isn't being created.
-
Narrative (Qualitative) Manifests - All Manifests are narratives—they are all stories about supply chains. But some are more "narrative" than others. Some Manifests (such as the historic Western Electric Telephone Manifest) include rich descriptions of nodes, with quotes and comments from sources like companies or journalists about materials, suppliers, or other nodes. They might also include images (sometimes many images) or videos associated with the node. These Manifests help to bring these nodes into focus—revealing the kinds of operations, geographies, and people they represent.
-
Quantitative Manifests - Manifest supports quantitative measures, and some projects do indeed have detailed quantitative measures for the nodes in their Manifest documents (for example the Vinyl Pressing Plant). This might be things like square-footage of facilities (for example, on the Amazon Fulfillment Center Manifest, the number of people (workers, consumers, etc.) enrolled at a given node, the impact of the node in terms of CO2e, electricity usage, water waste, and so on. These Manifests allow detailed comparisons to be made across specific measures.
-
Component/Constitution Manifests - Sometime just knowing what materials, sites, or people make up a particular product or part can be a valuable exercise. So too, can the general material demands of a product category (the "typical" materials used to make a car, a tablet, or a phone, for example). An component/constitution Manifest, like the Typical Laptop Computer, is a kind of "ingredients list" that provides readers with an understanding of what is inside a product and where it comes from, without necessarily detailing all of the steps required to document the complete supply chain. Sometimes useful (if general) measures, such as carbon footprint, water usage, and so on, can also be included, and this can be particularly useful to communicating the expected impact of a particular consumer choice, even for a general product category. Sometimes similar to a General/Sector Manifest, these are usually more specific because they focused on a specific product or product category
-
Upstream/Downstream Manifests - One kind of Manifest might look at one end (or both ends) of a product or product category, either raw materials (the bottom—upstream—of the supply chain) or top-level manufacture, final assembly, and retail (the top—downstream—of the supply chain) (for example, the Champion Petfoods Manifest or the (Nissan Supply Chain Manifest. This is often because it is the middle of the supply chain—the countless companies providing the various parts, engaging in various manufacturing processes, etc.—that are less well known and more difficult to pin down. Information about the top of the supply chain may be available in great detail (for example, this detailed listing of H&M Retail Locations, with the downstream operations of companies like Apple, Amazon, or Nike, comparatively well documented. Information on the very bottom may also be readily available—though it is usually general, rather than specific. While it is possible know that a particular product requires certain metals, for example, one may not know which particular mine (or even which country). Many Manifests, then, may identify only potential upstream sources based on general trends, production totals, import/export statistics, and similar indicators. Rational assumptions might be valuable indicators here that narrow the generality (with a node description that might read: "this country is the source of 90% of this material used in the electronics sector" or "this mine is the supplier of 60% of this mineral in this given country, which we know our target sources from according to their recent sustainability report").
-
Supplier/Client Manifest - Related to Upstream/Downstream Manifests, these supplier (or client) Manifests start in the "middle"—focusing on cataloguing the suppliers (and sub-suppliers, etc.) or the clients and partners of a particular company (for example, the Apple Suppliers Manifest). This Manifest type may be useful to consider from a data perspective because it has become increasingly common for certain kinds of companies to provide a glimpse into their supply chains. In addition to labor and environmental reports, this can include more detailed lists of their suppliers. Creating a manifest based on these listings can be valuable—it can provide, for example, a detailed representation of the various geographies, countries, cultures, etc. enrolled in that companies operations. It can also be problematic. These supplier reports are almost always partial, and they are rarely broken down by particular products. For companies with limited or specialized offerings, this could be a partial analogue for the supply chain of a particular product, but this is unlikely to be the case for companies with a large and diverse range of products.
-
Hybrid/Synthetic Manifests - Sometimes there can be value in creating a Manifest that incorporates multiple supply chains together. For example, it might be difficult to research, with any certainty, the supply chain for a particular model of disposable camera. A hybrid or synthetic Manifest might still aspire to document that model, but it might include nodes that are based on research from other models from the same company (with a node description that might read: "this firm is a supplier of a similar component for another model...") or other companies altogether (with a node description that might read: "this firm is a large supplier of this component for my target's largest competitor). Depending on the context, these could help "fill out" the Manifest with relevant details from sections where little direct information is available. Often, hybrid manifests will also make rational assumptions (with a node description that might read: "while the exact camera sensor supplier for this device is uncertain, this firm provides 70% of the sensors used in the industry, and is a known supplier for similar products by competitors").
-
Category/Sector Manifest - These Manifest expand on some of the approaches used in Hybrid/Synthetic Manifests to look not at individual products, but entire product categories or even sectors. These might be maps of the "disposable camera" or "hard disk" industry, for example, rather than a model (or two) (see the Hard Disk Drive Supply Chain Manifest). They are distinguished from Component/Constitution Manifests by not limiting themselves to the raw materials of a product or product category. Instead, this might be a Manifest that lists the major suppliers of various subcomponents or parts for a product category. For an entire sector (say electronics) it might list things like the major chip providers, major battery manufacturers and factory locations (or geographic regions) and so on..
-
Facilities/Infrastructure Manifest - While many types of Manifests focus on documenting the relationship between sites up and down the chain, we can also think about documenting a particular "slice" of a broader network. One approach to this is to document relevant logistics infrastructures. This includes things like warehouses, factories, and similar sorts of facilities. Some example of this are the Manifest of Amazon Fulfillment Centers or the Vinyl Pressing Plant Manifest.
-
Commodity/Production Distribution Manifest - Another kind of network "slice" could be focused on documenting sources for commodities or the movements of commodities (such as the Uranium Distribution Manifest). A related type of Manifest might focus on things which—while not necessarily something we might think of as a commodity—have a similarly well-defined (often constrained) production landscape (such as the Manifest of Algae Biofuel Producers).
The final—if not ongoing—step in the creation of a Manifest is to carefully document both sources for where the information for a given node comes from (including where any images or particular measures come from) and the assumptions you've made in the production of your Manifest. All Manifest documents have assumptions, and we should be transparent about what they are. Perhaps you know that a certain company provides 90% of a certain material worldwide, and you know it is a supplier for your target, and you know that your target uses that material. But you don't know for sure that they supply that material to your target. It might reasonable to assume they do—but you should be transparent about this assumption and why you think it is a reasonable (or useful) inference to make.