Considerations of Current and Emerging Transportation Management Center Data

Chapter 3. Current and Emerging Private-Sector Business Models

Private sector providers currently use the business models and delivery mechanisms displayed in table 13. Each one of these business models can have sweeping impacts on pricing, contracting, and agency use. These business models and delivery mechanisms are not mutually exclusive. For example, blending "perpetual ownership" with a business model that includes "archive only" means that the agency would have perpetual ownership of archived data, but not necessarily access to realtime data.

Business models determine how revenue is generated or how the private sector generates value from their product. Data providers may sell their products directly to agencies or to a third-party vendor that then sells the data to an agency. Private sector vendors may instead receive payment through advertisers.

Table 13. Industry data procurement and sharing business models and delivery mechanisms.
Model/Delivery Mechanism	Explanation
Direct Sales to Agency	Private sector vendor sells data directly to agency. Vendor can usually offer the lowest possible pricing for a given data set.
Third-Party Resellers	Data provider sells data to an intermediary company who then sells it to end customers. When a third party sells data to an agency, it is difficult to know the extent of additional markup or overhead. Both the originating data provider and the third-party reseller must remain profitable.
Advertising	In rare cases, agencies will receive data or services for free (or at a reduced rate) because the private sector gains revenue on the products through advertising services. Agency-branded websites (like 511 systems) usually provide these services. Many of these systems are falling out of favor. It is growing more complicated to sell advertising on agency websites that are falling in popularity—primarily due to the rise in popularity of newer navigation apps on smartphones.
Integration with Ancillary Products (e.g., Visualization Tools)	Adding additional services and products to data can make it more valuable, and can make the sale easier to digest by agencies. For example, a department of transportation may procure a new advanced traffic management system or 511 system from the private sector, and as a requirement of the request for proposal, may bundle third-party data into the contract. Similarly, a data provider may bundle data, analytics on the data, and other ancillary products together. Some data providers will offer realtime analytics, visualization tools, tiling services, and more. Each one may not have enough value by itself, but when bundled, they have greater earning potential.
Trade and Promote	Some providers are less interested in revenue directly from the agency, and are more interested in integrating received agency data back into private sector systems resold to others. In addition to providing private sector data to the agency in exchange for access to agency data, the private sector may also require that the agency prominently promote the private sector's products and provide significant, visible attribution.
Shared Revenue	The agency agrees to one or more revenue generation models (like advertising on agency-owned websites), but shares or outright relinquishes the title to any revenue generated from such advertisements.
Realtime Use Only	In this model, agencies use the data for realtime purposes only. The agency cannot archive the data for later use.
Single-Purpose Use	Single-purpose licensing may allow an agency to purchase the data for use in only one, usually very limited, application area. For example, licensing a probe speed data set for use in identifying potential incident locations would disallow the agency from using the data for congestion monitoring or performance management.
Archive Only	While it may not seem applicable, archived data can be incredibly useful for realtime operations. Newer decision support systems that leverage modeling and simulation or machine learning can benefit from an archived dataset. Archived data can train systems, help in prediction, and more. The private sector may sell access to archives of historic data at a lower cost than realtime data.
Limited Geography and/or Time Ranges	Even though data may be available for the entirety of the State's network, vendors will nearly always charge less for access to data that is limited in geographic scope—perhaps only to select corridors and/ or only during certain times of day.
Leased Equipment, Dual Use	Roadside equipment manufactures install and maintain the equipment at some cost to the agency and allow the agency to use the data for one or more purposes. However, the owner of the equipment may be entitled to resell the data to other third parties.
Perpetual Ownership	The agency buys all rights and the title to the data in perpetuity. Even after the contract with the data provider is complete, the agency still owns the data and can use it for other purposes, but cannot resell the data to others.
Single-Agency Use Only	The agency can only use the data for itself. Partner agencies, like metropolitan planning organizations, research institutions, and neighboring departments of transportation, cannot use the data.
Hosted and Aggregated	Only the vendor can store the data in the cloud. Aggregated data is available via application program interfaces, websites/visualizations, or other methods. The agency or application does not get raw data. The agency is otherwise unable to integrate the data into native systems.

To illustrate how these different business models affect cost and acceptable use, consider the following agency use case in table 14. This compares the procurements of probe-based speed data from HERE Technologies by two separate departments of transportation (DOT). Both agencies are purchasing identical data from the exact same data provider. However, each agency took a separate approach to procurement and business models.

Table 14. Comparison of Florida and North Carolina procurement strategies for probe-based speed and travel time data.
	Florida Department of Transportation (DOT)	North Carolina DOT
Purchased Data	Realtime probe data (speeds and travel times).	Realtime probe data (speeds and travel times).
Business Model and Contracting Approach	Procurement directly by the State of Florida. No provisions in the contract for sharing with partner agencies outside of the State.	Procurement through the I-95 Corridor Coalition's Multi-state Vehicle Probe Data Project. Data sharing agreement requires data to be available to any agency in the I-95 Corridor Coalition (not just the department of transportation). No limits on non-commercial use. Perpetual ownership.
Impacts to Business	Limits ability to share information across borders (or see data from neighboring States). Disallows multistate performance monitoring, cross-border congestion analytics, or cross-border traveler information services.	No information found.
Cost Implications	Lower cost.	No information found.

Another illustration, provided in table 15, is for two agencies that procured O-D data and services from two separate vendors. Each vendor has unique business models that drastically affected cost and capabilities.

Table 15. Comparison of origin-destination data procurement strategies for probe-based speed and travel time data.
	New York City Department of Transportation (DOT)	Virginia DOT
Purchased Data	Origin-destination (O-D) data (including breadcrumb trails) from INRIX.	O-D data through Streetlight.
Business Model and Contracting Approach	Provides raw O-D data (including breadcrumb trails) to the agency for unrestricted use. Perpetual ownership.	Single-use data license. Data provided for a specific study/study area. Data cannot be stored and re-used for other studies. Data provided through an on- line application in aggregated forms along with reports and visualizations.
Impacts to Business	Because of perpetual licensing and unrestricted use, the City owns the data for use in any study and can share the data with partners. Because it is raw data, it is up to the agency to decide how to use it. Analytical tools are not provided.	Use was restricted to a specific study for a specific date range. Data cannot be used for other studies. Because this was project-specific, procurement was easier and cost was significantly lower. Because an analytics platform offered the data, the data was initially easier to use and interpret than raw data would have been.
Cost Implications	Higher initial cost, but potential higher value.	Lower initial cost. Conducting a significant number of studies over many date ranges could increase the cost beyond that of the raw data purchase.

Data providers offer many different business models when selling their data and services. Most will also lower the cost of their data if it is restricted to single-purpose/single-study use. The examples above are more indicative of agency procurement strategies and how the business model chosen impacts long-term capabilities and costs.

Pricing Structures

The pricing structure of data varies significantly between businesses and can depend on the procurement method and type of data or service provided. Examples include:

Pricing data based on centerline miles.
Pricing data based on population of the covered region.
Pricing based on the number of calls (or requests) to application programming interfaces (APIs).
Pricing based on number of users.
Pricing based on number of studies conducted with the data.
Discounts for bundling of data or services.

These models vary. Some data providers may charge based on mileage of the National Highway System only and then offer arterials at lower prices. It is difficult to find published, stable price sheets on data products from the private sector due to several factors:

Costs change over time: Commoditization can result in some data products losing value. Pricing sheets today would become obsolete in 1 year or less. Other data sets may grow in value over time. Brand new data products may be difficult to sell as convincing an agency of the value of the data can be difficult. The private sector may initially offer the new data products at substantially reduced rates to get the public sector interested in them. The price can change after determining the actual value.
Persistence: Vendors do not want old pricing sheets circulating through agencies as they may be misinterpreted or considered current when pricing has changed.
Competition: In the low-bid marketplace, pennies can mean the difference between winning and losing a contract. The private sector is wary of the competition learning about their pricing for fear that their competition will use that against them.
Third-party markup: When selling data or services through a third-party or bundled in procurements with other contracts, the prime consultant may charge additional overhead, markup, or profit on the data and/or services.
Requests for proposal and low-bid contracts: Low-bid contracts accepted by agencies can disrupt marketplaces and result in constantly changing price sheets. This is because low-bid contracts make it difficult for data providers to win contracts based solely on the quality of their data or services and force providers to compete on price alone. This has many other negative side effects as well, including the reduction in innovation and reinvestments that providers may make. If a provider knows it will largely be judged on price alone, there is less incentive to propose an innovative and leading-edge solution.

Revenue Sharing with Agencies

The private sector can view shared revenues from data sharing as a marketing strategy rather than a viable business model. Many agencies have experimented with exclusively sharing speed, volume, and incident data in exchange for traveler information services to the public through 511 and other means. Revenue can be acquired through sponsorship of 511 systems, and while revenues that go beyond operating costs are supposed to be sent back to the State, all too often this income either falls short of predictions or any small amounts of revenue gained are reinvested in the system. Many revenue sharing schemes surrounding data sharing still end up costing the agency as much as (if not more) than traditional contracts.

Parking, tolling, and automated enforcement (speeding or red light running) are the notable exceptions to revenue sharing success. However, the business models that make these relationships a success have less to do with data sharing and more to do with contracts that enable the private sector to operate (or install) parking, tolling, and enforcement infrastructure for the agency. Sometimes the agency pays for these services directly, and other times the concessionaire pays for the entire infrastructure and shares a percentage of the revenue collected with the State.

Exclusive Access

A few agencies have entered into exclusivity arrangements with the private sector. For example, the private sector may offer to redistribute an agency's closed-circuit television (CCTV) video to partner agencies and the media at no cost to the agency. However, in return, the private sector will request exclusive access to the video for sharing with the media. The private sector then has exclusive rights to essentially re-sell the video to television stations, third-party traveler information providers, or other non-public agencies.

The Value of Agency Data

Most private sector data providers still have a desire for agency-produced datasets; however, the type of data deemed most valuable has changed. Table 16 shows examples of data types and their perceived values as considered by the private sector.

Table 16. Agency data value as described by private sector companies.
Agency Data Type	Value Narrative
Volume and Speed Sensor Data	Used to be valued by traveler information providers, the media, and third-party probe data. However, the growing penetration rate of probes (cell phones and Global Positioning System (GPS) data) has made this agency data unattractive.
Planned Event or Lane Closures Data	Still valued by media, third-party data providers/aggregators, and others as the agency is essentially the only source of this data.
Agency Advanced Traffic Management System/511 Incident Data	With the improvements in crowdsourced data collection, this dataset is slowly losing value, but many companies want it since agency- provided incident data often has information that goes beyond the basics of location and incident types. Agencies that provide more detailed information about which lanes are closed and which responders are on the scene are viewed has having greater value.
Computer Aided Dispatch Data	Considered the "gold standard" by both agencies and the private sector alike, not for the completeness of the data, but rather for the speed and coverage that are provided. The private sector still considers this data valuable even though it can be difficult to interpret.
Driving Records/Division of Motor Vehicles Data	Insurance companies find this valuable.
Police Crash Reports	Insurance companies find this valuable.
Live Closed-Circuit Television Feeds	Valued by the media and third-party traveler information providers. Studies have shown that the public chooses its local television and radio news programs based off the quality of its weather, traffic, and sports coverage (in that order). Broadcasters can raise advertising rates when higher quality traffic video is available. Some broadcasters and third parties even install and sell their own camera feeds.
Parking Availability	Traveler information providers, vehicle manufacturers, and parking services value this data. While many private companies are the main providers of this data, there are cases where agencies own or manage this data and can provide it to the private sector for use.

Use Cases of Value Exchange for Agency Data

As part of their connected citizens programs, public agencies are swapping data on a large scale with Waze; that is, no money is directly passing from the agency to Waze. Waze is one of the few private sector companies offering to swap data with public sector agencies. Waze offers agencies access to their crowdsourced incident data in exchange for access to agency incident data, attribution, and co-marketing of the Waze platform. While the Waze acceptable-use terms are evolving, some agencies are seeing restrictive data sharing agreements from Waze, which is limiting the value of the data for anything beyond realtime operations.

Several State DOTs have tried to generate revenue through marketing and selling CCTV video feeds as a subscription-based service. Several States have also arranged for in-kind services, which seem to have had more success than the subscription-based services. Some years ago, the Florida Department of Transportation (FDOT) had evaluated efforts to sell CCTV camera feeds to media partners based on a subscription service model dependent on the size of the market in question. For example, a larger market like Orlando would generate more revenue and be more desirable than the smaller Jacksonville or Tallahassee markets. FDOT ultimately decided against pursuing this effort, although it continues to provide CCTV images to media partners through written use agreements.

FDOT recently entered into a co-sharing and license agreement with Harris Corporation. Harris has developed the Helios platform, which enables traffic and other cameras to function as weather sensors providing better situational awareness. In an exchange of data, FDOT will grant Harris access to CCTV video images. In turn, Harris will allow FDOT to access the Helios web service. Harris also provides a web-based API in order to retrieve the Helios Analytic Observations or the weather condition observations. The agreement stipulates that API requests cannot exceed 100,000 per month. The weather condition observations gathered will be integrated into the FDOT FL511 site and provided to the regional transportation management centers (TMCs). As an additional contractual term, FDOT must provide credit to Harris for the use of Helios analytics. Neither the FDOT nor Harris Corporation is charged any fees under this agreement.

The State of California makes its CCTV cameras available to the public through the use of an application that is downloadable from their website. Similarly, the State of Oregon provides its CCTV feeds free of charge, making the data as accessible as possible. Media outlets and interested parties pay for any connection necessary to access the information.

The Utah Department of Transportation (UDOT) has a non-exclusive agreement with Total Traffic & Weather Network (Total Traffic) for traveler information dissemination in exchange for value in-kind services. UDOT grants Total Traffic access to the traffic operations center (TOC) for routine radio traffic reports from 5 a.m. to 11 p.m. Total Traffic provides and maintains, at its own expense, equipment and communications infrastructure necessary for the operation of its travel information dissemination activities. The equipment and communications infrastructure used by Total Traffic must not interfere technically or operationally with the UDOT system at any time.

During the radio traffic report broadcast, Total Traffic must mention the UDOT TOC or use any other agreed-upon language providing attribution to UDOT. As part of this agreement, UDOT also receives a minimum of 20 30-second commercial announcements per Clear Channel station using the traffic reporting services per month to advertise CommuterLink™ activities. In addition, all Clear Channel stations receiving the traffic reporting service and that have websites must provide a legible website link to Utah's CommuterLink™ website at no charge to UDOT. According to the agreement, UDOT does not guarantee continuity of access, nor the accuracy, timeliness, or quality of information. Information is provided on an "as is" and "with all faults" basis. Any reliance on said information is at the risk of the company. UDOT also accepts no liability for system down time.

In an agreement where the New Jersey Turnpike Authority administers the contract for both the Turnpike and the New Jersey DOT, the Authority receives $30,000 in revenue per year from TrafficLand based on the current number of cameras configured. On a per-unit basis, additional revenues may be provided should the New Jersey Turnpike Authority and the New Jersey DOT increase their number of configured cameras. The Agreement for Redistribution of Streaming Traffic Video indicates compensation in the amount of $2,500.00 per month for all cameras available to TrafficLand with additional compensation of $50 per month for each group of 50 cameras added to the original quantity.

The Michigan DOT has a video sharing contract with Skyline Technology Solutions (Skyline) where the company processes and hosts all of the Michigan DOT's traffic video and provides traffic video feeds for the Michigan DOT website at no charge. The contract does not restrict video redistribution by Skyline at a cost to external customers. Skyline provides a hosted site where MDOT's trusted partners (dispatch, county, police, and other TMCs) can access the video for free. This arrangement helps Michigan DOT avoid costs associated with purchasing and maintaining servers to host and distribute video to partners and the media.

Agency Data Valuation

Historically, agencies have found that the cost of managing even the most basic cost-recovery business models were overwhelming and not worth the effort. Most agencies are not equipped to receive funds—even in small amounts. Both legal and ethical issues have proved too great for the relatively modest returns.

However, if an agency does want to determine the value of a particular data asset, the agency should assess the following:

Identify the opportunity: What is unique about the data or asset that makes it valuable? Who desires the data? Have companies requested the agency's data already, or will the agency seek out potential buyers? How many potential customers are there?
Assess your uniqueness: Does the agency have exclusive rights to its data? Can anyone else offer an equivalent product?
Assess the market: Data for dense, urban areas with multiple, competing media platforms are likely more desirable and would provide greater competitive advantage. Do not expect data assets in rural parts of a State to be considered as valuable as those in congested urban regions.
Tiers of service: Will the agency offer an expected level of service and/or reliability for its data, or will it simply provide the data as-is with no warranty or expectations of availability? Will the agency offer different product types or combinations, e.g., silver, gold, and platinum data products?
Evaluate your costs: Will providing data to the private sector cost the agency anything? Potential costs could include:
- IT/networking costs.
- 24 hours a day/7 days a week operations and maintenance support.
- 24 hours a day/7 days a week technical/user support.
- Contracting/management/legal.
- Hardening/productizing of your data product.
- Periodic system upgrades.
- Insurance/liability.
Profit or cost-recovery: Does the agency simply want to recover the cost of making the data available, or is it looking to bring in funds to support other efforts?
Test the market: Before officially releasing a cost model, it is important to test the model with one or two private sector partners to gauge their reaction to the pricing and terms. Face-to-face conversations are preferred to help in determining the private sector's appetite for a proposal. Be ready with alternative business models as well as both floor and ceiling numbers.

It is important to think about management costs, contracting costs, and legal fees. Even if an agency does not officially charge its time to specific projects, there is a real cost to taking on additional efforts and contracting. That cost could be purely financial, or it could be realized when other work or contracts get dropped or bumped due to a lack of available resources.

It is also critical to evaluate the liability cost associated with charging for data: Does the agency stand to lose anything by providing the data (either free or at some cost)? Are there philosophical concerns with limiting access to the data to those who can afford to buy it? Would the agency be stifling innovation in restricting access?

Alternatively, does the agency have anything to gain by sharing/selling the data? Positive public perception and allies in the private sector can sometimes be more valuable than the cash or cost of the transaction.

previous | next