Open Contracting Data Standard Report on stakeholder engagement for Activity #2

Introduction

The Government of Canada has committed to leading the efforts in accelerating and expanding the publication of contracting data in order to strengthen the openness and accountability of government. It is striving to set a higher bar for transparency, recognizing that government and its information should be open by default.

Open contracting is about publishing and using open, accessible, and timely information on government contracting to engage citizens and businesses in identifying and fixing problems.

Open data is available to provide citizens and civil society the means of analysis and oversight of GC contracting spend. For government contracts, aggregate contracting data is often available, particularly related to what is being purchased. However, opportunities exist to provide more details regarding how the expenditure is moving through the procurement cycle.

As part of the project to fulfill this commitment, a data pilot project was set up according to the scope of work in Annex A. This report has been prepared in satisfaction of the last requirement of Activity #2 in Annex A.

Theauthors of this report have attempted to faithfully present the perspectives and impressions of the various participants involved in the consultation.

Open Contracting Data Standard

In order to place this report in context it is important to understand the nature of the Open Contracting Data Standard (OCDS) referred to in the scope of work.

The Open Data Contracting Standard is a free, non-proprietary data standard for public contracting developed and maintained by the Open Contracting Partnership (https://standard.open-contracting.org/latest/en/).

According to documentation on the Open Contracting Partnership site, the standard was developed to facilitate the disclosure and use of public sector contracting data to enable and facilitate:

• Value for money analysis
• Fairness and ease of doing business
• Deterrence of corruption and monitoring the integrity of the procurement system
• Boosting the efficiency and effectiveness of public contracts

The Open Contracting Data Standard is used to support an open contracting approach by specifying what data must be made available, and by demanding that it be made freely available for use and modification by anyone for any purpose. The standard is also intended to address the fact that data must not only be publicly available, but also accessible and useful to organisations outside of government. The standard therefore requires that the data be accessible and machine readable. Further, it defines a structure for the released data and certain definitions for the meaning of the data to ensure that it can interpreted easily and unambiguously.

Thus, the core purpose of the OCDS is to ensure that open contracting processes deliver complete and standardized datasets that are assembled and published in standardized ways so that organisations that analyse this type of data can build applications and tools that can use data from multiple sources with a minimum of source-specific pre-processing.

For these reasons, the portions of the standard that lay out how the data should be formatted and published are seen to be as critical as the portions of the standard that deal with what data is to be released.

The technical details of the OCDS with respect to formatting and publication of open contacting data are beyond the scope of this report. Those details are also beyond the scope of this pilot, which only attempts to identify what data is to be included in future Open Contracting Data releases. It does not consider how it will be assembled, formatted, or published. It is important to note that, for the civil society community that is concerned with using open contracting data, these details are vital to determining if the process implemented by the federal government will meet the Open Contracting Data Standard. No implementation of the OCDS will be complete until these details have been determined, reviewed with the stakeholders, and put into practice.

With this in mind, it should be noted that the data provided through the data pilot for review with stakeholders was limited to a set of data tables, which contained the pilot data. While this data is described in more detail in the following section, the dataset does not represent a fully OCDS-compliant release on its own, in that it does not follow the OCDS standard with respect to formatting and publication. Therefore, this data pilot is not a pilot of the full implementation of the Open Contracting Data Standard but rather as a test to determine what data from federal government data sources should be assembled and identified for inclusion into a product that would eventually conform to the OCDS.

As a result, the stakeholder engagement process that is being reported on in this report has been limited to discussions to identify any concerns raised by external stakeholders as to the nature of the data that had been identified for release.

The Data

The data that was delivered for review with the stakeholder community consisted of a single Excel file with separate tabs for each of the procurement phases:

1. Planning
2. Tender
3. Award
4. Contract
5. Implementation

The data was not merged in any way, but all the data initially referred to the same 300 procurement projects. Due to PSPC departmental stakeholder feedback, the procurement projects were reduced from 300 to 250. Because some phases contain multiple entries for each project there were more than 250 records in some tables. The following table lists the number of records in each of the Excel data tables contained in the file.

A list of data fields was provided in addition to the data itself. This table was annotated to indicate which data fields were included in current Government of Canada public data releases and which fields would be expected to be added in order to comply with OCDS.

The following analysis was performed to assist in providing a summary and context for stakeholder consultations.

Firstly, the data fields – old and new – were categorised according to what information they contained in order to provide external reviewers with some context regarding the type of information included in the various fields. The categories were defined as:

• Information describing the procurement itself
• Information describing the purchaser inside government
• Information describing the supplier
• Information describing financial transactions

Based on these categories, the new data fields can be summarized as follows:

Nearly 80% of the new data will therefore be fields that either aim to describe the procurement itself or which provide more information on the government purchasing department or program. It is assumed that the inclusion of new data in these two categories is unlikely to be of concern to the supplier community.

However, some new fields exist containing information on the supplier, which would therefore be expected to be of interest to the external stakeholder community. Most of the new fields are related to details in the implementation section. The information that is likely to be of most potential concern to suppliers are the new financial information fields. These are summarised below. The grey boxes are fields that are general descriptions of the transaction (i.e., dates, currency, or government purchasing document numbers). While these are unlikely to be of concern to suppliers, the yellow boxes contain information regarding the value of the transaction or items from supplier-provided documentation such as invoice numbers and item details, and as such, these fields are more likely to be of potential interest to suppliers.

It was then decided to perform some simple high-level analysis on the provided data in order to generate some examples of how the information might be used by external parties. This was done in order to be able to allow external stakeholders to have some representation of how the data might be used and what such analysis might reveal that might be of concern to stakeholders.

In order to do this analysis, the data in the five Excel tables was merged into one dataset. Some simple analysis was then performed on this data to determine, for instance, whether it was possible to proceed from a contract tender to determining how many contracts were let, to whom, and how much was eventually paid to each contractor and for what goods and services. It was found that this analysis could be performed.

Based on this analysis, it was determined that the most likely source of concern for suppliers would be the level of detail contained in the fields related to invoicing. It was felt that suppliers might have a concern if the information contained in these fields, which when presented with financial details, could reveal pricing strategies and other information that might be viewed as commercially sensitive and proprietary.

Several trial analyses were run to determine what level of detail would typically show up in such a report. While the testing was not comprehensive, and while it is based only on the samples provided in the dataset, it was found that the overall level of detail contained in the data was very high level.

The following sample report was prepared to share with external stakeholders. It shows the level of detail that would typically be available in the database. Note that the company names have been changed to generic values, but all other data is as it was found in the sample data.

Consultations

The information described in the above section (i.e., “The Data”) provided the basis for a memo, which was used as a precursor to engagement for consultations with external stakeholders.

Several groups were contacted over several months to determine their interest in further consultations. These groups included:

• The Minister’s Supplier Advisory Committee (SAC)
• Procurement Assistance Canada Regional Offices – for distribution to their respective stakeholder networks
• Representatives of several aerospace and defense contractor firms

Not-For-Profit organisations working in the open data space. In particular, Open North provided some background information and context in July 2021 and were invited to participate in a full review in October and November of 2021. Open North was also invited to include their partners and colleagues in the discussion.

Note that the 2021 federal election delayed consultations, as the Caretaker Convention limits interactions with external parties. ^{Footnote 1}

In all cases stakeholders were offered a high-level verbal briefing and, if desired, a memo describing the pilot dataset. Stakeholders were encouraged to reach out for more detailed consultations and were offered the opportunity to participate in a workshop to discuss the data.

Most stakeholders declined the opportunity to continue engagement beyond the initial verbal briefing. Of the approximately 30 organisations and individuals approached, 6 agreed to receive further data and 4 agreed to follow-up discussions beyond the initial briefing. The SAC members received the package of information describing the data. Some SAC members reported raising the topic with their member committees but reported that the information provided to SAC was sufficient to satisfy their members’ interest at this time and declined further discussions.

A few additional industry stakeholders requested the written material. Some also reached out for a detailed briefing based on the memo. No stakeholders – industry or civil society – expressed interest in participating in a full workshop.

Findings

The following findings are made based on the consultations with the various stakeholders that responded. They are summarized for length and clarity but have been reproduced as faithfully as possible to provide perspectives from both camps.

In general, the information provided by the data pilot – being the list of data fields that would be included in a fully OCDS compliant release - was not seen as being controversial by contractors or their representative organisations. Similarly, civil society groups working in the open data space considered the level of data provided by the data pilot to be insufficient to warrant significant discussions absent more details on how the full OCDS standard will be met.

Supplier Specific Findings

Of those contractor community stakeholders that did engage, it was clear that the one overriding concern was the presence of detailed invoicing information in the data. Contractor concerns were focussed on the fact that invoicing details might include commercially sensitive, proprietary, or personal private information. Examples of such data would be:

• Names and pay rates of consultants of key employees
• Detailed pricing information that could be used to reconstruct bidding and pricing strategies
• Details of bills of materials or other information that might reveal details of designs or other details not normally shared publicly or with potential competitors

However, based on the sample data provided, stakeholders were largely unanimous in that the level of detail represented in the pilot data did not cause much concern. In the sample data, the descriptions of the invoicing items are limited to very high-level descriptions, which seem to have been provided by the contractor themselves, or simply of line numbers from the invoice. The consensus seems to be that, at this level of detail, the data included in the dataset would not be commercially sensitive, proprietary or represent private details of individual employees.

However, it was also noted in discussion that the pilot dataset is only a representative sample, often from several years in the past and that no details were provided regarding how the information in the dataset was collected, or how such information would be collected in the future – particularly with the advent of PSPC’s Electronic Procurement Solution (EPS). Stakeholders, therefore, did express a concern that the level of detail contained in these data fields could change over time and that no guarantees had been (or could be) provided that such data might not become problematic in the future.

This concern was particularly strong with regards to changes that might be implemented through EPS. Stakeholders could envision scenarios where the content of such data fields might be specified by electronic forms and might include the requirement to provide information or descriptors that might be problematic – such as the identification of individual employees and their rates of pay.

Stakeholders therefore encouraged further engagement on this question in particular as implementation of OCDS and EPS proceeds.

Stakeholders also noted that even the level of disclosure represented in the pilot data was new to the contractor community. Since it appeared that at least some of the data contained in the invoicing data fields had been drawn directly from contractor documentation, they noted that it would obviously be unfair to release any data retro-actively as it would represent the release of contractor-provided data that had not previously been designated as being subject to public release.

Stakeholders were also adamant that once invoicing data is scheduled for inclusion in public data releases, contractors must be made aware of what specific data from their invoicing documents will be made public so that they can take this into account in the preparation of their invoices.

Civil Society Specific Findings

With respect to civil society groups working in this space, consultations made it clear that those groups welcome progress being made toward a fully OCDS compliant release. However, they did not feel that a list of data fields to be included in such a release constituted enough information to warrant a lengthy review or a detailed engagement at this time.

Stakeholders in this category explained that the details of how the data will be assembled, formatted, and published are as vital to their interests as the details of what data will be scheduled for inclusion in the data release. They pointed out that this concern was the motivating force behind ensuring that items related to the assembly, formatting and publication of data are included in the Open Contracting Data Standard.

They indicated that when the full details of how Canada expects to implement the OCDS are available, they will welcome the opportunity to review them and provide feedback. They further indicated that it was premature for them to engage in any more depth at this time given the amount of information that could be provided to them.

Conclusion and Recommendations

Throughout consultations and during the preparation of the report, it became clear that there is a great deal of interest on the part of industry and stakeholders alike on the OCDS project. Moreover, one should not assume that industry would naturally fight greater information disclosure, nor that civil society would always welcome more information. In the case of the former, most associations recognized the value of having greater or more complete contract data in the public sphere, as their own members could benefit from better analyses of government spending trends. For its part, civil society was more concerned about the broader OCDS effort and the many factors that play into how, when, and in what format information is provided.

Overall, the following recommendations can be gleaned from the stakeholder engagement process.

The supplier community’s interests can be summarized as follows.

• There were no significant concerns with the publication of the dataset as presented. However, suppliers noted the importance of avoiding any surprises in the data that is eventually released. This is particularly true if any of the data is drawn directly from documentation that they submit to government through normal contracting process. Suppliers will want to know in advance what data will be made publicly available.
• Suppliers will also want to ensure that any changes to what is released are not made retroactively and that the changes are only applied to data that is yet to be collected.
• Suppliers will have a concern to understand how the implementation of EPS and OCDS will interact with one another. In particular, when data that they provide is identified as being part of future public releases, they will want to understand if EPS will generate, or imply, requirements that affect that data.

Civil Society’s views can be summarized as follows:

• There were no concerns with the publication of the dataset as presented.
• The Open Contracting Data Standard goes well beyond the identification of what data will be released. For these groups, engagement on OCDS needs to include a discussion on how the entire standard will be met, as in their minds, making data available is not the same as making it accessible. They are ultimately interested in being consulted on how to make the data available and accessible.

Annex A – Scope of Work

Activity #1 - Pilot data that tests the implementation of the Open Contracting Data Standard (OCDS) is undertaken, which includes a cross-section of at least 300 contract records for a variety of contracts, including major projects. Pilot data will include all stages of the procurement cycle (planning, tender, award, contract, and implementation).

Description: This milestone will implement the latest OCDS (v1.1.3 or later) as a pilot with at least 300 contract records for a variety of contracts, including closed contract records and major projects. Pilot data will include all stages of the procurement cycle, planning (requisition data), tender, award, contract and implementation (spend data) and will be published on the Open Government Portal (open.Canada.ca).

In Scope:

• Contract data based on the contract records issued by PSPC (PSPC’s role as an organization responsible for acquisitions of goods and services on behalf of itself or other government departments and agencies).
• Contract data required by the Open Contracting Data Standard version 1.1.3 or later.
• Pilot of contract records will include all stages of the procurement cycle - planning, tender, award, contract, and implementation.
• Contract records (~300) for a “variety of contracts” includes closed contract records.
• Pilot data of contract records (~300) to be published on the Open Government Portal (open.Canada.ca).

Out of Scope:

• Contract records will not include active contract records.
• Contract data based on the contracts not issued by PSPC.

The results of this data pilot were the to be reviewed with stakeholders according to the following scope of work.

Activity #2 - 3-5 public workshops on open contracting are held to analyze what types of contracting data are currently available and assess barriers to releasing open contracting data. The workshops consider the results of the open data pilot. Workshops include participation from civil society, procurement experts, and industry representatives. A report is developed outlining input received and recommendations developed during the workshops.

Description: This milestone will comprise of evaluating the feasibility and degree to which open contracting data can be adopted by PSPC. PSPC will engage public (i.e., civil society, procurement experts, and industry representatives) and internal stakeholders via workshops. In doing so, it will leverage the lessons learned from the Open Contracting Data Standard (OCDS) version 1.1.3 or later pilot. A report will be developed outlining input received and recommendations developed during workshops.

In Scope:

• Work with public (civil society, procurement experts and industry representatives) and internal stakeholders to evaluate and document the scope, logistics, and potential barriers pertaining to the release of open contracting data.
• Consult with the Open Contracting Partnership to further explore the potential for system operability using the OCDS (v1.1.3 or later).
• Develop a report outlining input received and recommendations developed during workshops.

Out of Scope:

• Draft of policies, systems, or applications related to the adoption or implementation by the Government of Canada for the release of open contracting data.