01-RFP-SDT-2014
Questions and Responses: IDEA Public Schools Business Intelligence Solution RFP
Can you please elaborate on the write back capabilities? Do you want users to be able to make a change to a dashboard / application and have it update the source system of record?<Response>We have sandboxed a variety of BI Tools. Some provide the functionality of updating source system records while others do not. In certain circumstances, it will be preferable to allow users to update the source easily without having to traverse a variety of systems to accomplish the task. Use case: We have several custom internally built solutions which have varying levels of security defined. Those with administrative level privileges would be in a position to make changes to standardized data such as grade level targets/minimum requirements, or to correct missing detail information.
Is an OLAP technology / architecture a must have requirement? Our product is not an OLAP tool. Instead of browsing a cube for a pre-aggregated measure we perform a dynamic calculation, on the fly.Response> No, it is not a must-have requirement. However, we plan on leveraging a variety of data sources including transactional systems as well as existing cubes and/or information residing in our data warehouse.
What extent of statistical, regression and clustering analysis is needed? <Response> Our future plans include the inclusion of advanced analytics for the purpose of prescriptive and predictive analytics aimed at offering instructional recommendations to teachers based on student performance. We are currently working with a Big Data consultancy on the production of these analytics and have analyzed the potential need for custom development as many of the OTS solutions either do not have these included or are in the infancy stages of development. It is desirable that the chosen tool have these advanced capabilities. However, it is not a must-have feature.
What type of versioning and archiving of reports do you want to do as well as checking out of reports? <Response> Versioning and archiving will be primarily used for referencing older content in order to reduce time-spent redeveloping. The interest is being able to retrieve and reference major versions, not minor versions. In addition, there are times when older reports which have been taken from the end-user’s view to be requested for inspection. We have multiple developers who will be responsible for enterprise dashboard development. “Checking out” ensures that changes made by one developer are not in conflict with those made by another. In essence, it is a form of locking. Therefore, when a report is “checked out” developers are unable to make changes to the report until it is checked back in. Versioning and archiving in some form would be a must-have. Check-in/out is debatable and would be based on the tool’s development environment.
Please elaborate on the wireless technologies and methodology protocols supported? What is the use case? How do they apply to a BI tool? <Response> The intended solution will run within our intranet technology which is SharePoint 2013. This solution is currently 100% responsive and works on a variety of devices including PCs, laptops, tablets, and smartphones. IDEA issues iPhones to a variety of staff (administrative and instructional). The intent is for staff members to easily access and be able to read information including dashboards and reports from their devices, including issued iPhones and iPads (and in rare cases personal Android devices). Use case: The Chief Operating Officer travels often between sites using her iPhone/iPad for the majority of her work. She accesses SharePoint/Office 365 to work on documents, Outlook for email, BlueJeans for video conferencing, etc. These tools allow her to perform all tasks necessary to carry out productive meetings from her phone while in transit. If she chooses to leverage analytics in these meetings via either of these devices, it would be necessary for her to access and easily traverse/read a dashboard/report leveraging this technology.
Can you define limit on single library or provide a definition? <Response> This question references the tested performance and capacity limits of your product and how limits relate to acceptable performance (i.e., software boundaries and limits of your product). These could include the following: A) Boundaries: static limits that cannot be exceeded by design, B) Thresholds: configurable limits that can be exceeded to accommodate specific requirements, and C) Supported limits: configurable limits that have been set by default to a tested value.
Are you looking for an open source system?<Response> No.
Please elaborate on - Reports added to a menu for later use without programming? Are you looking to have reports bookmarked in the application for a user to easily navigate to when opening the application?<Response> Yes, we are interesting in either end-user book-marking capabilities or the ability for the end user to personalize their navigation experience without support from IT/development team. This is a nice-to-have feature.
Please elaborate or provide an example for ‘User can select a report group order or sort order via a menu?<Response> Example 1: Data grid is reporting detail data of First Name, Last Name, and Grade Level. The data set contains 1500 records and is sorted alpha by first name. Can I change the sort order so that the list is sorted alpha by last name? Example 2: Data grid is reporting detail data of First Name, Last Name, and Grade Level. The data set contains 1500 records and is grouped alpha by grade. Can I sort the group in descending order and sort within those groups by last name and first name in alpha order? In other words, can the user dynamically set the sort and/or grouping order in reports, data grids, or other components, overriding default sorting and grouping order?
For ‘Expiration dates for reports’ can you please elaborate on what you are looking for in regards to interacting with groups of reports<Response> Expiration dates for reports is a management tool. For example, I may have a static data set that I want to publish for end-user consumption, but I know that the data will only be valid for the next 48 hours. Rather than having to manually take the report offline in 2 days, it is preferable to automatically take the report and all of its links offline by a given date timevia a predefined setting (i.e., remove it from end-user view/access). If I have the option of setting this at the time of publishing, it makes life easier.
What is the use case for extended, programmatic, and triggers functions? Can you please elaborate?Response> I am going to refrain from going into lengthy detail as to the use of such functions as there a myriad of purposes. However, a few simple examples include the ability to pull table schema from databases, leverage metadata, or even initiate a programmatic change to a chart upon user parameter entry. If your tool does not inherently have these capabilities, then please feel free to choose “Not Supported”.
Please clarify what you mean by cycle management or master data management? <Response> Master data management (MDM) is a method of enabling the enterprise to link all of its critical data to one file, providing a common point of reference. This streamlines data sharing across multiple system architectures, platforms and applications. A tool of this nature can be used to manage the removal of duplicates, standardization of data, and incorporation of business rules/constraints for example. MDM's objective is to provide processes for collecting, aggregating, matching, consolidating, quality-assuring, persisting and distributing organizational data and ensure consistency and control over the data for the long-term. IDEA is performing master data management at the data warehouse level, but not at the source system level. Since we will be accessing a variety of data sources, some of which may not be integrated into the data warehouse, we are interested in the ability to have the same facility through a BI Tool. This is a desired, but not a must-have feature.
What ETL does IDEA currently use and want to integrate with the product? <Response>We are currently developing ETLs using Microsoft’s SQL Server Integration Services.
What do you mean by the software creating branches and versions? <Response> Assuming you are referring to rows 1476 and 1477 under “Data Services & APIs”, the reference to branches and versions relates to managing the service/API repository lifecycle in a multiuser development environment. This facility enables individual developers to design and test manageable subsets without impacting each other, yet share their changes with other developers in a controlled, incremental fashion. It also allow developers to migrate changes to test and production systems in bulk, or incrementally. Therefore, development on data services and APIs can occur in parallel.
Please elaborate on controls on data integrity input.<Response>This refers to source data updates. If the proposed BI tool has the ability to update source data as described in question #1 above, the questions about data integrity apply in those cases. For example, if the user is allowed to update grade level targets, we would want to restrict the values entered to integers rather than letter grades.
What do you mean by virtual tables? <Response>Assuming this question is referencing row 1639 under Data Governance, the item asks, “Does the proposed product provide attributes to display to be manipulated as virtual tables?”. This question is basically asking if system metadata can be viewed in table format (i.e., the equivalent of creating a view on SQL Server that queries the tempdb system table.)
What is your proposed MDM system? <Response>We are leveraging Microsoft SQL Server’s Master Data Services (MDS) for data populating our data warehouse.
What kind of library are you referring to of analytic components?<Response>Assuming that you are referencing the items under predictive analysis (i.e, row 1872), the item is basically inquiring about the collection of visualizationsavailable for predictive analysis (if any) within the tool. If none are available, is there opportunity to integrate with other tools like R which have the capacity? This item is further confirmed by the items related to the integration of R which follow.
Please define what is meant by reusable components?<Response> Assuming that you are referencing the items under predictive analysis (i.e, row 1874), the “reusable component” being referenced is the creation of a data model. In other words, can a single data model be created and be reused in different analyses easily?
What do you mean by fully human readable SQL? <Response>This question is asking about the formatting of SQL generated by the system. If the developer or power-user has the ability to view the SQL statements generated for a report or behind a chart/graph/visualization, it is easier to identify and troubleshoot data problems. Many proprietary systems do not format the SQL in any way making statements difficult to parse. Below are two examples, one that is not parsed and one that is.
SELECT * FROM table JOIN other_table ON table.id = other_table.id WHERE column = 'value' AND column = 'value'
Versus
SELECT *
FROM table
JOIN other_table
ON table.id = other_table.id
WHERE column = 'value'
AND column = 'value'
If you add multiple tables, with multiple complex joins, a variety of filters, and throw in subqueries, the former example becomes extremely difficult to read.
Please elaborate or define what you mean by User Business View Maintenance option?<Response> “User-friendly Business Views” refers to the ability to organize and present information based on the audience. From the Production Reporting standpoint, having the ability to segregate maintenance options and/or selection criteria (for example) by business need/audience would be helpful.
What do you mean by an automatic breadcrumb bar?<Response>Breadcrumbs are a graphical control element. Breadcrumbs are used as a navigation aid in user interfaces allowing users to keep track of their locations within programs or documents.
What do you mean by a right trimmed interface?<Response> “Rights trimmed interface” refers to security. Based on the user’s security, only content appropriate for that user’s rights/role would be displayed. (The opposite would be that all content is displayed and access is denied upon unauthorized access to the content.)
What is view improvements? Please elaborate.<Response>Is there a way to control the aesthetics of the presentation either through configuration changes/customization or is it “what you see is what you get”?
Please elaborate on large-list and cross-list indexing?<Response> Reports based on large or mult-list data sets which are not properly indexed can create performance problems, bottlenecks, and other issues. The purpose of the question about large and cross-list indexing is to determine if the tool has the ability to create indexes to avoid these situations and optimize these data sets. If the technology you employ leverages a different methodology for addressing problems related to very large and/or complex data sets, please feel free to include a description of the methodology in the notes section.
Please elaborate or define Subsite promotion?<Response> Subsite promotion is the act of converting a subsite to a site collection / main site. The question of “Does the proposed software provide subsite promotion?” is referring to potential pitfalls to conversion as described where your tool is concerned. In other words, if I create a dashboard in SharePoint on a subsite and later determine that I want to promote my subsite to a site collection, will there be a need to recreate or drastically modify the existing dashboard to accommodate the promotion based on url changes/security/etc.?
Please elaborate/ define List events and site events?<Response> List events are core events, including changes, additions, and removals of items and columns (schema changes). Simple site events include deletion of sites and site collections. These two items are listed under platform capabilities (rows 2575 and 2576).
Is IDEA looking for a SaaS model, cloud, hosted, on prem?<Response> Our current preference is on-premise. However, we are open to other solutions if the offer fits the need.
What is your definition of real time? Daily, hourly, every 15 minutes etc?<Response> Real-time is dependent on the data set in question. From the data warehouse perspective, we will have daily updates while others will be hourly. We even have instances where data sets are updated every 5 minutes. So, the answer varies. We would like to limit the latency between source systems/data warehouse and availability through the chosen BI tool as much as possible.If querying a transactional system directly, real-time would equal seconds or minutes rather than hours or days.
Who provides IDEA Public Schools’ current data warehouse?Three years ago, IDEA Public Schools contracted with Hitachi to develop and host the district’s intranet portal in SharePoint and the district’s data warehouse in an MS SQL Server environment. In the fall of 2013, IDEA brought both in-house, implemented a new physical infrastructure, and in the spring of 2014 began to build upon the existing solutions.
Would IDEA Public Schools consider replacing its current data warehouse with an education data warehouse solution that is already in use in Texas, and that is already coupled with a Business Intelligence Tool that together address substantially similar goals as described in the RFP?We would have to see if the solution meets our data needs. I am open to a preview of the solution, though. As a side note, our methodology is based on individualized learning and the strategies for lab-based learning are decided at the campus level (i.e., a vendor’s software is not standardized across the district). This has the biggest impact on our decision-making as many data warehouse off-the-shelf solutions do not take into account the high level of dynamism necessary. In addition, we have a growing number of custom developed applications used both operationally and instructionally. The proposed DW solution would need to have the flexibility to incorporate a variety of data sources as well as the ability to create custom data models.
Can we provide supplemental information beyond what is requested in the RFP Excel workbook? For example, can we provide, sample screen shots.Yes, please do.
The RFP contains the following requirement: “The Customer requires the Contractor to demonstrate quality of past performance. This must be done by providing at least 5 references, of which at least 2 are similar to the Customer. If you are supplying references at this time enter a response of YES and click the "References" Tab to enter references.” The RFP Excel workbook we received, along with the one on the website, does not contain the Reference Tab. Can you provide an updated workbook?<Response> An updated RFP template has been provided on the website.
There is reference to several plans that are to be provided (see lines 686 & 687). When are these plans to be provided – at contract award or with submission of the proposal? Similarly, should sample service level agreements be provided?
For SaaS solutions, the plans are being requested as a part of the proposal submission. We will be evaluating these plans as a part of the decision –making process. So, please include the necessary disaster recovery, business continuity, and sample service level agreements as a part of your submission.