Crystal McDonald:    Good afternoon, everyone. My name is Crystal McDonald. I'm a policy advisor on the staff of the Weatherization and Intergovernmental Programs Office, the Policy and Technical Assistance Team. Welcome to our webinar today. We're happy to have you, and we will be presenting [break in audio] "Best Practices in Energy Data Collection and Tracking" session today. The objective of this work is to discuss a systematic methodology to energy data collection and tracking that will help analysts, energy planners, and community officials conduct benchmarking towards understanding the performance of their assets, and, thereby, improve them.

Technical assistance to state and local governments. The guide presents a range of clean energy opportunities and renewable power, sustainable transportation, and energy efficiency, along with key DOE resources and information. You see a snapshot here of our state and local solution center research guide, which points to the state and local solution center available online. I want to make you aware of the State and Local Spotlight, as well, which is a monthly publication that features clean energy news, events, and resources for state and local leaders. Sign up to hear about news resources and initiatives as they are launched and stay up to date. DOE values state and local governments as a critical part of developing the nation's clean energy future.

Everything you hear today will eventually be posted on the state and local resource, the solution center online, and please use that as your go-to resource for clean energy. With that said, I'll hand over the session to my colleague at the Lawrence Berkeley National Lab, Shankar Earni. Thank you.

Shankar Earni:           Yes. Thanks, Crystal. Again, my name is Shankar Earni. I'm with the Lawrence Berkeley National Lab. Thanks for joining us today for this talk on the data collection and tracking. This is the course outline for today. We're going to start with introduction, highlight some of the concepts of benchmarking, discuss the whole process associated with benchmarking, and tell you how data collection and tracking fits into that overall benchmarking process, discuss these five aspects related to benchmarking, the first aspect being how one can create a comprehensive data asset inventory. The second aspect is streamlining how one can access utility data. The third aspect is how can you decide on what data tools and analytics to use to help process some of the data that you are collecting.

The fourth aspect is we're going to touch briefly on the various organizational structures that one can imagine to help not only collect this data, but also to make sure that this data that you collected is kept up to date. The fifth aspect is engagement and communication. As most of you know, engagement and communication is very important to – that can help implement _____ program. Finally, we're going to close with some closing thoughts, and, as we go through this presentation, we're going to show you some case studies showing how various public sector entities are implementing some of these aspects, and also show you some of the benefits that they've realized in this process.

The course objectives, this will help personnel from various cities, communities, and states that are involved with benchmarking process, and specifically with the asset data collection as a way to improve the performance of their assets. The learning objectives, this will help understand the methodology towards putting together a robust data collection and tracking system. It will also help you to develop and maintain an inventory of energy-consuming assets, and also discuss some of the options that are available to access related utility data. We're going to touch briefly and help you understand some of the options and strategies for establishing a robust and sustainable data management process. We're going to show you some of the hurdles that one might face as you go through this process of data collection and tracking.

So to kind of give you some context, this is _____ introduce the concept of benchmarking. Benchmarking is a concept of comparing the performance of an asset with some established norms. These norms can be either past performance or some other established benchmark, so the benchmarking can be done either longitudinally or cross-sectionally. Benchmarking, when you compare the performance _____, assets _____ performance with its historical performance, that's what we call it as longitudinal benchmarking, and the cross-sectional benchmarking is when you're comparing the asset's performance with other similar assets within that category. So it can be cross-sectional or longitudinal.

So some of the benefits associated with benchmarking are it will help you to facilitate energy accounting. It can help you assist in identifying energy conservation opportunities, and, thereby – and it can also be helpful in using as a tool for measurement and verification to assess the performance of certain projects. Some additional benefits. This will help you if you can assess how well the asset is performing, it will help you to manage the energy consumption more proactively. It will help you to assess and compare a building's performance. It also can be a helpful tool in identifying some of the billing errors and other anomalies that you might see in day-to-day practice. As I told you, it can be a very useful tool for conducting measurement and verification to assess the effectiveness of your current operations and policies and practices, and, thereby, assist you in help setting reasonable goals and targets. Also, it helps you to communicate some of these results in a meaningful manner.

So this is the overall benchmarking process. Some of you might have seen this. This is obtained from the solution center's website. It is basically a seven-step process. The first step starts with developing a benchmarking plan. This is where you set objectives of benchmarking, identifying the key actors, deciding what the key milestones are, so you're essentially putting together a plan as to how this benchmarking process is going to unfold. The second step in this process is where you're going to identify the benchmarking tools that are currently available that fit your portfolio to assess the performance of those buildings or assets. This will be an important step to decide what kind of inputs are needed and what kind of outputs that they're going to generate, these tools are going to generate to decide on your next steps.

The third step in this process is outreach. Outreach is very important. It's important to put together an outreach plan and reach out to all the parties involved to ensure that everybody knows what benefits that they can expect from this benchmarking process. The fourth step in this process is data collection. That's where we're going to focus the rest of our talk today. This is where you are essentially collecting all the data that's needed to assess the performance of these buildings based on the benchmarking tools that you've chosen. The fifth step in this process is insuring that the data that you've collected doesn't have any errors or missing values, so you want to have some kind of a QA/QC procedures in place to ensure that whatever data gets analyzed subsequently is sound. So we have dedicated a module for this, and we did a presentation last April. We have the link for that if you want to access that presentation later.

The next step in this process is analyzing and interpreting the results. Whatever data you've collected and cleansed, that has to be analyzed to understand the performance _____ _____ these buildings so that you can separate the performing buildings from the non-performing buildings so that you can focus on the non-performing assets. This is something that we're going to cover in more detail during our next week's webinar, so if you haven't registered, you might want to register for that if you want to know more about that. The seventh step in this benchmarking process is about communicating the results. This is where you're essentially closing the loop and bringing back all the parties that are involved and showing them what the final results are, and also some of the lessons learned so that everybody knows what the final results are. It's also important to present these results in a manner that can speak their language, so you might have to customize these reports in such a way that they're understood by different parties.

So why track energy data? _____ has commissioned a study to understand why tracking data is important, and these are the top five reasons that the various energy personnel have given up. The first reason is improve strategic energy management capabilities. This will help you to build the case to leadership on the value of energy management and gain additional support. It will improve control and transparency of energy costs and budgets. It will also help to improve operational efficiencies, and _____ it will also help facilitate demand, _____ and facilitate demand response for programs and also efficient energy purchasing. So, as I said, the rest of the talk is going to focus on the data collection and tracking piece of the benchmarking process. As we see it, there are five different aspects associated with this data collection and tracking.

The first aspect is developing a comprehensive asset inventory. This includes the assets that you have at your disposal and their characteristics, and also the commodity-related information that these assets are consuming. The second aspect is the data, how can you access the utility consumption data in a streamlined fashion so that the data can be gathered in a timely and efficient manner. The third aspect is deciding on what tools and analytics to use so that you can process some of this data. The fourth aspect is organizational structure, what kind of organizational structure that needs to be in place so that this data can be collected efficiently, and, also, going forward, how this database can be updated in an efficient and effective manner. The fifth step is engagement and communication. As I said, engagement and communication is very important. It's important to have targeted, clear, and transparent messages to all the parties involved so that they know what the effects of this data collection and tracking steps are.

Now we're going to focus a little bit, dig deeper on this asset inventory aspect, why it is important. It is, as I said, the asset inventory is a centralized database that has assets and their characteristics information. It also has information related to the utility consumption and costs. Why it is important? It will help you to establish a good benchmarking program, thereby, manage your energy consumption properly, optimize your operations, help you prioritize energy efficiency projects and investments. Who should be involved? Anybody that's involved with energy or sustainability should be part of this effort. It could take about six to two years to put together a comprehensive data on assets. Obviously, it's going to depend on portfolio size and number of staff that you have dedicated to this effort, and, also, the degree of cooperation that you have among different entities and personnel.

The asset inventory, again, has the aspect of asset inventory as these four different elements. The first element is locating the proper sources of asset data. As you know, this asset data is disseminated across the entire organization, so the first step would be to understanding what those sources are in order to develop a comprehensive asset inventory. The second step, or second element of this aspect is creating a standard organizational structure to house this data or to store this data. The third step is as you sift through this data coming from different sources, you might find that this data has gaps and a lot of data discrepancies, so one needs to have a process in place to ensure that these data gaps are filled and these discrepancies are addressed. The fourth element of this asset inventory aspect is maintaining inventory's integrity. As things change, as new assets are brought online and old ones are disposed, it's very important to keep this data up to date, so one has to have a proper protocol in place to ensure that this inventory's integrity is maintained. Inventory dataset is, the dataset's integrity is maintained.

So, as I mentioned, there are different sources of asset data that one might have in their organization. It might be finance department or accounting department, which might be in the form of a purchase order, so invoices. It can be tax records. You might have some audit data that lists some of the asset information, and sometimes this can be in the project reports or proposals, performance contracting reports or proposals. So, again, it's important to note that this data can be anywhere in your _____ _____, so it's important to sift through an entire organization to find areas where this data might lie. The second part of this asset data is the vendor or commodity-related information. This is where you're trying to identify what commodities are being used in your organization, and, also, the vendors that are providing you with those details.

As you go through this process, it is important to keep an eye on how the baseline energy consumption varies by different sectors, so it's important to dedicate your efforts appropriately. For example, you might have a few datacenters _____ _____ might contribute about two percent of your energy consumption, so it's important not to spend too much effort and time and resources on assets that are not contributing towards your energy consumption very much, so you might have to prioritize things as you move through this process. The second element towards developing a comprehensive asset inventory is making sure that you understand how information is being organized in your data, how data is being organized in your organization. Sometimes, this data might be listed by department or funding source or activity types, so a well structured inventory takes into account different existing organizational principles and designs a consistent set of principles that can be applied throughout your organization.

Another aspect is as you look at this data that you've collected from different sources, you might find different discrepancies on how various terms are defined, and how various terms are labeled and what kind of data types their using. So it's better to have a consistent approach throughout your organization, and this can be done by creating a standard data dictionary that can be uniformly applied toward your organization so everybody is on the same page. If you cannot create a dictionary internally, there is this data dictionary that's been developed by DOE. It's called Building Energy Data Exchange _____. That might be able to help you so that you can have a consistent approach, a consistent and standard data dictionary that can be applied throughout your entire organization. Another important thing that might be useful is to tag giving these different assets a unique identifier so that it's easy to track their data as you move through this process.

As you retrieve data from different sources, you might think about ways as to how to reorganize that data so that it's easy to retrieve and easy to manage going forward. One way you can manage this data or organize this data is to set it up in some kind of a hierarchical manner so that it's easy to manage and organize. Another way that you can organize this data is using some kind of a tagging mechanism so that, again, you're trying to find ways as to how you can retrieve this data in an efficient manner. As you go through this process, it's important to have some kind of a tracking tool to help house this data. Again, this tool need not be sophisticated. A simple Excel or Access with some sorting and filtering capabilities will suffice.

The third element towards the step of developing a comprehensive asset inventory is as you go through this process of collecting data from different sources, as I mentioned, you're going to see gaps and data inconsistencies, so you might want to have a process in place so that you can close these gaps and adjust these data discrepancies. So this here, the schematic here kind of helps you to explain some of the concept that I'm trying to get at. The asset inventory has two components. One is the master asset list, and the second component is the utility bill data. So you pull together different sources of data to develop a master asset list, and you put together different utility bill data to develop a master utility data information. So, as you _____ these two components, you might find that these two datasets may not align, so you might have to do additional research to understand how you can _____ these two sets of tables, if you will.

So, as I said, some of this data, the utility data may not match with the asset level information, so you might have to do some additional investigation to close some of those gaps. One example may be there might be a utility meter where it doesn't have a corresponding asset, so this might be cases where the utility was not notified of _____ in assets and you continue to get billed for this asset and its consumption. The other possible reason why this might have happened is the meter or the meter number has changed by the utility and that that information is not communicated to you, or the meter might have been deactivated. What are the consequences for this? Your energy consumption and costs are somewhat inflated because you're paying for an asset that you're not using.

There might be cases where the asset does not have a corresponding utility account. This might be because the asset might be newly constructed or acquire, or this might be because this _____ _____ another master meter. So, in this case, the consequences are you might owe some additional money back to the utility. There are some cases where the utility might send you a bill with zero consumption. The implications are you adding additional burden for your personnel to track this bill with zero consumption. So these are all the things that you might encounter as you try to map your assets with your utility or commodity data.

To help you to resolve some of these differences within the asset and utility data, you might use some of these resources or tools, helpful tools where you can address these data discrepancies. The utility rate codes, sometimes the utilities have rate codes based on the asset that – sometimes the utilities assign different rate codes to meters based on the assets that they serve. So, by knowing that rate code, it will understand what asset you're looking at. For example, the street lights might be given a certain rate code compared to an office building, so by looking at that, you might be able to resolve some of these data discrepancies. The other tools that might be helpful is where you can use online mapping or GIS system, if you have it at your disposal, where you're mapping the address of the unmatched asset with the address or the service address of the unmatched meter. So, by identifying these clusters, you might be able to get a better idea as to what meter goes with what asset.

The fourth aspect is where you might be able to use some other databases, like CBECS, to help understand what assets that you're looking at. For example, you might be looking at – if you can know what the UR is, you might be able to know what kind of asset it is. For example, a higher EUI may mean that the asset that you're looking at is a datacenter and not an office building, as per your records. Again, these are all the tools that will help you to close some of these data discrepancies and gaps without collecting any additional data. But, even if all these resources – there might be cases where you may not be able [break in audio] discrepancies with these existing tools without gathering any data, you might need to collect additional data. So this is where _____ survey tools and give to the various staff that are responsible for managing these assets. Again, this can be a time consuming process. This probably looks _____ makes more sense for large organizations.

Another way you can resolve some of these data gaps or data discrepancies is to walk the site with the responsible manager to gather more insight. So, as you walk through the site, one thing you might gather is that the office space that you thought you were visiting might turn into a retail space or vice versa, so walking the site will help you to gain a lot of insight into the asset and its meter. Another possible way that you can use to resolve some of these data gaps is to _____ some of the utility relationships that you have so that they can help you collect more data. So, again, as you go through this process, it is important to keep in mind what the contribution of energy that that particular asset is providing to the overall energy consumption so that you don't spend too much time and effort on an asset that's not contributing too much towards your energy consumption or costs.

The fourth aspect, as you put together a comprehensive data inventory, it's important to have a system or protocol in place to make sure that the integrity of this database is maintained. As you know, this asset information that you gather can quickly get obsolete as more assets are acquired and old assets are disposed, and buildings are built and new meters are added. Again, all these things happen and you have to have a necessary protocol in place to make sure that asset information gets updated. You want to have a communication policy. The inventory manager has to have a communication policy to ensure that all the sales and purchases and construction activities, everything gets communicated to him or her so that he can essentially update the asset inventory database.

Also, it's important to understand if key changes to personnel that are responsible for managing this inventory has been made so that the inventory manager can find other ways to update this database. Also, one needs to have a process in place to ensure that the utility information or the commodity information, changes to commodity information gets updated in the database, as well, so you might need to work with the various utility vendors to put this process in place so that the communication happens in a timely and efficient manner. As you keep updating this inventory, it's important to circulate the revised inventory on an annual basis so that the responsible staff can know what their main energy consumers are, and, also, it helps you to inform the various personnel on the various trends that you're seeing as you analyze the data. So that's how one creates a comprehensive asset inventory. The second aspect in this process of data collection and tracking is how does one streamline the access to utility bill data.

What is it? Again, this is where you're trying to optimize the flow of data from the utility to the customer by creating shortcuts or some kind of an automation so that the data can get to the customer in a most efficient and effective manner. There are various solutions to help transfer some of this data. Why does one need to streamline it? As you know, it reduces the time spent collecting data from different vendors. It also helps you to reduce some of the errors due to manual data entry, and also allows for consistent and timely data collection. Who should work with these vendors? Energy manager, sustainability manager, database administrator, individuals with comparable roles within an organization should leverage their knowledge and utility relationships so that they can design a system that's streamlined and efficient. This takes about six to a year to put together a streamlined process to access data. This can be done in parallel with the asset inventory development process.

So these are the various data access solutions that are available to access the utility data. There's the manual way of handling this data where the utility generates a bill on a monthly or quarterly basis, and each accountholder gets this bill and the data gets entered some kind of a database manually. There is this consolidated billing, where the utility aggregates multiple accounts into one utility bill data, and the utility – the utility aggregates multiple accounts and the corresponding bill data into a single spreadsheet file, and that file gets transmitted to you electronically. Again, this is a little bit more efficient than the manual way of doing things. The third, more efficient way of handling this data is through electronic data interchange, where it's a software-to-software communication protocol where the data from the utility gets directly transmitted to your database with very minimal manual intervention.

There is this fourth way of communicating which is called application program interface, where this API will allow utilities to export cost and consumption data directly into your system an EPA's Energy Star Portfolio Manager uses APIs to communicate with different software applications. Third-party services is another option that is basically this third party acts an intermediary between the utility and you that basically gets hold of all the utility data and then enters that data into a central database that can be easily accessed by you and your stakeholders. Again, each of these solutions have their own advantages and disadvantages. As you go through this process, it's important to weigh these pros and cons to decide on what solution might work well for your needs.

There are these other emerging solutions as more and more advanced meters come online. They have the capability to generate high-frequency data, usually 15-minute interval data. This high-frequency data will help you to monitor and manage your asset energy consumption more effectively by identifying these problems quicker and designing a solution more quicker. The Green Button is another option. It's where the utility sends the data out in an XML format that can be easily digestible by various software applications. This also reduces the manual intervention and, thereby, reduces any errors associated with that. There's also this better buildings energy data accelerator, where local governments are combining forces with local utilities to help make it easier for building owners to get access to this _____ building energy use data. Again, these are all different solutions that are quickly emerging in addition to the ones that were shown on the previous slide.

So how does one identify an optimal data access solution? One needs to _____ how current process is to collect the data for various commodities. As you identify the vendor for those commodities, it's _____ to investigate. The next step is to investigate what other data access solutions or options that they have that might be able to suit your needs. You might have to do some kind of a cost benefit analysis to understand what the pros and cons are as you determine the right solution that fits your needs. The fifth _____ step is implementing the solution, testing, and making the necessary improvements for correcting any problems or issues.

The first step we're going to dig a little bit deeper to see what the step entails. This is where you're assessing what process is currently being followed to collect energy data. You need to identify all the actors that are involved in this process of reviewing and paying bills. You might also want to identify all the systems that are used to support bill payment and additional processing of this data. You might need to work with the relevant staff to obtain a list of utility vendors along with the organization's portfolio of assets they serve, and then you also need to identify what the process is on how those vendors provide utility data to the organization.

So, at the end of this step, you might be able to generate a table something similar to this, where you list the utility and the commodity that they're providing, and the number of meters that you have _____ _____ the method of collection as to how data is getting to you right now, and then the frequency of collections, whether it's a monthly bill or a quarterly bill and the database that they associated with this process. There might be one database for cost data and there might be another database for usage data, so this is something that will help you to assess how efficient your current process is.

The second step in this process is investigating the data access options, other data access options that the vendor has. As you know, one unified solution may not work with all vendors. You might have to customize, you might have to pick and choose different options for different vendors. So, as you go through this process, it's important to take into consideration any future utility infrastructure upgrades that the utility is planning so that you can take that into account while deciding on what option to choose. For example, the utility might be embarking on installing these smart meters or they might come up with new EDI capabilities, or might be able to provide data on the Green Button format. So all these things have to take into account when deciding what options to choose to streamline your data access.

So, as I said, EDI is the most efficient data access solution that's out there. Again, it is expensive to implement a full-fledged EDI solution. The factors that affect the feasibility of EDI is the number of meters, the cost. Some EDI requires that in order to transmit the data in the EDI format, they need to have electronic funds transfer, so you might need to put that in place to have this EDI functionality. So, if you have a very small number of meters, EDI may not be a good option. You might think about other options, including consolidated billing.

So this is an example of what City of Virginia Beach has done. They have been processing these large amounts of data from various bills, paper bills. They're entering this data manually. So they decided to go to an EDI solution. In the process, they were able to cut down the processing time that their personnel were spending on this data entry from 158 hours per month to 24. It's an 85 percent improvement in their productivity. So, again, this additional time could very well be spent on analyzing the data and identifying buildings that are underperforming, so it's important to look at what options you have and your other constraints on deciding what solution might be feasible for your use case.

In this case, EDI worked out very well, and they were able to save a lot of their time. But, in some cases, it may not be cost effective to go towards EDI. So, if EDI is not an option, the other option might be to use a third party to help you enter some of this data. That way, you free up some of your staff's time and expertise and dedicate their time to analyzing some of this data so that you can identify buildings that are underperforming and do some corrective actions to address their inefficiencies. Again, this is a cost benefit study, a sample cost benefit study that you can use to understand if going towards a third party is a viable option.

So, once you've chosen a solution, it's important to have a good implementation plan to implement that solution. The general guidance is it's important to obtain and leverage leadership's buy-in to improve data access. You need to identify a data champion. You need to have a comprehensive asset inventory. That's the first step in this process. Also, identify various utility contacts that you can use for helping with these data needs, and, all along, it's important to communicate the value proposition of this data tracking to help drive participation. So, if you were to stick with a manual data entry, it's important to find resources that can devote time each week for entering the data. As they go through this process, you also want to make sure that you have some kind of a QA/QC procedures in place so that _____ data can be caught early.

You can also access the utility's web portals to download some of the data. That might be faster. The other, it might be that you have chosen the consolidated billing. In this case, you might need to look at all your meters and remove some of the inactive accounts so that you're not adding additional burden of tracking these inactive accounts. You can also work with the utility to help put together a template or a format for this consolidated billing that works well with your existing systems. You might also want to ask the utility to align all the bills to a consistent time period so that it's easy to analyze. It can also determine the medium of exchange, whether it's a download from their website or is it e-mail, so that's another thing that you might have to work out, work those details with the utility. There also has to be a process in place to review any account changes that you might have or that they might have so that that information gets updated as it happens.

So this is an example. This is a case study from City of Knoxville that went towards consolidating their different bills. The city worked with the municipal utility and developed a consolidated billing for electricity, gas, water, sewer, and other consumption data related to fire hydrant and other lighting infrastructure. In the process, they were able to reduce their time considerably on entering data into their database. According to them, it takes about one hour per month to import data. Previously, their project manager was spending 8 to 10 hours per month on these data management activities. So you can see there's _____ [inaudible due to audio distortion] additional time that they were able to save can be spent on analyzing this data and identifying inefficiencies in their assets. The other thing that they were able to do is use some of this data for measurement and verification for their performance contracting projects, so since all this data exists in one place, you might be able to track the data and see how effective these performance contracting projects are from an _____ standpoint.

If you have chosen the EDI to be the option to access data, again, it's the same process you might – there are several ways as to how one can implement this EDI solution. Either you can do it in-house or you can use a third party _____ _____ subscription model that can put together a solution for you. There are various ways how data gets transmitted, how this EDI file can get transmitted. You also need to determine the frequency as to how this data gets downloaded, and sometimes this EDI, when you sign up for EDI, the utility wants you to do an electronic fund transfer, so you might have to do some additional work to ensure that the funds get transferred electronically. Also, there might be cases where the accounting office might need the hard copies of this bill, so you might need to keep that channel open so that the hard copies of these invoices go to the accounting department.

As you implement this solution, you might have to refine it as you find some errors and other things. As I said, EDI is one of the most effective solutions but it can be expensive to implement. It depends on your number of meters and the vendors that you have, the number of vendors you have on your roster. There might be cases where you're better off going to a third-party service. If the utility doesn't have an EDI option and consolidated billing is not something that the utility is willing to provide, then the other option is to go towards engaging a third party that acts an intermediary between you and the utility that essentially collects all this data and enters the data, and sends you the processed file that you can use. Again, there's no one-size-fits-all approach.

There might be cases where you might have some meters that might be using an EDI, you might have some information that's being entered manually, so you have to take all that into account when selecting this third party. So, if EDI is predominantly what's being used, so you might want to contract with an EDI specialist that can act as this third party to process some of this data. Some of these companies also offer a complementary data tracking tool that can be used internally within your organization to access and analyze this data as it gets updated.

So this is case study from Portland Public Schools. The Portland Public Schools used to enter their data manually. They have eliminated this manual process by going towards a third-party vendor, and they were able to reduce some of their staff time, they were able to reduce some of the time that their staff was spending on this data entry, and they were able to allocate that time towards identifying usage anomalies and identifying performance deficiencies in their schools. It decreased their invoice processing time. It also automated some of the usage and cost notifications. It helped identify some of the process flow inefficiencies and the access to data is much more quicker, and you can identify any anomalies much quicker and design solutions in a quicker manner. As you can see here, before they've done to this third-party vendor, they were spending about 12 weeks, having about 12 weeks of delay in getting this energy data for review, so by going to this third-party contract, they were able to cut down that time in half to six weeks. So their efficiencies have been improved and the staff are dedicating their time to more important things, like finding inefficiencies and anomalies in their buildings.

So the third aspect of data collection and tracking is understanding what kind of tools and analytics that need to be in place to process some of this data that's being collected. The first aspect to think about is the structure. As you start collecting data, it's important to make sure that the data structure matches with your organizational structure and the data is easy to enter, and the system that you have in place can work well with other systems that you have in your organization. The flexibility is how easy is it to add additional records or additional fields. That's also a key when deciding what data tools to use. Obviously, backup and security is key when looking at some of these tools.

Analytics, the second piece, the second component of these data tools and analytics, you need to make sure that these data systems have the _____ processing capabilities and can do some basic analysis, including search and sort functions, and also need to make sure that you have some kind of quality control in place to ensure that the data that's coming in doesn't have any issues. The third element of this is making sure that whatever data gets reported gets reported in a manner that's useful to various stakeholders, so you might have to design different views or dashboards so that they can clearly understand the value proposition for this data. So DOE has this open source platform that can help you that gives you some of these functionalities. It's called SEED. It's Standard Energy Efficiency Data. Again, this is an open source tool that's developed by DOE, and you can also use Energy Star Portfolio Manager to get some of this functionality.

Again, there are different ways as to how this data management system can be implemented. Either you can develop something on your own or you can go to a third party that can give you this system based on a subscription model. Depending on your in-house IT capabilities, you can go either way. Again, the data systems that you put in place have to have the necessary functionality in place to make sure that new data can be easily entered and that all data can be revised, that it's compatible with other systems and databases that you have in your organization. You also need to make sure that adding additional fields and records is easy, and the data backup and security is paramount when deciding what kind of systems that you are going to implement.

The analytics, this is where you're analyzing some of the – analytics provides you with some of the capabilities to process some of the data that you've collected throughout this process. You should be able to combine, map, and cleanse different datasets that you gathered from different sources. You should be able to synchronize different time intervals, because each dataset might be sampled differently and it might start at a different timestamp. You might be able to do some interpolations and extrapolations of this data, compute some basic statistics. Other data analysis capabilities that you might want is to compare the performance of the asset with its historical performance, or you should be able to compare the performance with other buildings or assets within its category. You might be able to do some various search and sort functions. You might be able to identify some anomalies. For example, if there's a big jump in electricity consumption in a building, you should be able to have a functionality to quickly find that out and act on it as you find those. Data auditing is also a good functionality, where if a number that gets entered is way off, you should be able to raise some flags so that you can focus on addressing those data quality issues.

So, as I said, presenting this data in a manner that's easily relatable is an important – presenting this data in a manner that's easily relatable by different stakeholders is a key, important step. So this is an example of a building dashboard that we put together for a different report. This is where we're identified different stakeholders. There are different stakeholders in this example: the public, occupant. Based on their rules in the organization, you might be able to provide access to this data and provide customized reports so that it's easy to understand what the value proposition is for various stakeholders that are involved. For example, if it's a general public, all they need to see is the top-level energy consumption information. They don't need to look at each individual building or asset level information.

For example, if it's an occupant, they need to be able to compare their energy consumption for their space and see how well they're performing compared to other occupants or to their historical performance. In that case, you need to provide them with daily or weekly data so that they can do some meaningful comparisons. If you are an agency administrator or a building operator, you should be able to track building consumption at a building level or a sub level so that you have the necessary tools to understand what buildings are consuming more energy than others, and then focus your attention on those underperforming buildings. If you are a building manager or a facility personnel, you need to have all levels of energy consumption data so that you can support the energy management at both macro and micro scales.

In this case, there is a unique dashboard that's created for researchers. This is where the researchers have access, can use tools and analytics. By having the right access to the data, they can use tools and analytics to identify wasteful areas or areas that are consuming high energy and focus on some specific problems that might need some additional research. Again, these are different stakeholders, and each of these different stakeholder have their own needs or different value propositions for the data that you're generating, so it's better to customize the same data in multiple ways so that they get to see the data from their angle.

So this is the case study of the State of Maryland. They put together a centralized database to track their consumption of different commodities. They have 120 vendors and about 16,000 utility accounts, 120 accounts payable departments, and 58 agencies. So they were able to put together this centralized database that help them with their data collection and tracking process, and they've gone to a third party to implement this centralized system. The cost of the contract is about $1 million to manage a budget of $200 million. In the process, they were able to save about $17 million by optimizing their energy purchasing _____. In this case, they went to this block and index commodity purchasing program.

As I said, that helped them to save about $17 million by going towards this new purchasing mechanism. So they were able to free up their time, and their personnel were able to dedicate their efforts and time towards identifying the anomalies in their buildings, and identifying the assets that are causing some of the inefficiencies. As you can see, their energy consumption has gone down sequentially starting in 2009, and, cumulatively, they were able to reduce their energy consumption by about 15 percent by the end of fiscal year '15.

So the fourth aspect of data collection and tracking is making sure you have the necessary organizational structure in place to not only collect data efficiently and effectively, but also to make sure the data gets updated as things change. We have two _____. There's a decentralized way, organizational structure where the utility bill data gets transmitted to different accounts payable departments and the data gets entered into a database, and all that information gets rolled up to a central energy or sustainability office. Again, _____ [inaudible due to audio distortion] there are too many human touch points, and this can cause some data entry problems, or cause some errors in data entry. The other _____ is more centralized, where the energy or the sustainability office receives all the bills for that organization and the data gets entered into one centralized database, and that information about the payments is disseminated to individual accounts payable. This is a more centralized way of doing things. This is less human touches and maybe less error-prone. The other option is somewhat partially centralized.

This is what the Department of General Services – I believe it's the State of Maryland – is using. They have this third party that acts as an intermediary between the customer and the utility. All the utility data gets sent to this third party that enters this data in a centralized database, and all the other entities or other personnel interacts with this database, and it's also the utility, along with sending the bills to this third party, they also send the bills directly to the account payables, and the account payable department also interacts with the database as they do different activities. So, again, this is one centralized database that's mostly maintained and administered by this third party that acts as a central depository for the various data, and those various departments and personnel interacts with the database for managing their day-to-day operations.

The fifth aspect of data collection and tracking is engagement and communication. As I said, it's very important to engage and communicate the results of this program in effective manner. The dashboards that I've showed you on a previous slide is a good way to essentially put together reports that speak their language and they can understand the value proposition for them. So the one-size-fits-all doesn't work _____ have to tailor your messages in such a way that they're clear, concise, and timely so that everybody knows the value of this data collection and tracking effort, what the value proposition is for this data collection and tracking _____.

In summary, a robust data tracking strategy is a foundation for strategic energy management and pays additional dividends. If implemented right, it will help you to develop a robust tracking system. Development of this robust tracking system will take time, so it's better to start early. As I mentioned, it's important to start with assets that contribute the most towards your utility consumption and scale your efforts accordingly. Develop an asset inventory and a system, and it's important to keep that system up to date as things keep changing. It's important to dialogue and collaborate with utilities to understand what right or optimal way to access this utility data. Medium to large entities use a combination of approaches _____ one-size-fits-all. You might have to resort to different ways to access this utility data. A robust analysis tool is critical for making good use of this collected data. An integrated or centralized structure can be a win-win situation for all the stakeholders.

It's important to implement available solutions while remaining flexible so that you can adopt these new solutions as they become available, like Green Button data or EDI. As they become available, you should be able to nimble enough to adjust accordingly to take advantage of those options. So this is the final slide showing the references and resources. As I said, we have a module dedicated to data cleansing, and we presented on this last April, and this is the link that will take you directly to that webinar. As I said, next week, we're going to have this other webinar on how can one manage, analyze this benchmarking data analysis, so if you're interested in that, please sign up. It's the same time next week. Thanks for your attention. I'm happy to take any questions.

Crystal McDonald:    Thank you, Shankar. We do have a few questions that have come through. Let me just open my window here. The first question was around links for the resources. That question came through early on, so I'm glad you summarized the list of resources along with the hyperlinks that you just mentioned. When we share the PowerPoint slide, you'll have access to the hyperlinks for the resources that Shankar covered. The second question is around what are the most user-friendly energy data management programs available? Do you have any recommendations or can you point to a few sources around energy data management programs?

Shankar Earni:           Yeah. As I mentioned, the one thing is SEED, the DOE SEED tool that will help you to not only cleanse the data but also it has some analytic capabilities that you can use to process some of this data. There might be some other solutions that are being offered by various private companies. I don't have it handy, but I can –

Crystal McDonald:    Sure.

Shankar Earni:           − _____ _____.

Crystal McDonald:    No, that's fine. Okay, thank you, and then I wanted to call out something on Slide 23. You mentioned the Better Buildings Energy Data Accelerator, so I just want to inform our audience the data accelerator has a toolkit that is now available on the Better Buildings solution center. Okay, do we have any _____ [inaudible due to audio distortion] the screen here. Okay, so with that said, I just want to remind you again of the next and final session in this series of energy data management. We have the session next Tuesday at the same time, "Benchmarking Data Cleansing: A Rite of Passage Along the Benchmarking Journey," so we invite you to join us then. Again, I want to thank Shankar for laying out the steps for energy data collection and tracking. I hope you found this information to be very useful, and it will be posted on the state and local solution center in the near future, so look for a follow-up e‑mail from us. We hope to join you again next Tuesday.

Shankar Earni:           Can I say something –

Crystal McDonald:    Oh, sure.

Shankar Earni:           Next week's session is going to cover the topic related to the analysis, specifically. This is –

Crystal McDonald:    Okay.

Shankar Earni:           − when we have a cleansed dataset and _____ trying to analyze that dataset. The one on cleansing has already been done, and that's the one that's posted on your website, _____'s website.

Crystal McDonald:    Okay, we'll need to get that information updated. Thank you for that clarification. Any final thoughts or questions from anyone? All right. Well, you have a few minutes of your afternoon back, and, again, thank you for your time and attention today. We'll see you next week.

[End of Audio]