To provide a practical and hands-on approach to learning these concepts, the module will use Grantable as a demonstration platform because it is purpose-built for these uses. You do not need to subscribe to Grantable in order to apply most of these techniques and principals, however, to realize the time-savings and process improvements described in this module, it will be difficult to achieve the same results by piecing together other kinds of software.
​
Throughout this module, we will return to a core theme of grant-seeking efficiency, which is to reuse high-quality grant proposal content as much as possible when addressing similar questions between proposals. Large language models excel at this kind of formatting and reformatting work, which is why generative AI is such a powerful tool for grant-seekers to use.
​
By the end of this module, you will know how to assess whether or not using AI to augment your grant-seeking work is appropriate for your situation, how to identify high-quality source materials from your existing writing, and seamlessly incorporate AI into legacy grant-seeking workflows to do everything from drafting responses to brainstorming outlines to grant reports.

No one absolutely needs to use AI tools in grant-seeking. Declining to use this emerging technology is a reasonable position to take, though people who abstain from AI will find it increasingly difficult to do so as the world's largest software companies quickly add generative features to workplace staples like email, word processors, and spreadsheets.

Some of the largest software companies in the world, including OpenAI, Microsoft, and Google, are rapidly investing in ways to bring more and more generative AI features to all of their most popular apps, including Claude, ChatGPT, Word, Google Docs, Excel, Sheets, Outlook, and Gmail.

As a grant professional, it's likely you have experienced all of these platforms at one time or another. Soon, all of these programs, many of which we are familiar with, will have features allowing users to input prompts or instructions in natural language, and the software will be able to automatically create more and more sophisticated outputs.

Many grant professionals have already found ways to use a combination of word processors and chatbots to accelerate their grant-seeking work at almost no additional cost. For people just beginning to work on grant applications, this can be a great way to start and learn, not only about AI, but also about grant-seeking best practices.

While chatbots can be a valuable assistant, it's important to exercise critical thinking, verify information from reliable sources, and ensure the quality and accuracy of the final grant proposal.

Some organizations have particular workloads, like grant-seeking, that are so important to operations and take up so much time that it can be worth looking into specialized tools designed for making specific kinds of work easier and more efficient. Let's look at what traits good AI tools have in common.

Perhaps the most obvious characteristic of a good AI tool is that it does not promise to be an 'easy button' for complicated work products. While generative AI is already extremely capable, and getting better each day, we are not yet at the point where these systems can completely take over for people in many complex work situations.

It's generally a good idea to avoid products that hide everything behind a curtain, asking only that you input a bunch of information, click a button, and voila! What this kind of software really does is deprive you of opportunities to influence the quality and direction of the work product. It is over-automated.

Good AI tools should take time to master. They should have lots of ways for you to control the performance of the AI, to iterate, and arrive at outputs that incorporate human intelligence and skill.

Another trait of high-quality specialized tools is that they give you precise control of specific AI abilities when you want it, and get AI out of the way when you don't. For example, you may want to call your AI assistant to place on a page when you want help brainstorming some ideas, and once you've found one you like, you should be able to get back to writing as normal.

AI chatbots like Claude and ChatGPT are popular examples of large language models designed to be extremely versatile so that anyone in the world can interact with it and find it engaging and useful to some degree. For quick queries or having fun, chatbots can be ideal.

For certain kinds of work with well-defined processes and phases, it can actually be more effective to constrain large language models in ways that focus their value for the user. This can be done on the back end with prompting, and on the front end with visual controls that can be faster and more efficient to use to guide AI than by typing out written prompts each time you want something to happen.

A good analogy is to think about a chatbot like a multitool or pocketknife that you can use to cut, screw, file, and grab things around a workshop. A specialized AI tool is like a power drill that can only do things that require rotation. While the multitool might be the way to go for, say, simple cutting work, if you need to make a bunch of holes, you're better off using the drill.

There may come a time when piecing together features from a patchwork of different free software becomes too burdensome. This is especially true when the volume or complexity of grant proposals increases significantly, or when a grant professional is working on grants across multiple programs or for various organizations. In these situations, it can be extremely challenging to manage source material and the constant switching between an AI chatbot and a word processor can become a huge, hidden time-waster.

One solution is to consider using Grantable, a software specifically designed for grant-seeking with generative AI features. Grantable can be thought of as a combination of three essential components: a smart content library, a word processor, and an AI writing assistant.

When an organization is ready to apply for grant funding, and they have found a high-quality opportunity to pursue, the work of producing a grant application can begin. The most efficient organizations build upon existing work by repurposing language from recent proposals to produce new grant applications more quickly.

Generative AI tools can supercharge this process by leveraging the incredible abilities of large language models to help locate and reformat grant proposal content in fractions of a second.

You're working on a grant application and you come across a question that you know you've answered before. The problem is that you can't recall which of the dozens of recent grant proposals in your content library contains the previous answer.

You proceed to check file after file, searching through shared drives and inboxes looking for keywords and clues to help you locate what you’re looking for. Finally, after a few minutes, or in the worst cases, several hours or days, you locate the passage, which you copy and paste into your new draft.

If you are working on a grant application and need to find a previously answered question, consider using Grantable, a generative AI grant writing software. Grantable can instantly search through all the content in your library and retrieve meaningful excerpts that you may wish to reuse. This saves you from having to go through each document and folder individually. You can search your entire library at once and either select excerpts yourself or allow the AI to recommend resources. The entire process usually takes only a few seconds.

Storing grant proposal content in an organized manner that facilitates easy referencing is a fundamental practice for effective grant-seeking. Typically, grant professionals store documents on a local or shared drive and employ a labeling system for folders, often categorized by year and grantor. Grantable takes this best practice a step further by enabling users to tag documents and folders for an additional level of categorization, and with a powerful search feature that searchers your entire library by keyword.

Additionally, Grantable utilizes AI to read and reference these materials.

Grant professionals can now quickly find and reuse specific excerpts, saving time and effort compared to traditional methods. With Grantable's search feature and AI assistant, maintaining a well-organized content library is easier than ever. There is no longer a need to create and update boilerplate documents or spreadsheets for template language. Accessing your best and most recent content is simple with Grantable.

Using Grantable is intuitive for anyone familiar with word processors like Microsoft Word and Google Docs. Writing progresses seamlessly from the cursor on the screen, and AI assistance can be accessed either by creating a line break and pressing the spacebar, or by highlighting text and pressing Command/Control + J.

One notable difference from legacy word processors is a simplified set of formatting options.

Unlike chatbots, which do not allow for easy text editing and requires constant copying pasting into other text editors, Grantable combines the generative AI abilities of large language models, while placing control of the document in the hands of writers.

Grantable’s AI grant writing assistant is a large language model that has been tuned to act and assist in grant-seeking. It can help to draft responses to prompts using relevant source material from the applicant organization, and also more open-ended tasks such as creating outlines and brainstorming ideas.

These improvements help to keep the AI grant writing assistant on track by limiting its use of outside data, information not from the applicant organization, and to preserve the same style of writing as the source material.

Grant professionals working in Grantable, alongside the AI grant writing assistant, have the benefit of all their work automatically becoming part of their smart content library. This library stores the work and makes it readily available as potential source material. This continuous cycle of improvement leads to increased efficiency and better outputs from the AI grant writing assistant.

To determine the right set of tools to use, grant professionals should ask the following questions to identify which option is the right one for you:

The more of these questions you answer affirmatively, the more likely it is worthwhile for you to invest in generative AI software that is specifically designed for grant professionals.

Setting up a Grantable workspace can be as simple as uploading a single high-quality grant proposal as initial source material. While even one strong grant proposal can suffice to begin, we suggest uploading a few recent proposals to offer a range of sources for you and your AI grant writing assistant to draw from.

When writing, it’s far easier to reduce or remove detail than it is to add more of it. By uploading your most thorough writing samples, you give yourself and your AI grant writing assistant the most optionality when working on your next grant proposal.

Include your most recent writing samples that contain information your organization considers current.

Upload writing samples you and your organization have invested the most effort to write regardless of whether or not these grant applications were ultimately awarded or not.

If you have never written a grant application, you may also consider other common documents that contain information about an organization such as:

Ensure these other types of writing samples are well-formatted and in .docx format before you upload them

When working on grant proposals, it is common for grant professionals to encounter poorly formatted documents. For instance, this often occurs when grantors provide highly stylized PDF files that are challenging to edit and result in awkward formatting when translated to a .docx file. Similarly, issues arise when grant applications are submitted through online portals and copying and pasting into a word processor leads to strange spacing and formatting.

Not only is this a headache for grant professionals, poorly formatted documents are also less useful as reference material because they are more challenging for your AI assistant to decipher when helping you locate potential source material from your content library. For this reason we highly recommend taking a bit of extra time to ensure the formatting of your documents, or at least the sections you may wish to reference in the future feature clean formatting.

Now, let's take a step back and examine how your documents are stored in your Grantable account. The way you organize your folders and files, as well as how you tag them, does not affect the AI's ability to read and reference documents.

However, organizing your content in labeled folders with informative tags is intended to help you stay organized and easily locate specific sources, enabling you to direct the AI assistant to use or avoid certain sources more efficiently.

At the top level is a folder for each year’s worth of grant applications. For applications that span two calendar years, we recommend filing the materials in the year in which the application was begun.

Within each year’s folder are folders for each grantor, and within each grantor’s folder are all the files associated with the application.

This structure is highly effective as it enables grant professionals to easily monitor all components of a specific application and ensure that source materials include up-to-date information about the applicant organization.

Adding tags to a filing system, as described above, enables grant professionals to enhance the precision of information storage and referencing. One common use of this feature is to differentiate between organizational programs and grant types.

For example, an organization may have multiple program areas for which it seeks grant funding. In such cases, a grant professional can add tags to documents and/or folders to indicate the corresponding program for each application.

By using tags, a Grantable user can filter search results to specific tags when looking for source materials or when instructing the AI assistant to find references. This allows for excluding results from other programs and only retrieving results relevant to the desired program for a specific application.

With a well-formatted and well-organized content library, you are ready to enlist the help of your AI assistant to work on a grant application.

Most grant professionals consider it best practice to work on a grant application in a reliable text editor rather than the original format of most grant applications, such as PDFs and web portals. Working in these other formats can be very difficult and may result in inadvertently losing hard work if the application does not have the ability to save progress.

If the grant application is in .docx format or you know how to convert it to .docx format, this file type can be uploaded directly to Grantable in the home screen. Once uploaded, you can double click into the document to begin working on it.

Grant applications that are web-based or in PDF format can also be copied and pasted into a blank document in Grantable.

From the home screen, select the ‘New File’ option, and then select all of the grant application text, copy, and paste it into the new file in Grantable. You should now be able to begin editing it as you normally would.

There are two ways to call upon your AI assistant within the workspace.

Locate the sparkle AI icon in the toolbar at the top of the page.

Select any text on the page and press Command/Control + J to call your AI assistant. Using this method will direct the AI assistant to act upon the highlighted text however you instruct it to.

Learning how to instruct generative AI large language models to achieve desired results involves a learning process and lots of trial and error. These models are highly flexible in terms of the inputs they can receive and the outputs they can generate.

Grantable has created a few shortcuts for common prompts that grant professionals often use. These shortcuts include "Answer this," "Improve writing," "Shorten this," and "Lengthen this." These shortcuts not only save you time but also give you ideas on how to instruct the AI assistant.

Many users ask the AI to modify the tone or complexity of the text using prompts such as "Please make this more conversational" or "Please simplify this selection."

The AI assistant is also capable of more open-ended tasks. For example, you can create a new line, press the spacebar, and instruct the AI to "Draft an outline for a letter of interest to a private foundation." The AI assistant will then proceed to draft the outline, which you can use to fill out and revise with AI assistance if desired.

Feel free to experiment with various prompts within a given AI interaction as Grantable will continue to track each AI response as a version, which you can toggle through to return to an output.

Becoming a good prompter requires practice, precision, and patience. Over time, you’ll develop an intuition for how to instruct the AI to yield better results with fewer prompts.
​

One of the most powerful capabilities of large language models is the ability to revise text quickly and effectively in a nearly limitless number of ways. In Grantable, this is achieved by highlighting a selection of text you wish to act upon with AI assistance, and to instruct the AI how to proceed.

For example, to shorten a passage that is several hundred words long to fit within a limit of 100 words or fewer, highlight the text, press Command/Control + J to expose the AI command line, and prompt the assistant to:

The AI will attempt to reduce the length of the passage, while preserving the essential information.

If you are not satisfied with the initial response, you can try additional prompts and each version outputted will be stored for the duration of this interaction with you AI assistant, so you can review several different options to select one that is most suited to your needs. If you exit the interaction, these versions will be reset.

An advanced technique some users enjoy is to use your document as a brainstorming space, or whiteboard, and to rely on the AI assistant to combine disparate threads and thoughts together. In this practice, the user will search for content and paste all the content onto the page, highlight all of it, and prompt the AI in a way that asks it to combine all the material according to the user’s instructions.

Once you have a feel for how to use AI to help you work with content you provide to it, you can also experiment with more unstructured or open-ended tasks, ones in which you may not yet have a clear idea of the desired outcome.

For example, the AI assistant can be asked to help with structuring a response or generating ideas for formatting a letter of interest. To do this, the user would create a line break using the Enter key and then press the spacebar to access the AI command line. An open-ended prompt might look like this:

Prompting the AI assistant in this way should result in it generating an outline for a letter of interest, along with ideas on how to complete each section. Once you have added the outline to the page, you can use AI again to assist you in completing each of these sections.

This short quiz is only meant to help you check your understanding of these materials. Your score is not recorded, so please write it down if you want to keep track for your own records.