DatabookAI takes the power of Databook to the next level, delivering new trusted insights, proactive guidance, and personalized assets through a conversational AI interface you can access where you already work. Initially available via a conversational AI web application, DatabookAI will also be available through integrations within sellers' existing workflows, including collaboration apps such as Slack and AI apps such as Microsoft Copilot for Sales in the near future. With DatabookAI, sellers can quickly find the insights they need to save time on account research, pipeline generation, outreach creation, customer meeting preparation, and much more.

DatabookAI helps sellers secure the right meetings by identifying the best accounts and buyers and the most relevant use cases and solutions to discuss with buyers. It also helps sellers prepare for those meetings so they can execute effectively in every customer meeting.

DatabookAI, through an easy-to-use conversational interface, offers insights that help sellers increase opportunity conversion rates, generate more pipeline, and grow the average contract value (ACV) of their opportunities.

We believe that DatabookAI's conversational AI allows it to combine different data points into sophisticated yet easy-to-understand insights perfectly suited to the use cases in which sellers have already used Databook. We have proof points showing that the use of the Databook web app is associated with higher ACVs and more pipeline for accounts where it is used, and we believe that DatabookAI will offer even more value for enterprise sellers and sales organizations.

DatabookAI is built for enterprise sellers to help them across their sales cycle by providing relevant, up-to-date trusted insights. The key advantages that DatabookAI provides over ChatGPT (and other foundational LLMs) include:

We have built DatabookAI as a Retrieval Augmented Generation (RAG) application to reduce hallucinations to near zero, and improve the accuracy and relevancy of DatabookAI responses. In simple terms, the RAG app uses a foundational large-language model (LLM - for example GPT-4o) to understand the intent of the user’s query and plan how to answer it. DatabookAI will then find relevant data to ‘ground’ its response from Databook’s high-quality datasets including premium financial and firmographic data and proprietary data such as strategic priorities and management intent. The LLM will then synthesize the data into a response for the user.

Relating to the proprietary and third-party datasets that are used to ‘ground’ DatabookAI responses, Databook has measures in place to ensure high data quality when ingesting and processing data. For datasets such as company strategic priorities or financials we are able to trace their origins back to authoritative sources such as company investor documents, earnings transcripts or company filings. Key datasets are refreshed daily. Databook and its third-party data suppliers also have automated data checks to verify the accuracy and recency of the data used to inform DatabookAI responses. Databook also has its own program of quality assurance checks by human analysts to provide additional reassurance.

This approach is different from conversational AI apps such as ChatGPT or Perplexity that rely on general web searches to answer some queries, especially those that relate to recent developments after the LLM was trained and so the answers won’t be found in the LLM’s training data. For example, for queries relating to company strategic priorities these conversational AI apps will look at sources beyond the company’s official documents (for example, media reports or SWOT analyses by MBA students), thus increasing the chance of inaccurate data being included in a response.

We have designed DatabookAI to help enterprise sellers better understand their customers and prospects, empowering them to be more productive and effective in their roles. We believe that DatabookAI, when used for its intended use cases, has a positive impact on our users, customers, and buyers who interact with them.

Databook is intended for a specific set of business use cases relating to enterprise sales. As such, some broader concerns about bias and misinformation in generative AI are less relevant. The design of our retrieval-augmented generation (RAG) application ensures highly relevant and accurate answers. We also have guardrails in place to mitigate against misuse of the product and have tested prompt injection attacks where a bad actor tries to misuse the product.

Databook follows industry best practices in relation to information security and data privacy. You can find more information about specific controls on our security portal. Our security controls are continuously monitored and audited for our SOC 2 Type II certification. A copy of our SOC 2 report is available on request.

We prioritize product security through a robust set of measures to safeguard your data and application:

Databook has completed its most recent annual SOC 2 Type II audit, which now includes DatabookAI in its certification scope. Our core Databook platform and DatabookAI share common infrastructure and security controls, ensuring comprehensive protection for all customer data across our product ecosystem.

Our SOC 2 Type II compliance demonstrates our commitment to maintaining robust security measures throughout our platform, including:

This certification confirms that Databook maintains the highest standards of security, availability, and confidentiality for all customer data, regardless of which product features you use.

DatabookAI may refer to the solutions, use cases and case studies that you have configured on the Databook platform to personalize its responses. This data is only available to inform responses to queries from users in your organization. (This data is also already being used to provide insights to your users on the Databook web app and use of this data for that purpose is covered by your existing cloud services agreement.)

The response may also refer to information included in the user’s query - for example, if a user mentions the name of an account they are interested in. This data is only used to inform responses to that user’s queries.

Databook stores all customer data, including user queries, in logically separated tenants with strong data isolation (including row-level security). Customer data is only accessible to users within the same customer tenant.

Depending on your environment, we use various AI services, such as OpenAI or AWS Bedrock.

We use Enterprise APIs to generate responses. Data shared with AI service providers is not shared with their customers, made available to the service providers, and not used to train their models.

You can read more about data privacy with OpenAI here.

You can read more about data privacy with AWS Bedrock here.

A complete list of sub-processors for the Databook service can be found here.

In the future, we intend to allow enterprise customers to choose an alternative LLM to generate DatabookAI responses. We will announce details of this bring-your-own-model capability, including which LLMs we intend to support, in due course.

No. Your data would not be used to train OpenAI models. Please see this information about data privacy with OpenAI.

As an enterprise customer, the use of your data will be governed by the agreement you sign with us. Our default position is that we will not use data from our enterprise customers to train models that are shared with other customers.

To provide a higher-quality service, we may in the future choose to train our own AI models to support DatabookAI, unless specified otherwise in your customer agreement. For enterprise customers, we would train individual models exclusively for that customer’s use.

Individual models for enterprise customers may be trained on data from non-enterprise users (e.g. trial or freemium users) to improve the quality of responses we provide still further. Data from enterprise customers would not be used to train this shared, underlying model.

If an enterprise customer chooses not to allow any training with their data, then the insights and responses that DatabookAI provides their users would come only from this shared, underlying model and not be as customized to their GTM strategy and operations.

We are building integrations to bring DatabookAI’s account intelligence and conversational AI to where your sellers work (for example - in collaboration apps such as Slack or in AI apps such as Microsoft Copilot for Sales). We will be working with partners such as Salesforce and Microsoft to get the appropriate partner certifications and security approvals for these integrations.

The integrations are optional and, in some cases, may require some bidirectional syncing of data. Our team can provide more details of these integrations if you would be interested in integrating DatabookAI insights further into your team’s workflows.

We comply with data privacy legislation such as GDPR and CCPA. Databook completed a Data Protection Impact Assessment (DPIA) for DatabookAI to identify, analyze, and minimize privacy risks.

Databook’s data is hosted by AWS in the us-east-1 region.

You can contact the Databook Privacy & Security team at [email protected].