Project Icon

genai-for-marketing

Google Cloud生成式AI助力营销优化的开源项目

这个开源项目展示了Google Cloud生成式AI在营销领域的应用。项目提供了从环境配置到实际使用的详细指南,涵盖营销洞察、受众分析、趋势发现和内容生成等功能。通过整合Vertex AI、BigQuery和Workspace等服务,项目旨在提升营销效率,支持数据驱动决策。项目还包含多个Jupyter笔记本,用于演示核心概念。此外,项目集成了Looker仪表板用于数据可视化,以及Vertex AI Search用于改进内容搜索体验。整体架构设计支持从数据分析到内容生成的全流程营销优化。

Generative AI for Marketing using Google Cloud

This repository provides a deployment guide showcasing the application of Google Cloud's Generative AI for marketing scenarios. It offers detailed, step-by-step guidance for setting up and utilizing the Generative AI tools, including examples of their use in crafting marketing materials like blog posts and social media content.

Additionally, supplementary Jupyter notebooks are provided to aid users in grasping the concepts explored in the demonstration.

The architecture of all the demos that are implemented in this application is as follows.
Architecture

Repository structure

.
├── app
└── backend_apis
└── frontend
└── notebooks
└── templates
└── installation_scripts
└── tf
  • /app: Architecture diagrams.
  • /backend_apis: Source code for backend APIs.
  • /frontend: Source code for the front end UI.
  • /notebooks: Sample notebooks demonstrating the concepts covered in this demonstration.
  • /templates: Workspace Slides, Docs and Sheets templates used in the demonstration.
  • /installation_scripts: Installation scripts used by Terraform.
  • /tf: Terraform installation scripts.

Demonstrations

In this repository, the following demonstrations are provided:

  • Marketing Insights: Utilize Looker Dashboards to access and visualize marketing data, powered by Looker dashboards, marketers can access and visualize marketing data to build data driven marketing campaigns. These features can empower businesses to connect with their target audience more efficiently, thereby improving conversion rates.
  • Audience and Insight finder: Conversational interface that translates natural language into SQL queries. This democratizes access to data for non-SQL users removing any bottleneck for marketing teams.
  • Trendspotting: Identify emerging trends in the market by analyzing Google Trends data on a Looker dashboard and summarize news related to top search terms. This can help businesses to stay ahead of the competition and to develop products and services that meet the needs and interests of their customers.
  • Content Search: Improve search experience for internal or external content with Vertex AI Search for business users.
  • Content Generation: Reduce time for content generation with Vertex Foundation Models. Generate compelling and captivating email copy, website articles, social media posts, and assets for PMax. All aimed at achieving specific goals such as boosting sales, generating leads, or enhancing brand awareness. This encompasses both textual and visual elements using Vertex language & vision models.
  • Workspace integration: Transfer the insights and assets you've generated earlier to Workspace and visualize in Google Slides, Docs and Sheets.

Notebooks and code samples

The notebooks listed below were developed to explain the concepts exposed in this repository:

The following additional (external) notebooks provide supplementary information on the concepts discussed in this repository:

  • Tuning and deploy a foundation model: This notebook demonstrates how to tune a model with your dataset to improve the model's response. This is useful for brand voice because it allows you to ensure that the model is generating text that is consistent with your brand's tone and style.
  • Document summarization techniques: Two notebooks explaining different techniques to summarize large documents.
  • Document Q&A: Two notebooks explaining different techniques to do document Q&A on a large amount of documents.
  • Vertex AI Search - Web search: This demo illustrates how to search through a corpus of documents using Vertex AI Search. Additional features include how to search the public Cloud Knowledge Graph using the Enterprise Knowledge Graph API.
  • Vertex AI Search - Document search: This demo illustrates how Vertex AI Search and the Vertex AI PaLM API help ensure that generated content is grounded in validated, relevant and up-to-date information.
  • Getting Started with LangChain and Vertex AI PaLM API: Use LangChain and Vertex AI PaLM API to generate text.

Environment Setup

This section outlines the steps to configure the Google Cloud environment that is required in order to run the code provided in this repository.
You will be interacting with the following resources:

  • BigQuery is utilized to house data from Marketing Platforms, while Dataplex is employed to keep their metadata.
  • Vertex AI Search & Conversation - are used to construct a search engine for an external website.
  • Workspace (Google Slides, Google Docs and Google Sheets) are used to visualized the resources generated by you.

1- Select a Google Cloud project

In the Google Cloud Console, on the project selector page, select or create a Google Cloud project.

As this is a DEMONSTRATION, you need to be a project owner in order to set up the environment.

2- Enable the required services

From Cloud Shell, run the following commands to enable the required Cloud APIs.
Replace <CHANGE TO YOUR PROJECT ID> to the id of your project and <CHANGE TO YOUR LOCATION> to the location where your resources will be deployed.

export PROJECT_ID=<CHANGE TO YOUR PROJECT ID>  
export LOCATION=<CHANGE TO YOUR LOCATION>  
gcloud config set project $PROJECT_ID  

Enable the services:

gcloud services enable \
  run.googleapis.com \
  cloudbuild.googleapis.com \
  compute.googleapis.com \
  cloudresourcemanager.googleapis.com \
  iam.googleapis.com \
  container.googleapis.com \
  cloudapis.googleapis.com \
  cloudtrace.googleapis.com \
  containerregistry.googleapis.com \
  iamcredentials.googleapis.com \
  secretmanager.googleapis.com \
  firebase.googleapis.com

gcloud services enable \
  monitoring.googleapis.com \
  logging.googleapis.com \
  notebooks.googleapis.com \
  aiplatform.googleapis.com \
  storage.googleapis.com \
  datacatalog.googleapis.com \
  appengineflex.googleapis.com \
  translate.googleapis.com \
  admin.googleapis.com \
  docs.googleapis.com \
  drive.googleapis.com \
  sheets.googleapis.com \
  slides.googleapis.com \
  firestore.googleapis.com

3- In Cloud Shell, authenticate with your account and set Quota Project

From Cloud Shell, execute the following commands:

  • Set your project id. Replace <CHANGE TO YOUR PROJECT ID> with your project ID.
export PROJECT_ID=<CHANGE TO YOUR PROJECT ID>
  • Follow the instructions in your Shell to authenticate with the same user that has EDITOR/OWNER rights to this project.

gcloud auth application-default login

  • Set the Quota Project

gcloud auth application-default $PROJECT_ID

4- Clone the Gen AI for Marketing repository

From Cloud Shell, execute the following command:

git clone https://github.com/GoogleCloudPlatform/genai-for-marketing

5- Update the configuration with information of your project

Open the configuration file and include your project id (line 16) and location (line 17).

6- Prepare BigQuery and Dataplex

From Cloud Shell, navigate to /installation_scripts, install the python packages and execute the following script.
Make sure you have set the environmental variables PROJECT_ID and LOCATION.

cd ./genai-for-marketing/installation_scripts
pip3 install -r requirements.txt

Run the python script to create the BigQuery dataset and the DataCatalog TagTemplate.

python3 1_env_setup_script.py

7- Create an Vertex AI Search engine for a public website

Follow the steps below to create a search engine for a website using Vertex AI Search.

  • Make sure the Vertex AI Search APIs are enabled here and you activated Vertex AI Search here.
  • Create and preview the website search engine as described here and here.

After you finished creating the Vertex AI Search datastore, navigate back to the Apps page and copy the ID of the datastore you just created.
Example:
Vertex AI Search ID

Open the configuration file - line 33 and include the datastore ID. Don't forget to save the configuration file.

Important: Alternatively, you can create a search engine for structure or unstructured data.

8- Add your Looker Dashboards

In order to render your Looker Dashboards in the Marketing Insights and Campaing Performance pages, you need to update a HTML file with links to them.

  1. Open this HTML file - lines 18 and 28 and include links to the Looker dashboards for Marketing Insights. Example:
  • Add a new line after line 18 (or replace line 18) and include the title and ID of your Looker Dashboard.
  • For each dashboard id/title you included the step above, include a link to it at the end of this file.

The allow_login_screen=true in the URL will open the authentication page from Looker to secure the access to your account.

  1. Open this HTML file - lines 27 and 37 and include links to the Looker dashboards for Campaign Performance.

[Optional] If you have your Google Ads and Google Analytics 4 accounts in production, you can deploy the Marketing Analytics Jumpstart solution to your project, build the Dashboards and link them to the demonstration UI.

9- Create a Generative AI DataStore Agent

Next you will create a Generative AI Agent that will assist the users to answer questions about Google Ads, etc.

  • Follow the steps described in this Documentation to build your own Datastore Agent.
    • Execute these steps in the same project you will deploy this demo.
  • Enable Dialogflow Messenger integration and copy the agent-id from the HTML code snippet provided by the platform.
    • The HTML code snippet looks like this: HTML Code
  • Open the HTML file - line 117 and replace the variable dialogFlowCxAgendId with the agent-id.

10- Workspace integration

Follow the steps below to setup the Workspace integration with this demonstration.

10.1- Create a service account and upload the content to Secret Manager

  • Create a Service Account (SA) in the same project you are deploying the demo and download the JSON API Key. This SA doesn't need any roles / permissions.
    • Follow this documentation to create the service account. Take note of the service account address; it will look like this: name-of-the-sa@my-project.iam.gserviceaccount.com.
    • Follow this documentation to download the key JSON file with the service account credentials.
  • Upload the content of this Service Account to a Secret in Google Cloud Secret Manager.

IMPORTANT: For security reasons, DON'T push this credentials to a public Github repository.

10.2- Change the DOMAIN that folders will be shared with

This demonstration will create folders under Google Drive, Google Docs documents, Google Slides presentations and Google Sheets documents.
When we create the Drive folder, we set the permission to all users under a specific domain.

  • Open config.toml - line 59 and change to the domain you want to share the folder (example: mydomain.com).
    • This is the same domain where you have Workspace set up.

Be aware that this configuration will share the folder with all the users in that domain.
If you want to change that behavior, explore different ways of sharing resources from this documentation:
https://developers.google.com/drive/api/reference/rest/v3/permissions#resource:-permission

10.3- Google Drive

  • Navigate to Google Drive and create a folder.
    • This folder will be used to host the templates and assets created in the demo.
  • Share this folder with the service account address you created in the previous step. Give "Editor" rights to the service account. The share will look like this: Share Drive
  • Take note of the folder ID. Go into the folder you created and you will be able to find the ID in the URL. The URL will look like this: Drive ID
  • Open the configuration file app_config.toml - line 39 and change to your folder ID.
  • IMPORTANT: Also share this folder with people who will be using the code.

10.4- Google Slides, Google Docs and Google Sheets

  • Copy the content of templates to this newly created folder.
  • For the Google Slides template ([template] Marketing Assets):
    • From the Google Drive folder open the file in Google Slides.
    • In Google Slides, click on File and Save as Google Slides. Take note of the Slides ID from the URL.
    • Open the configuration file app_config.toml - line 40 and change to your Slides ID.
  • For the Google Docs template ([template] Gen AI for Marketing Google Doc Template):
    • From the Google Drive folder open the file in Google Docs.
    • In Google Docs, click on File and Save as Google Docs. Take note of the Docs ID from the URL.
    • Open the configuration file app_config.toml - line 41 and change to your Docs ID.
  • For the Google Sheets template ([template] GenAI for Marketing):
    • From the Google Drive folder open the Google Sheets.
    • In Google Sheets, click in File and Save as Google Sheets. Take note of the Sheets ID from the URL.
    • Open the configuration file app_config.toml - line 42 and change to your Sheets ID.

11- Deploy the APIs to Cloud Run and Firebase Hosting

11.1- Cloud Run

cd ./genai-for-marketing/backend_apis/

  • Open the Dockerfile - line 20 and include your project id where indicated.
  • Build and deploy the Docker image to Cloud Run.

gcloud run deploy genai-marketing --source . --region us-central1 --allow-unauthenticated

  • Open the Typescript file - line 2 and include the URL to your newly created Cloud Run deployment.
    Example: https://marketing-image-tlmb7xv43q-uc.a.run.app

11.2- Firebase Hosting

Enable Firebase

  • Go to https://console.firebase.google.com/
  • Select "Add project" and enter your GCP project id. Make sure it is the same project you deployed the resources so far.
  • Add Firebase to one of your existing Google Cloud projects
  • Confirm Firebase billing plan
  • Continue and complete the configuration

11.3- Firebase Hosting app setup

After you have a Firebase project, you can register your web app with that project.

In the center of the Firebase

项目侧边栏1项目侧边栏2
推荐项目
Project Cover

豆包MarsCode

豆包 MarsCode 是一款革命性的编程助手,通过AI技术提供代码补全、单测生成、代码解释和智能问答等功能,支持100+编程语言,与主流编辑器无缝集成,显著提升开发效率和代码质量。

Project Cover

问小白

问小白是一个基于 DeepSeek R1 模型的智能对话平台,专为用户提供高效、贴心的对话体验。实时在线,支持深度思考和联网搜索。免费不限次数,帮用户写作、创作、分析和规划,各种任务随时完成!

Project Cover

白日梦AI

白日梦AI提供专注于AI视频生成的多样化功能,包括文生视频、动态画面和形象生成等,帮助用户快速上手,创造专业级内容。

Project Cover

有言AI

有言平台提供一站式AIGC视频创作解决方案,通过智能技术简化视频制作流程。无论是企业宣传还是个人分享,有言都能帮助用户快速、轻松地制作出专业级别的视频内容。

Project Cover

讯飞绘镜

讯飞绘镜是一个支持从创意到完整视频创作的智能平台,用户可以快速生成视频素材并创作独特的音乐视频和故事。平台提供多样化的主题和精选作品,帮助用户探索创意灵感。

Project Cover

讯飞文书

讯飞文书依托讯飞星火大模型,为文书写作者提供从素材筹备到稿件撰写及审稿的全程支持。通过录音智记和以稿写稿等功能,满足事务性工作的高频需求,帮助撰稿人节省精力,提高效率,优化工作与生活。

Project Cover

阿里绘蛙

绘蛙是阿里巴巴集团推出的革命性AI电商营销平台。利用尖端人工智能技术,为商家提供一键生成商品图和营销文案的服务,显著提升内容创作效率和营销效果。适用于淘宝、天猫等电商平台,让商品第一时间被种草。

Project Cover

Trae

Trae是一种自适应的集成开发环境(IDE),通过自动化和多元协作改变开发流程。利用Trae,团队能够更快速、精确地编写和部署代码,从而提高编程效率和项目交付速度。Trae具备上下文感知和代码自动完成功能,是提升开发效率的理想工具。

Project Cover

AIWritePaper论文写作

AIWritePaper论文写作是一站式AI论文写作辅助工具,简化了选题、文献检索至论文撰写的整个过程。通过简单设定,平台可快速生成高质量论文大纲和全文,配合图表、参考文献等一应俱全,同时提供开题报告和答辩PPT等增值服务,保障数据安全,有效提升写作效率和论文质量。

投诉举报邮箱: service@vectorlightyear.com
@2024 懂AI·鲁ICP备2024100362号-6·鲁公网安备37021002001498号