AI and Go-To-Market Functions: Leverage Today’s Tech to Save Time and Money

AI advancements move rapidly, with “shiny new objects” and billion-dollar investments making headlines daily. But for every exciting new AI video generator promising to upend Hollywood filmmaking, tried-and-true AI technology of this decade quietly augments – or even fully automates – enterprise workflows, tasks, and processes. 

Ultimately, these successful implementations, which deliver a measurable return-on-investment, help keep the pace of AI investment and innovation flowing. Small to midsize businesses (SMBs),are increasingly turning to AI solutions, especially generative tools like OpenAI’s ChatGPT, to drive efficiency, improve accuracy and elevate quality. ChatGPT and similar AI assistants excel in tasks that require research, gathering data from multiple sources, summarization, and creating original content.

But ChatGPT isn't always enough for bootstrapped SMBs, marketing agencies, and digital creators who require a more customized solution to create deeply personalized and targeted assets for sales and marketing — such as hyper-personalized AI proposals that can instantly leverage, summarize, and organize extensive information with minimal human input.

However, not everyone is a coding expert and many businesses can’t budget the time and money to build out and integrate a comprehensive platform to meet their needs. They want a turnkey solution that allows them to drive the creation and take responsibility for ongoing maintenance and upkeep of their AI-powered solutions. 

Contextual: The Low-Code Customization Capabilities You Need

Contextual is an AI-orchestration platform that specializes in building workflows and processes that bridge the complexity of multiple data sources, multiple AI models, human interactions and AI-generated recommendations. As a low-code, visual editor platform, Contextual makes it possible for members of a business development organization to develop, implement, manage and maintain their AI solutions. 

AI Use Case: Creating Targeted Proposals and Marketing Strategies for Musical Artists 

Digital marketing agencies represent one example of businesses who may not have the inclination to create custom AI platforms but would vastly benefit from the ability to create a customized AI platform for targeted, personalized proposal generation. 

For instance, one digital marketing agency that specializes in bringing corporate brands together with musical artists for promotion opportunities leveraged Contextual to create an AI Proposal Generator. The tool can tailor campaigns that exactly match the brand and the musical talent. These promotions include artist’s influencer marketing campaigns, custom content, and live show activations.

The Challenge: Create Relevant, Targeted Proposals for Musical Artists Quickly 

With a database of artists across any genre you can imagine, the sales team faced significant challenges in effectively mapping a given company, influencer, product, target audience, and brand message to a potential artist. 

Add in the additional nuances of targeted geographies, appropriate campaign strategies, artist-specific expectations, and budget limitations, and finding the perfect match for a given proposal took too much time. Even worse, the tactics employed often produced inconsistent results. As they say, time kills deals, so being able to rapidly and accurately create tailored and focused proposals that hit the client’s expectations was critical.

Solution: Implement a Highly Specialized AI Proposal Generator Powered by AI 

The agency used Contextual as the command center to create a highly specialized AI Proposal Generator. The power of this solution is driven by the reality that creating a truly production-ready AI solution is not as simple as slapping a pretty face on an existing GPT or LLM chat tool. 

Successful AI solutions span multiple data sources, both from internal and external sources, and require multiple AI steps tied together with iterative logic steps and workflow management. 

These often include a combination of general GPT calls, function-specific AI calls such as data categorization or summarization, and tightly defined and controlled requests to retrieval augmented (RAG)-enhanced assistants. For the marketing agency, having an AI assistant uniquely trained on their internal data made the AI dramatically more efficient in its proposal generation abilities.

The Details: How the Digital Marketing Agency Deployed the Proposal Generator to Achieve Objectives

By creating a thoughtful design that took into consideration thedata, user experience, anticipated results and how the solution would be used by sales team members and clients, the agency’s AI solution went far beyond generalized content generation to hyper targeted and personalized proposals that drive results. 

The solution was designed with the following objectives:

• Pull from the best of a generative AI / large language model’s ability to engage in a thoughtful and context-rich dialogue with end users, be they customers or sales team members

• Vectorize previously successful proposals along with critical musical artist data so certain requests are driven to retrieval-augmented (RAG) enhanced AI assistants to optimize responses

• Ensure the system has the required guardrails to control the production, curation, and storage of data required to generate the proposal

• Integrate traditional system logic and software functions where AI isn’t required or could cause challenges, especially around calculation of budget allocation and critical proposal metrics like cost per thousand (cpm) impressions

• Thoughtfully pull from third party APIs and other AI endpoints as required to generate the final proposal results

The AI Proposal Generator solution achieves all of those objectives through the following workflow:

1. The initial user prompt encourages free-form response of any length and depth about a client’s goals. By making this an open entry point, the system feels more interactive, more intelligent and more able to handle proposal generation for a range of client maturities. This flexibility capitalizes on the freedom of dialogue-based exchange that LLMs provide.

2. Through engineered instructions, or ‘prompts’, on a proprietary OpenAI Assistant coupled with a structured data object, the AI Proposal Generator determines what information it can readily identify from user information and what information it needs to further engage to identify. This data is passed back into the back-end system using OpenAI Function calls, which are critical for controlling data consistency.

3. As information is gathered, it is “‘checked off” in the user-facing UI, which creates the sense of both interactivity, intelligence in the background and progressive movement and value.

4. Where appropriate, the AI Proposal Generator leverages third party tools like WebPilot  and RapidAPI Classify to summarize product websites, determine brand tones, check recent news about a candidate artists and validate tour dates.

5. Based on a completely captured data structure that includes information about the client’s goals, advertising methods, campaign strategies, and audience targeting a general GPT pulls a set of traits that can be matched to the company’s proprietary artist database. This combines the “secret sauce” of the business with the power of AI generated content.

6. Ultimately, a different proprietary AI Assistant translates the combined data of the campaign objectives, artist information, third party data from websites and social media, and financial calculations into a complete and client-ready proposal.

Establishing Guardrails for AI Use

Success with any AI-generated, client-facing content requires guardrails to prevent the AI Assistants from producing unexpected or unacceptable content. This type of prompt engineering is critical, and can be the difference between a pleasant AI-driven experience for clients and platforms quickly forgotten or even deleted.  The Assistants are heavily reinforced with information on the types of requests they can respond to, topics they should avoid, and how to keep the AI Proposal Generator discussion on track.

Conclusion

The pace of innovation in AI-assisted solutions is accelerating. Digital marketing agencies and other SMBs who rely on proposals to make sales and require efficient go-to-market solutions can’t sit on the sidelines while their competitors adopt AI solutions to sprint ahead. 

 Forward-looking organizations, from digital agencies like the one described above to traditional “non tech” companies across a range of industries, are already making investments to integrate AI into their existing workflows, processes, tasks, and systems. 

AI solutions are being built that impact go-to-market sales motions, service delivery tasks, customer support and back-office activities. With the evolution of tools that make building AI solutions easier than ever, the ability to rapidly trial, iterate, expand and change puts AI ROI directly in the hands of those who choose to use these tools. 

Andrew Brooks is the founder and CEO of Contextual.io. Contextual's low-code AI automation platform makes enterprise AI solutions fast to build, easy to deploy, and ready to scale. The founding team, including Brooks, consists of seasoned entrepreneurs with a history of successful exits from technology platforms such as SMBLive and SmartThings.