Key insights

• ⚠️ Identifying low close rates despite high lead volume
• 🔄 Realizing inefficiencies in the sales process
• 🚀 Expedited proposal process to minimize drop-offs and secure deals faster
• 🔍 Implemented an exploration Milestone to vet potential clients and save time
• 🔄 Transitioning clients into the exploration Milestone after a discovery call to assess their suitability and willingness to invest
• 💡 Leveraging the company's brand position as an AI Solutions company
• 📊 Addressing non-deterministic outputs and setting client expectations
• 🔬 Morningside used cosine similarity testing to evaluate the semantic similarity of text

Q&A

• How can developers benefit from leveraging metrics like cosine similarity?

  Developers can get paid for exploratory work and present findings and proposals to potential clients, demonstrating professionalism to secure contracts, avoid endless revisions, and filter out potential clients by leveraging the exploration milestone and relevant metrics like cosine similarity.

• How is cosine similarity used in the proposal process and milestone?

  Cosine similarity is used in the exploration milestone to evaluate the similarity of outputs, present findings, and build trust in proposals. It becomes part of the contract to mitigate subjective client opinions, showcasing professionalism and demonstrating the value of the AI services.

• How did Morningside tackle non-deterministic AI outputs and client feelings?

  Morningside used cosine similarity testing to evaluate the semantic similarity of text, achieving consistent and predictable results, mitigating subjective opinions, and comparing AI outputs with client expectations.

• How has the exploration milestone evolved and what benefits does it offer?

  The exploration milestone has evolved from $800 to $2000-$5000, enabling the team to weed out unqualified clients and address non-deterministic outputs. It helps in setting client expectations, solving downstream issues, and determining the client's seriousness and financial qualification.

• How did the company address spending too much time on unqualified leads?

  They implemented an exploration Milestone to vet potential clients and save time. By transitioning clients into the exploration Milestone after a discovery call, they can assess their suitability and willingness to invest, leveraging the company's brand position as an AI Solutions company.

• What was the key change that helped the agency scale past 10K per month?

  The key change was expediting the proposal process to minimize drop-offs and secure deals faster. This helped address the low close rate despite receiving 20 leads per day and identifying inefficiencies in the sales process.

• 00:00 The speaker shares a simple change that significantly improved their agency's performance, allowing them to scale past 10K per month. They realized their low close rate despite getting 20 leads per day and identified inefficiencies in their sales process. The key change was to expedite the proposal process to minimize drop-offs and secure deals faster.
• 06:19 The company was spending too much time on unqualified leads, so they implemented an exploration Milestone to vet potential clients and save time. By transitioning clients into the exploration Milestone after a discovery call, they can assess their suitability and willingness to invest, while also leveraging the company's brand position as an AI Solutions company.
• 12:07 The exploration milestone allows the team to assess if the project is feasible, set expectations with clients, and determine the client's seriousness. The milestone has evolved from $800 to $2000-$5000, enabling the team to weed out unqualified clients and address non-deterministic outputs. It also helps in setting client expectations and solving downstream issues.
• 19:52 Dealing with non-deterministic AI outputs and client feelings can be challenging for building an AI business. Morningside tackled this by using cosine similarity testing to evaluate the semantic similarity of text, which helped in achieving consistent and predictable results and mitigating subjective opinions. Cosine similarity converts words into numerical equivalence, allowing for the comparison of AI outputs with client expectations.
• 27:06 A discussion of cosine similarity in the context of using it to measure similarity between outputs, presenting findings in an exploration milestone, and using it in a proposal to build trust and pitch for a full project.
• 33:16 Developers can get paid for exploratory work and present findings and proposals to potential clients, leveraging metrics like cosine similarity to demonstrate professionalism and secure contracts, which helps avoid endless revisions and filter out potential clients.