Compare the options
Understand the options to find the best fit for you.
Build your own agents
Nearly every organization Pentaria works with has AI expertise in-house. We call this group the “AI vanguard.” These individuals and teams understand the intricacies of skills vs. agents vs. gems and can give a detailed explainer on MCP servers. They love to experiment with the latest technologies and are early adopters.
Often, the AI vanguard have created their own libraries of prompts and agents, and operate at the highest level of AI maturity within the organization. If you are a member of this AI vanguard, we want to work with you to create the change your organization needs. Your valuable insights, the agents you have already created, and your deep knowledge inform how we create change together.
AI embedded in repeatable workflows. Builds automations and agents for others.
Frequent users, one-off tasks. No repeatable workflows yet.
Not yet equipped to get value out of AI.
Source: Section, The AI Proficiency Report 2026. n=5,026 US knowledge workers at enterprises of 1,000+ employees; proficiency scored on usage, knowledge and hands-on skill tests rather than self-rating. Published July 2026.
The AI vanguard represents about 5% of employees. Unfortunately, the tools developed by the AI vanguard have usually not been broadly adopted by the rest of the teams and individuals across the organization. This is not their fault. The underlying issue is that arming teams with better tools, such as AI agents, is only about 10% of the work to create adoption, change, and business impact.
If providing libraries of AI agents worked to change behaviors, the 634 GitHub repositories of product management and product marketing skills and agents would not have left 94% of employees in the AI explorer or curious categories.
Our experience tells us that 90% of the work required to create change is onboarding, training, and showing people how, where, and when to use the tools appropriately in their work.
Pentaria combines three key disciplines required to drive impact and adoption:
- —Deep expertise in product management and product marketing, with over 100 years of experience informing the work
- —Strong AI knowledge, having helped product teams implement and use AI at hundreds of organizations
- —Decades of experience in training and coaching, helping professionals learn and adopt new tools, frameworks, and techniques
By combining these three disciplines, Pentaria provides more than just agents — we provide a platform to change how the team works.
For leaders who are satisfied with the adoption and business impact of the AI tools your team is using today, you should continue to use them.
“Your business is not creating AI agents for product management and product marketing; it is running your business.”
At Pentaria, we are devoted to creating AI-enabled product management and product marketing workflows and agents designed and tested to work together, provide the highest quality outputs, and drive adoption and business impact.
If you are interested in learning more, book a demo with our team to show you how it can work for you.
Software
Pentaria augments, rather than replaces, the software you use in your product lifecycle. Many of our clients use tools such as Aha!, Jira, Confluence, and more. Pentaria agents make a wonderful partner to the software you already use today. Our agents will take the outputs of your software and provide them with inputs where needed.
Pentaria’s platform is designed to amplify the effects of these software tools by ensuring that the work which happens throughout the product lifecycle before and after them is performed at the highest level.
For example: A backlog management tool does not test if the requirements listed in its backlog are the result of researched problems in the market, or if a business case was completed. A roadmapping tool does not tell you if positioning has been completed for approved roadmap projects.
Software is largely deterministic, meaning it expects a fixed input and provides a fixed output. Pentaria agents are designed to be flexible, adapting to your work. For example, if you are performing competitive analysis, you can feed the Pentaria agent a website, a Word document, a PDF, a slide deck and a set of your own notes and the agent will perform deep analysis and ask the right questions to produce the appropriate output.
Last, most software for product teams is only focused on a small part of the product lifecycle, such as roadmapping or requirements tracking. Pentaria agents cover the entire product lifecycle and assist from ideation through sunset.
Interested in learning more about how your existing technology stack and Pentaria can work together? Book a demo and we can explore with you.
Training
The Pentaria team comes from a training background. Training is a powerful technique to instill an overarching philosophy, such as: “Be market-driven.”
We believe that training will always be needed for product managers and product marketers. Many of the clients we work with combine both approaches: training to create the understanding and philosophy, and Pentaria to drive real, durable change and implementation.
Where standalone training struggles is retention and implementation.
Retention of Lessons Learned in Training
Experimental psychologist Hermann Ebbinghaus documented training’s retention issues as far back as 1880. His research shows that people forget most of what they learned in training within a few hours or days, and has been validated again and again.
Source: Hermann Ebbinghaus, forgetting curve; replicated by Murre & Dros, Replication and Analysis of Ebbinghaus’ Forgetting Curve, PLOS ONE, 2015.
Implementation of Training and Business Impacts
Even more concerning than retention is what led our team to start Pentaria after training tens of thousands of product professionals: only a small percentage are able to successfully implement what they learned.
The Kirkpatrick Model is the standard framework for evaluating training. Donald Kirkpatrick introduced it in a series of articles in 1959 and expanded it in his 1994 book Evaluating Training Programs. It asks four questions to evaluate the impact of training, each with increasing difficulty:
| Level | Question it asks | Typical measure | EXAMPLE OF TYPICAL RESULTS |
|---|---|---|---|
| LEVEL 1Reaction | Did participants like it and find it relevant? | Survey immediately after the class | “4.7/5, would recommend” |
| LEVEL 2Learning | Did they gain the knowledge or skills? | Quizzes, tests, skill demonstrations | Passed a certification exam |
| LEVEL 3Behavior | Do they do things differently on the job? | Manager observation, work-product review, follow-up surveys months later | Are they actually writing problem statements and doing win/loss interviews? |
| LEVEL 4Results | Did the business outcomes change? | Business metrics tied to the behavior | Faster launches, higher win rates, fewer failed features |
Most trainings stop measuring at Level 2. Unfortunately, a product manager or product marketer giving a training course a high Net Promoter Score or celebrating a certification on LinkedIn does not create behavioral change or business impacts.
Other researchers suggest that the implementation problem may be even larger:
“Less than 20% of trainees take what they learned in training back to the job and use it to achieve worthwhile results.”
The team at Pentaria saw this gap during their careers and the frustration that teams had in taking the tools they learned in training to the implementation phase.
Pentaria was created to address this problem. Instead of learning that is forgotten or never implemented, the expertise sits inside the daily work.
The team can focus less on learning the work and more on doing the work with their own experience and judgment. Because the deep expertise of the Pentaria team is built into the workflows and agents, you know that your team is always following best practices, and the outputs they create raise the floor and the ceiling for quality.
Want to integrate what your team learned in training with AI workflows that put the methods into the work itself? Book a demo with our team and learn how.
Do nothing
Not everyone is an early adopter for AI. Depending on your industry, product, culture, team, regulatory environment, or other factors you could be in the 21% we call “AI curious.”
AI embedded in repeatable workflows. Builds automations and agents for others.
Frequent users, one-off tasks. No repeatable workflows yet.
Not yet equipped to get value out of AI.
Source: Section, The AI Proficiency Report 2026. n=5,026 US knowledge workers at enterprises of 1,000+ employees; proficiency scored on usage, knowledge and hands-on skill tests rather than self-rating. Published July 2026.
If you are happy with your results, and you perceive low disruption in your space, Pentaria may not be the right fit for you. We recommend you stay as-is and come to us when you are ready to change how your organization works.
However, if you want to take advantage of this new technology, Pentaria can help you step into the AI world with low risk. Our workflows put guardrails on how the team uses AI, so you know which outputs to expect throughout the process.
If you’re curious but would like to explore further, book a demo with us and we will walk you through an implementation with no obligation to go further.