Creating Your Initial Roadmap

If we help you create your initial product roadmap, at the end of the process, you’ll have a functional product roadmap: a list of projects with target release dates, selected to maximize business impact and minimize execution uncertainty. You’ll have an Excel-based tool for maintaining your roadmap.  More importantly, you will be ready to maintain it on your own with relatively little effort, integrated with your existing development process.

The entire process takes 2-6 weeks, depending on how many projects you have and how much concentrated time you spend on it.

Here’s what a typical engagement looks like, divided into phases:

Click on the tabs below to see more information about each phase.

This is where we learn about your business and the way you develop software. We sit down with your business and technology owners and ask a series of questions that give us an idea of what you are trying to do, who you do it for, and how you do it.

We also examine materials you use for employees, customers, and partners to tell them about yourself and what’s unique about your company and your products.

This phase can usually be completed in one day, although more time may be necessary for larger companies. Although it can be done remotely, it’s best performed on site.

Of course, it’s all done under NDA.

Deliverables: A report summarizing the findings and an outline of next steps.

Initial Assessment Goals

  • Understand your business: current vision, mission, business strategy, customers
  • Understand your process: product planning/approval, product management, development, release
  • Identify preliminary “Uncertainty Areas” based on historical projects
  • Create straw-man “Product Related Business Goals”
  • Calibrate estimated cost buckets based on historical projects

Methodology

  • Interview the business owners and drivers of the software process (depending on the company, CEO, business unit VP/GM, CTO/VP/Directors Engineering, CPO/VP Product, product managers)
  • Examine product requirements, planning tools/materials, business presentations (as needed and as appropriate)

This is where we construct the list of candidate projects (what projects are called before they are approved) by “mining” your organization: all the people in your company who “generate demand” for engineering resources. At most companies, product managers generate most of the projects. The VP Engineering/CTO always has some. The CEO, CMO/VP Marketing or general manager/managing director of the company or business unit usually does too. If you have a sales function, they often are a source of candidate projects based on customer feedback.

The list must include “pure engineering” projects too. Why?  At many companies, engineering projects are not discussed broadly outside the engineering team because a lot of people don’t understand the technical details. Yet these projects compete with customer-facing features for engineering resources. As a result, many companies have “two sets of books” for their software projects, which means there is no transparency in the prioritization process. For that reason, just as you only have one set of books for your company’s financials, there should only be one list of projects per engineering team. Ultimately, it will represent the complete set of work that your engineering team is asked to perform over the next several quarters.

This phase generally takes about two days for a list of 100-200 candidate projects. At this point in the process, it’s OK if there’s very little detail on exactly what the project really is; in fact the term “project” is a little awkward to use for what is often very vague. (In practice we use the term “feature” or “capability” because for things that are not very well understood, it’s easier to say “we need a way to do Y” than “we need X which will let us do Y.”)

This phase can be done on-site or remotely.

Deliverables: Rough draft of the project roadmap pipeline: all projects, in all phases, from “concept” to ones that are actively under development. Most projects will be assigned to business objectives, grouped into categories, and many duplicate and overlapping projects will be eliminated/combined. Weekly status reports as needed.

“Mining” Goals

  • All raw data for the product roadmap pipeline assembled for all projects/features the company might  deliver over the next 6-12 months

Methodology

  • Assemble and refine product-related business objectives
  • Interview product managers, CTO/VP Engineering, and business owners such as CEO, GM or MD, VP Marketing, VP Sales, etc.
  • Initial classification of every project relative to business objectives and categories

Tools

This phase is the heart of the Roadmap Integrity Process. We walk through all the projects and assess every project relative to three factors:

  • Estimated business impact against the business objectives identified and refined during the previous two  phases
    • Minor – few customers or stakeholders will notice
    • Noticed – customers or stakeholders will definitely notice
    • Significant – this project will make significant progress towards reaching a business objective
    • Game changing – relative to our current state or competitors, this could change the rules of the game
  • Estimated project uncertainty, the “secret sauce” of the Roadmap Integrity Process to help ensure successful execution
    • Low - we’ve done things like this before and don’t expect this to be significantly different
    • Medium - we’ve done things that are kind of similar, but there are significant differences
    • High - there are a large number of unknowns and open issues including ones we haven’t thought of yet
  • Estimated engineering cost, using the customized “cost buckets” identified in the Initial Assessment phase

This phase can usually be completed in 1.5 to 3 days, either all at once or over a longer timeframe. It can be done remotely or on-site, although some on-site interaction is recommended.

Deliverables: A project roadmap pipeline with all projects assigned estimated business impact and execution uncertainty. Excel pivot tables are available for preliminary analysis. Weekly status reports as needed.

“Refining” Goals

  • Estimated business impact for all candidate projects
  • Estimated uncertainty for all candidate projects
  • Estimated engineering cost for all candidate projects (Minor, Medium, Major, Epic)
  • Business objectives and cost “buckets” refined as needed

Methodology

  • Go through all candidate projects with senior product leaders and estimate business impact, uncertainty, and cost
  • Depending on the company, this could be with product managers and engineering leads. At a startup, it’s might be with the CEO, CTO, and VP/Director of Product

Tools

At this point, you have a roughly prioritized list of projects. To transform the project list into a roadmap, however, you need to map candidate projects against the calendar. That’s what happens during this phase.

Many companies have an existing “release calendar” which defines how often significantly-enhanced versions of the product(s) are made available to customers. If one exists, it can usually be used. If not, we will create one that works for you, ideally one that has the concept of “major” and “minor” releases so that there can be a small number of “dominant themes” for major releases.

Essentially, this phase consists of “slotting” candidate projects into desired releases in the order of descending business impact, accounting for uncertainty. This phase will require multiple iterations or “passes” — the first pass, the first pass is a “wish list” because there will be be no way to deliver everything in the desired timeframe. Projects that are not slotted against releases are moved to the “backlog” for future use.

After the first pass is complete, we involve engineering to move things around on the calendar and/or brainstorm ways to reduce uncertainty and/or cost. It may be appropriate to discuss procuring additional resources or load balance existing resources at this point too. Projects may be divided into multiple phases, and usually the description is refined to more accurately reflect what work can actually be completed in the desired timeframe. More projects are moved to the backlog at this point too.

After a few iterations/passes, you’ll have a set of projects to plan against. If desired, “uncertainty milestones” can be identified for bigger projects indicating when certain milestones must be complete if the target release is to be kept: product requirements complete, engineering design complete, etc. Not every project needs that additional overhead, but it can be useful for the bigger and more complex ones.

This phase can usually be completed in 1 to 3 days, although it may take longer based on the number of iterations/passes. This phase can be performed all at once or over a longer timeframe. It can be done remotely or on-site, although some on-site interaction is recommended.

Deliverables: All desired projects slotted against the calendar such that initial planning can proceed. Weekly status reports as needed.

Release Targeting and Integration Goals

  • Target dates for all projects you want to pursue over the next 9-12 months
  • All other projects moved to the backlog for future use/reference
  • Major releases with dominant themes (optional, but strongly recommended)

Methodology

  • (If needed) Define release timeframes (e.g. “Major” and “Minor” releases)
  • Slot desired candidate projects against the calendar with senior product and engineering leaders
  • Move remaining projects to the backlog
  • Iterate project slotting and scope/uncertainty until workload/resources/scope are roughly in balance
  • Depending on the company, this could be with product managers and engineering leads. At a startup, it’s might be with the CEO, CTO, and VP/Director of Product

Tools

At this point, you’re ready to “launch” the process internally by going through the set of releases and projects and discussing the rationale and process behind them. This is usually done with a “kickoff meeting” with product managers and senior engineering staff. This meeting will be run by you (usually a VP or one or more Director of Product), but facilitated by Roadmap Integrity. The entire process is designed to be 100% transparent, so it’s important that people understand the process enough to know how project prioritization and scoping decisions were made.

Separate launch meetings with business owners and/or senior management may also be needed. In general they will be less interested in the specific details of how the projects will be defined and implemented and are more interested in the timing and the scope of the projects.

Although it takes some effort to complete the initial version, maintaining it takes relatively little effort. Generally, one person is put in charge of the following activities:

  • Regularly scheduled meetings where current and upcoming projects are discussed in terms of status and the current level of uncertainty. Since projects should not proceed to implementation phase with a high level of uncertainty, this is is the most important thing to monitor. There’s no one right way to do this; at a high level the question you need to ask for each project is, “if we want to release this by date X, are we on track?”  If yes, great; if not, then mitigating steps must be taken. (In that context, the mitigating steps are usually fairly obvious.)
  • Another meeting with a smaller number of people where new projects are considered. For each of these projects, uncertainty, business impact, and cost are assessed. Target releases for projects can be chosen at that point, or perhaps you’ll want to have a less-frequent “release planning” meeting where this is done, as well as fine-tuning of the business objectives (if needed).
  • During the annual planning cycle, the business objectives are refined to align with the overall company business objectives.

Roadmap Integrity will train the person responsible for these activities so that they can run them on their own and maintain the spreadsheet tool. Your company may already have processes in place to do this type of tracking; if so, we will work with you to integrate these activities into your existing project management process.

Deliverables: Completed (initial) project roadmap and the knowledge to run the Roadmap Integrity Process on your own and to maintain and enhance the project roadmap pipeline.

Longer-term coaching services for subsequent maintenance are available on request.

Release Targeting and Integration Goals

  • Ensure all stakeholders on the team understand the process, the decisions that have been made, and specific project/release plans
  • Ensure at least one person in your company is trained to maintain and enhance the project roadmap pipeline spreadsheet
  • Ensure a process exists to keep projects on track

Methodology

  • Create a presentation with an overview of the work that’s been done, and conduct that presentation with the team
  • Train the person or people in your company to maintain the project roadmap pipeline

Tools

Leave a Comment