Blog Planview

Votre parcours vers l’agilité métier

Gestion du portefeuille de produits, Gestion de portefeuille de projets, Planification stratégique

7 Views of Project-Portfolio Variance

Publié le By Dr. Richard Sonnenblick

Knowing the current cost and estimated value of every initiative in your portfolio is nice, but when executives meet for a review, what they really want to know is: What’s changed since the last review, was it good or bad, and why?

The above is true whether a month, a quarter, or a year has passed since the last review; looking at what has changed is one of the fastest ways to communicate portfolio health and progress towards strategic goals. It also builds trust in the portfolio process: executives remember where you left off with them, and starting from that same point establishes a feeling of continuity and stability.

In this post, I’ll share different methods for communicating variance. These methods have different use cases, but they all riff on the idea of variance. My examples will come from our valuation-based portfolio management solution, Planview Advisor, Planview Advisor but you could build similar views by hand in Excel or other self-service BI tools.

Views 1 and 2: Operational Variance with Stoplights and Gantt Charts

Often, the least complex view is the best. When reporting on operational budget and schedule impacts, the simplest way to communicate variance can be a stoplight or heat map. In the figure below, we use a simple red-green color scheme on both financial and schedule variance columns to make troubled projects easier to spot and get the attention they need.

 The schedule and financial status columns provide clear and simple variance information. In the Enrich Analytics Platform, users click on a project's row to drill down to project details.

The schedule and financial status columns provide clear and simple variance information. In Planview Advisor, users click on a project’s row to drill down to project details.

Rather than overload the table with more detailed information on the budget overrun or the schedule slippage, we make the additional information available via drill downs on each project row. In some cases, it is useful to add a column on schedule slippage, expressed in days or weeks, and a column on budget shortfalls, in dollars. But at the top-line portfolio level, it is often distracting.

If you are looking for even more detail, schedule shifts across a set of stages can be viewed by comparing old and new Gantt charts. In the picture below, the current schedule for a specific project is compared with the original, baselined schedule.

 
ops_gantt_detail2

Each row of the Gantt chart recalls a different forecast for an initiative’s critical phases and milestones.

View 3: Metric Variance

To communicate variance of a project metric such as net present value, peak revenue, or total cost savings, use a time-series plot. This is especially helpful when you have multiple values for each metric, illustrating an evolution over time. (For the manufacturing-oriented among us, these will remind you of Deming’s control charts.) In the figure below, two key metrics for a project are plotted over the last several years. The upper chart indicates that project value (as represented by risk-adjusted net present value, here labeled eNPV) has increased as the project successfully navigated its life cycle, while remaining project costs decline.

 
 Charts like these help managers observe changes in a project's key metrics over time. If the market size or market share has been revised since the last business case review, you'll see that here in spades.

Charts like these help managers observe changes in a project’s key metrics over time. If key assumptions have been revised since the last business case review, you’ll see the impacts here in spades.

View 4: Portfolio Metric Variance

If we want to see a key metric, say, eNPV, across the entire portfolio, a table view, color-coded with a databar for positive and negative variance is a good choice. In the table below, we’ve included eNPV variance as a column next to eNPV, making it easy to see which projects had a change to eNPV, and whether that change is important. The subtotal rows also make it easy to see the total variance by division (in this portfolio of drug candidates, the acronym ‘TA’ stands in for a division).
 
port_key_metrics_variance

The eNPV variance column makes it easy to identify the biggest movers in forecast value since the last review took place. You can also see which divisions experienced the greatest gain or loss of value.

View 5: Vector Variance

Some important metrics, like revenue or cost by year, are not single-valued. For these metrics, use a line chart to compare multiple forecasts. In the figure below, costs and revenues over time are compared across a project’s quarterly checkpoints. What was hoped to be a 500M /year opportunity a few months ago now appears to suggest less than 300M/year.

 
vector_variance2

These charts compare revenue and cost over time forecasts and make it easy to discern, for example, whether a project’s short-, medium-, or long-term revenue prospects have changed.

View 6: Portfolio Vector Variance

To compare historical forecasts over time for an entire division or other group of projects, a line chart like the one above is your first stop. If you want to break out changes for each project contributing to the variance, switch to a bar chart that shows the absolute changes for each project. Since some projects might have a positive variance while others will have a negative variance, you’ll need to plot bars both above and below the origin. In the picture below, we see that two different projects contribute to the Endocrine Disorder division’s change in costs and revenues: Eaglogen and Resdexel.

 
port_vector_variance

When looking at a division or corporate portfolio, use a stacked bar graph to illustrate the specific projects with changes in cost or revenue.

View 7: Waterfall Charts tell a Portfolio’s Story

The final view we’ll consider is a waterfall chart with pillars designed to highlight changes in key metrics since the last review. In the picture below, last year’s portfolio is depicted on the far left, with a value of around 6B. The current portfolio is on the far right with a value of around 9B. The difference in value is explained in the columns in between: a couple of projects had been dropped or reduced in value, but many more have increased in value, and a single project was added to the Dermatology portfolio.

 
eNPV_waterfall

Waterfall views like this summarize a portfolio’s change in value since the last portfolio review. After a moment’s perusal, you’ll know the projects added and removed, as well as those increasing and decreasing in value. Coloring and sorting by division, as done here, makes it easy to see division contributions to value accretion.

Often, executives will oscillate between views like this waterfall, which provides a top-level view on what has changed in a portfolio, and detailed project-specific views on the variance of costs, benefits, and schedules. In Planview Advisor, you can explore the reason for variance by clicking on the blocks in the waterfall, bringing up views on the variance of each individual initiative, much like the ones we’ve just reviewed.

Portfolio Management is a Journey

Not everyone who practices portfolio management regularly produces variance views. Without the help of a solution like the Planview Advisor, it can be a truly arduous task to even calculate project and portfolio variance. Nevertheless, I hope you’ve found these views thought-provoking. Perhaps there is one small, yet important, aspect of variance analysis that you can incorporate in your portfolio reviews: How recently each project has been modified or updated by the project team. That one piece of information might be an entry-point to a worthwhile conversation about each asset.

Each year, tens of billions of dollars in investments are shaped, allocated, and refined with the help of Planview Advisor. Are you interested in learning how we can help streamline your portfolio reviews, turning months of long nights and frustration into a value-enhancing, confidence-affirming exercise for your organization? Contact us, and learn what our clients already know about the value of our flexible and intuitive solution for product portfolio management.

 

Articles similaires

Rédaction du contenu Dr. Richard Sonnenblick Chief Data Scientist

Chief Data Scientist de Planview, Richard Sonnenblick possède une solide expérience acquise auprès d'organisations majeures des secteurs pharmaceutiques et des sciences de la vie. Fort de son expertise, il a développé d'excellents processus de priorisation et de revue de portefeuilles, systèmes de scoring, et méthodes d'évaluation et de prévision financières pour améliorer à la fois les pronostics produits et l'analyse de portefeuilles. Richard Sonnenblick est titulaire d'un doctorat et d'un master en ingénierie et politiques publiques de l'université Carnegie Mellon, et d'une licence en physique de l'université de Californie à Santa Cruz.