How to Improve Your Demand-Generation Process

By Ted Chen March 03, 2016

Demand generation—from strategy creation to marketing program management—is a primary responsibility of product marketing. So how can you improve the demand-generation process to help sell even more products? The same way you would go about improving any process: apply Plan-Do-Study-Act (PDSA):

Plan: Develop a plan with specific objectives

Do: Execute the plan while measuring performance metrics against the plan

Study: Analyze deviations, identifying causes or barriers to performance

Act: Implement and standardize on the process improvements

When you apply PDSA to demand generation, focus on measuring metrics that reflect the quality, quantity, effectiveness, ROI and timeliness of marketing deliverables. In particular, keep your eye on product revenues (or orders). This is, after all, the intended end result of demand-generation activities. And the more you tie or correlate actual product sales to your marketing programs, the more you are able to define your required demand-generation budget, create and articulate strategies, and gain management support of your strategies.

You should also systematically forecast product revenues to produce metrics for comparison against plan. This forces you to analyze and compare actual and forecasted product revenues in a periodic and timely manner. It also forces you to apply models and judgment to forecasting, making you think more thoroughly about how you can affect demand. You will detect deviations earlier, gain insights into their causes and get a jump on reacting to them.

Below are two methods for forecasting product revenues. The methods complement each other and provide different measurements to improve demand generation.

Sales Funnel Metrics

Many marketing and sales organizations use lead nurturing and CRM software to track metrics that correlate product-line revenues to demand-generation programs, such as win rates, average revenue per deal, new sales opportunities and marketing qualified leads. Since they pertain to a company’s sales funnel, let’s call them sales funnel metrics.

To begin, each quarter calculate the sales opportunities needed to support your product line’s revenue goal. Use actual results to calculate average deal size, win rate and cycle time. Then apply them to the formula shown in the next column. Working backwards, you’ll be able to model the number of new marketing qualified leads needed to generate the required sales opportunities.

New Sales Opportunities = Product Line Revenues / (Sales Win Rate % x Avg Revenue Per Deal)

in Month M        in Month N
Where: Month N is ahead of Month M by the number of months in the Average Sales Cycle

Qualified Mktg Leads per Month = New Sales Opp per Month x Sales Opp Conversion Rate

You now have the monthly sales opportunities and marketing qualified leads objectives that your demand- generation plan must support.

Then on a monthly basis, determine the actual funnel metrics for the prior month and reforecast your sales funnel metrics—sales opportunities, marketing qualified leads, win rate and cycle time—based on the actuals. Use those to generate a forecast for product-line revenues and compare it against the revenue plan.

If you see any deviations between actual sales funnel metrics and plan, perform an analysis to determine the cause and formulate a demand-generation strategy to get back on track. A forecast that shows a lower trend than plan provides an early warning. It allows you to identify problems sooner and provides a head start in creating and executing a get-well plan.

Product Forecast Metrics

Product forecasts are created using a combination of judgment and time series statistical analysis algorithms applied over historical sales to discern levels, seasonality and trends. At most companies, they are created monthly and result in product quantity and product revenue forecasts. The forecasts are usually rolled up to create revenue forecasts at the product- family level (i.e., a group of similar and related products) and further rolled up to the product-line level.

The product forecast is often developed in marketing or sales, since they are in the best position to shape, drive and therefore predict demand. They can be sliced by region (e.g., North America, Europe, Asia, rest of world), by distribution or manufacturing location, by channel, by major account, or any combination of the above.

Finance uses product forecasts not only to predict revenues, but to provide a forecast for overall gross margin based on the forecasted product quantity, net price and cost of goods sold. Product forecasts are used by supply-chain operations as input to the manufacturing or procurement plans. Marketing departments use product forecasts for product planning and to track product, product-family and product-line performance.

In addition, marketing can use product forecasts to develop and adjust strategies for demand-generation programs. More sophisticated marketing organizations track and model the revenue impact of specific demand-generation programs, such as pricing promotions, and fine-tune their forecast models based on actual outcomes.

A Plan-Do-Study-Act Example

When used in combination with forecasting, PDSA can help improve demand generation. The following example is hypothetical but closely mirrors an actual experience of mine.

Sally, a director of product marketing at a B2B company, is responsible for tracking and managing a compute server product line that consists of two product families. TowerServer is an older, more mature compute server product. ThinServer, a newer product, was recently launched by Sally’s team. ThinServer is a more competitive product that enables Sally’s company to penetrate the hot cloud-computing market. The company markets and sells to enterprise companies and service providers worldwide and has just completed the first six months of its fiscal year.

Six months into the fiscal year, Sally’s product-line revenue goal is still the same as the one published in the annual company business plan. Sally created a demand-generation plan at the beginning of the year to drive the number of marketing-qualified leads and sales opportunities required to meet the revenue plan. Her product line is tracking closely to plan.

The product marketing team is executing well on the demand-generation plan. As part of her management-by-objectives process, Sally has diligently tracked the status of her department’s demand- generation deliverables—such as website content, marketing collateral, direct mailing campaigns, advertising, etc.—using a management dashboard. On a monthly basis, her department works with the product management team to review and develop the product forecast.

In addition, Sally works closely with marketing operations to model the product-line revenues based on the sales funnel metrics. She compares the sales funnel model’s product-line revenue forecast against the one created by rolling up the product forecasts. If both forecasts point to roughly the same number, that gives Sally more confidence in her demand-generation plan.

Product management published a new product forecast at the six-month mark and Sally’s team quickly noticed that the negative gap between the product forecast and the business plan continued to widen. The last month of the quarter produced orders that were below plan, resulting in a lowering of the product-line forecast. At the same time, however, the forecast that resulted from the sales funnel metrics forecast was still on track to plan.

Based on the latest product forecast, Sally rallied her team to perform an in-depth analysis of what caused the gap and make recommendations on how to close it. The team immediately observed that the product family forecast for the older TowerServer was holding to plan. However, the new ThinServer product family’s forecast was not ramping as expected.

The team then turned its attention to the sales funnel metrics. Why did the sales funnel metrics forecast predict revenues that were aligned with plan, but the product sales forecast did not? After analyzing the actual sales funnel metrics, including the latest win rates and sales opportunities, the team determined that the win rate had declined significantly compared with the expected win rate used in the forecast model.

They analyzed the last quarter’s losses and wins and found that the sales loss rate for ThinServer was significantly higher than for the mature TowerServer product. Although ThinServer was a newer, more comprehensive product, it was being used to penetrate the new cloud space, which was not only hot but also more competitive. TowerServer, on the other hand, was sold mainly to existing repeat customers, so its sales win ratio was naturally higher.

Sally confirmed her team’s hypothesis with a few regional sales directors. In response, the team decided to focus on increasing the win ratio for ThinServer by creating more effective materials, including more competitive analyses for sales, specifically aimed at cloud computing. In addition, the team agreed it needed to generate more marketing-qualified leads for ThinServer than previously expected.

Based on her thorough analysis, Sally convinced the vice president of marketing to allocate more money for demand-generation programs for cloud computing. She created a simple statistical model that quantified the number of additional total leads needed to generate the required marketing qualified leads and sales opportunities. Several additional mailings and a traveling mini trade show were added to the marketing calendar.

Also, Sally and the vice president of sales set a monthly goal for the win ratio and tracked it. Not only was it important to have an accurate win number for her forecast model, it served as an additional metric to track her demand generation’s effectiveness in helping sales close deals.

Sally’s story is a good example of PDSA in action and demonstrates some of the benefits of forecasting sales funnel metrics and improved product performance. In this instance, tracking and forecasting product sales helped Sally make faster decisions on allocation and improvement of demand generation programs at the product-family level.

It’s important to remember that when organizations don’t track, analyze and forecast their product sales, they are slower to react to deviations and lose out to competitors. Combining the forecasting process with PDSA will help you think more thoroughly about how to affect demand, detect deviations earlier, gain insights into their causes and get a jump on reacting to them.

The Benefits of Forecasting

Short-Term Demand Shaping
Model the impact of past product promotions, such as pricing promotions, to more successfully predict the revenue and profitability impact of promotions and determine which promotions are worthwhile. Companies can use this information for short-term demand shaping, like shifting demand from a product in low supply to another product through a price promotion. Dell is a master at this.

Channel Programs Improvement
Product marketing is often responsible for supporting the indirect channel marketing program, which covers value-added resellers and major distributors. Track and forecast product sales through each major distributor and reseller to gauge and improve the effectiveness of channel partners in marketing and selling specific products.

Regional Programs Improvement
Review regional sales funnel metrics and regional product forecast metrics against plan in order to react faster and reallocate or improve demand generation in lagging regions. This becomes especially important when companies rely on international expansion for much of their growth.

Annual or Quarterly Business Planning
In many organizations, product management and product marketing teams are expected to create product-line forecasts for the annual business plan. Once created, the product forecast plan also serves as an important document for creating the marketing plan. Having a product forecast process in place makes it easier to keep plans up to date.

Ted Chen

Ted Chen

Ted Chen is president of Demand Analytix LLC,, which helps companies develop product demand shaping strategies to achieve their financial objectives. The company fulfills its mission through its Business Analytics SaaS and professional services. Before founding Demand Analytix, Ted led marketing, product marketing, product management and engineering teams for computing and storage systems, virtualization and management software at HP, EMC, Cisco and multiple small and medium-size companies including startups. Contact Ted at

Looking for the latest in product and data science? Get our articles, webinars and podcasts.