Forecasting - A Product Management Task

The role of Product Management brings many responsibilities, some great, some awful, and many that are just “there”. One of the “there” responsibilities is forecasting. Being at the center of the organization, Product Management is the logical focal point for this activity, even though it requires input from groups outside their control. i.e. Sales and marketing.

One of the challenges is the method of forecasting. Different groups have different opinions and assumptions as well as differing models of consumption of said forecast.


Clearly, as the group that has to “produce” whatever it is you “ship“, even if it is a SaaS product, they need to ensure enough capacity, and bandwidth, Operations, or Ops, needs an accurate picture of the forecast, and as important the when orders or demand comes in.

In the hardware world, they need to manage the supply chain, inventory, WIP, and labor to optimize their budget. Thus they care how linear orders are, or that you get the total demand close to accurate, so they don’t fail to deliver (demand outstrips supply), or build too much inventory to cross a quarter boundary (demand doesn’t materialize).


[mks_pullquote align=”right” width=”300″ size=”20″ bg_color=”#8224e3″ txt_color=”#ffffff”]Forecasting is an art, with some science and methodology behind it. Practice improves your accuracy and credibility[/mks_pullquote]Finance cares about revenue. They have either been given a revenue target to meet (top down), or have supplied their own forecast to the Executive team. Regardless, they don’t really care about new product introductions, ramp rates, funnel leakage, lead generation, or campaigns. All they know is they have to justify $xxM to the executive team, and they want you, as in Product Management, to tell them what they can expect.

If you are short of their “Plan”, expect intense pressure to adjust the numbers. Even if you don’t believe that the assumptions on volume are correct, or achievable. They will often not take no for an answer.

Or, if you push it, they will start playing the CR card, or “Cost Reductions”. This is thinly veiled blackmail that staffing will be cut to meet targets, if revenue doesn’t come in. I don’t know about you, but I get very nervous being the one who might get the blame for layoffs.


This is a love-hate relationship. Sales and Product Management virtually always fight about forecasts. Sales of course wants to set their targets low, so they can reliably blow them away and make bank. Expect them to low-ball, and to argue that the leads are poor quality, that the funnel is leaky, and that Marketing isn’t giving them enough leads to make their numbers.

Then a miracle occurs. At the end of the quarter, the orders FLY in, they blow away their (reduced) forecast, and pocket a big check.

In this case, Product Management needs to become intimate with the CRM, build their own funnel (yes, build your own, don’t rely on Sales’ grim/rosy projections), and combine with historical knowledge to build an accurate funnel. I will tell you, this isn’t fun, but after a few times of your week 1 forecast being dead nuts accurate at the end of the quarter, and getting to call Sales on their sandbagging, it can be very gratifying.


Forecasting is a fact of life, but it is difficult with all the strings tugging in different directions. However, as much pain as it is, we in Product Management need to grasp the reigns, and make sense of it. Without our involvement, not to mention ownership, there are too many diverging opinions to balance.

Author image
Product Manager, physics educated, avid cyclist, dog rescuer, guitar noodler.
San Jose, Callifornia Website
You've successfully subscribed to The Product Bistro
Great! Next, complete checkout for full access to The Product Bistro
Welcome back! You've successfully signed in.
Unable to sign you in. Please try again.
Success! Your account is fully activated, you now have access to all content.
Error! Stripe checkout failed.
Success! Your billing info is updated.
Error! Billing info update failed.