Ask sales for the big events only: tenders, launches, campaigns, customers won or lost. Handle tenders with three likelihood buckets (full volume, half volume, none) agreed across departments. Plan launches and big new customers with best, expected and worst-case forecasts and mitigating actions decided in advance. Campaigns need only a small override and a comment.
Watch the 17-minute video this article is based on, or see the full S&OP training series.
Coach the people giving you input
Once you know who you need to talk to, the next step is to coach them. Local forecasters usually do this alongside their main job. Teach them two things:
- What information you need and what dialogue you want to have every month.
- What internal process they need to collect it consistently. They rarely hold all the information themselves; colleagues hold pieces of it.
Ask for events, not tweaks
The input you want is the events big enough to move the needle on the products you care about: your flagship products, cash cows, expensive items and upcoming launches. That means:
- Tenders
- Campaigns
- Product launches
- Potential new customers, and customers at risk of leaving
You are not looking for 5-10% tweaks across 500 items. Statistics captures those over time, and manual overrides of that kind often make accuracy worse, because people are biased one way or another. What you want is market knowledge nobody else has: which campaigns are being planned, which big customers are being courted, which tenders are being bid on.
Few things are certain in sales. That is where a good S&OP process shines: scenarios, likelihoods and mitigating actions.
Tenders: plan with three likelihood buckets
Tenders are the hardest, because you either win or you don't, and both outcomes have consequences.
Take an example: the NHS needs 5,000 hospital beds over two years and you produce 10,000 a year. About 200 extra a month is no problem. But if the tender requires the full volume in one delivery, those 5,000 beds equal half a year of demand.
- Don't plan for it and win: late-delivery penalties, a frustrated sales manager, operations running flat out to catch up. Worst case, the customer treats it as a breach of contract and re-tenders.
- Plan for it and lose: half a year of inventory, and operations that just ran flat out to build it now has to cut capacity to work it down. Worst case, operators stand idle for so long that jobs are lost.
The answer is a standard: one agreed way to judge the likelihood of a tender and decide how it enters the forecast. The local sales team gives its best estimate of the chance of winning, in percent. It isn't hard science, but nobody is better placed to make that call. Then sort each tender into one of three buckets:
- High likelihood: include the full tender volume in the forecast and build stock for it.
- Medium likelihood (for example 50-70%): include half the volume. Agree with production in advance which levers would close the gap if the tender is won: overtime, extra shifts, or whether current capacity can absorb it.
- Low likelihood (below about 50%): keep it on the radar, but put no volume in the forecast.
The exact percentage limits should be agreed across departments and leaders, because they become the foundation. Also define when a tender is big enough to be escalated through the S&OP forums for management approval. It may sound bureaucratic, but a large tender handled wrong hits the whole business, and that is when you want the organisation aligned up front rather than arguing afterwards.
New product launches: three forecasts and a plan for each
Whether a launch is a small portfolio filler or a major flagship, settle three things before it starts:
- Three forecast scenarios: the best case, the expected case, and the worst case if the market doesn't respond.
- The launch buffer: how much extra stock you need in the first months, because customers fill their shelves ahead of the launch.
- Mitigating actions if the best or worst case starts to materialise.
Launches follow a predictable pattern: an early demand spike from shelf-filling and novelty interest, then a waiting game to see which scenario plays out. The organisation is already geared to the expected case, so prepare actions for the other two.
If the best case happens
Selling more than expected is good news, as long as you can serve the demand. Know in advance:
- How much extra demand current capacity can absorb.
- Which short-term levers you can pull, and how many units they give you.
- How long it takes to add capacity permanently, such as a new production line.
Being caught out by demand far above forecast is a real risk. Novo Nordisk is a well-known example: selling everything you can make is great at first, but running operations flat out for long periods burns people out, mistakes creep in, the culture shifts towards output over people, and competitors take note that there is a big market to chase.
If the worst case happens
Customers have stocked up and have no outlet for the product, so operations has to slow down or pause production. Prepare for it with production up front: can operators be cross-trained to work elsewhere, and is there a point in sourcing or sub-assembly where materials stay neutral and can be used for other products?
It isn't the S&OP team's job to solve everything. It is their job to make sure everyone understands the scenarios and has the best possible chance to prepare.
New and lost customers
Handle these like tenders: a likelihood framework with mitigating actions at each step. A big new customer behaves much like a launch, with an initial stocking phase followed by a steadier run rate you have to monitor. How much effort to spend depends on the size of the customer: little for a small one, a lot for one that can move your operations.
Campaigns: the easy one
Campaigns usually bring a small bump in demand followed by a small dip when they end. A small manual override with a comment explaining it is enough.
Why this structure pays off
Repeatable standards mean nobody reinvents the wheel each time a tender or launch comes along. Everyone knows the plan, and the monthly dialogue serves two goals:
- Sales can serve customers consistently and reliably.
- Operations gets as much stability as possible, so it can produce cheaply while keeping inventory low.
In short: serve your customers today and a year from now without breaking the bank. Curious what missed deliveries cost you today? Try the stockout cost calculator.
Want to see this implemented for your team?
OmniaOps is a supply chain planning platform built by ops practitioners. 30-minute call, no slide deck.
Book a 30-min call