Reading a competitor's launch
A competitor's launch appears in measured data as a new item with a rising store count. What that means for you depends on where its volume comes from, and the data can answer that better than most brands attempt.
The instinct is to watch the new item. The useful analysis watches everything else.
What should be watched first?
The category total in stores that took the new item, against stores that did not. If the category grew where the item landed, the launch brought incremental volume; if it did not, the volume came from somewhere inside the category.
That single comparison separates the two scenarios that matter and it uses data a brand already has.
How do you find where the volume came from?
By comparing each existing item's velocity in stocking versus non-stocking stores over the launch period. The items that fell disproportionately in stocking stores are the ones that ceded volume.
The result is usually more specific than expectations. Launches rarely take from everybody evenly, and knowing which of your items is exposed is more actionable than knowing your brand lost share.
What about shelf space?
In small-format retail a new item almost always displaces something, and the displaced item may be yours. A store that took the competitor's item and dropped one of yours shows a distribution loss that is easy to miss in a brand-level figure.
Track store counts per item rather than per brand for exactly this reason. A brand count that is stable can hide an item gaining stores while another loses them.
How quickly can a launch be judged?
Not from the first weeks, because those blend distribution build with trial. What can be assessed early is the distribution build itself, which indicates how much retailer support the launch has and therefore how seriously to take it.
A launch with rapid, broad distribution has support behind it. One building slowly may still succeed and is a different competitive threat in the near term.
What is the wrong response?
Reacting on price before understanding the mechanism. A launch that took volume from a different segment does not call for a price response on your item, and a price cut made on a share movement whose source was never established is expensive and frequently ineffective.
Establishing the source takes days and it is available before any response is committed.
What should be prepared in advance?
A standing analysis that runs on any new item above a size threshold in your categories, producing the stocking-versus-non-stocking comparison automatically. Launches are not rare and building the analysis each time delays the response.
What should be reported internally?
The distribution build, the category effect, the specific items affected, and the stores where displacement occurred. Four outputs, and together they define the response rather than just describing the event.
Frequently asked questions
How do I identify a competitor launch in the data?
By new item codes appearing with a rising store count, above a threshold that filters out noise from code changes and re-listings. A monitoring rule that flags them automatically is straightforward and it is usually faster than the field intelligence.
Is a competitor's rapid build always bad news?
It indicates support, which matters for how quickly the threat materialises. It also means the launch is committed and visible, which makes the source of its volume easier to measure than a slow build spread across months.
Should I respond in the stores that took it?
That is where the effect is, so it is where the response belongs if one is warranted. A national response to a partial distribution launch spends money in stores where nothing happened.
What if the launch grew the category?
Then it brought new volume in, which may still cost you share while leaving your absolute sales intact. Whether that matters depends on whether you are managed on share or on volume, and the honest report shows both.
How long should the monitoring window be?
Long enough for the competitor's distribution to stabilise and for trial to decay, which is the same standard applied to your own launches. Judging their launch faster than you would judge your own is a common asymmetry.
Does this work for a relaunch or a reformulation?
Yes, with the caveat that a relaunch may carry the same code and therefore not appear as a new item. That is a case where field intelligence is needed to know when to start looking.
Can I tell whether it is working for them?
You can see velocity per stocking store, which is the same measure you would use on your own launch. What you cannot see is their margin or their internal threshold.
What about their promotional support?
You can infer price activity from observed prices and not display support, which is the usual limitation. Treat an unexplained lift as unexplained rather than attributing it.
What is the first output to build?
Category volume in stocking versus non-stocking stores. It is one comparison and it determines everything about the appropriate response. Channel context is collected by NACS.
How much should be invested in monitoring competitors?
Enough to detect a material launch within weeks, which is an automated flag rather than a research programme. Beyond that, effort is better spent understanding your own items' exposure than tracking every competitor movement.
### What should a brand actually do with this?
Run the exposure analysis before a launch happens rather than after. Knowing which of your items sits closest to a likely entrant tells you where to defend, and it is a calmer analysis conducted outside a live competitive event.
Then agree in advance what would trigger a response and what the response would be. A pre-agreed playbook produces a faster and cheaper reaction than a discussion held while share is moving.
Interested in NRS Insights data? See case studies or contact NRS Insights.