Measures how fast a product is moving in the stores where it is in distribution. Use only within a market or retailer, when comparing products with different distribution levels. Used most often by Sales (in item ranking reports). Calculated by dividing some measure of sales (e.g. Units or Dollars) by some measure of distribution (e.g. % ACV Distribution).
Here’s a post comparing different measures of velocity, including SPPD:
Hi guys,
Can you help me understand this hypothetical example?
Value Evolution: +12%
Weighted Distribution: +32%
ROS: -15%
Av Price: +5%
Volume Evolution: +7%
In this example, given that price is growing and value and volume are positive, how would you explain the decline in ROS? By this I mean, what factors would you predict to be important, assuming category growth is flat?
First I’ll explain some terms in case some [U.S.] readers are unfamiliar with your terminology.
value evolution = % change vs. year ago in dollar sales (or whatever the currency is)
volume evolution = % change vs. year ago of physical volume, could be in cases, pounds, gallons, kg, liters, etc.
ROS = rate of sales = velocity
Here is what we can say, given the data you have shared:
Revenue is up +12%, due (almost evenly) to pricing (+5%) and true volume growth (+7%). Since volume is up even though the price is up, other things are more than compensating for the higher price to drive volume growth. We know that distribution is up a lot but there could also be more advertising, more/better trade merchandising, a competitor has been doing a lot worse (even higher price increase, losing distribution, less merchandising, less advertising, etc.).
It is very common for the ROS to decline as distribution increases. This is usually because the stores that take in the product later are not as productive. (By the way, a similar thing happens with penetration and buy rate – as more people buy a product, the amount bought per buyer tends to decline as the new buyers are lighter buyers.)
Hi Robin,
Could you show the formula on how ROS(Rate of Sales) is calculated?
ROS is just another term for velocity. See this post for more on that. The exact calculation may differ by country but in all cases the numerator is some measure of sales (value, units or volume) and the denominator is a measure of distribution (TDP, %ACV, # of Stores).
Hi I need your help, can I computed SPPD by using this: Sales Volume/ WD Total Stocks? Or it has to be % ACV Distribution
I’m not familiar with the metric “WD Total Stocks”. That being said, you can use any measure of distribution in the denominator and it will still be a measure of velocity. You just need to make sure you are clear with your audience what distribution metric you are using and understand the implications of using that measure. For example, you can use Sales per Store instead of SPPD. You just to keep in mind that Sales per Store will inevitably skew sales rates in favor of larger stores. And, if you don’t have %ACV as the denominator, don’t call your metric “SPPD”.
The answer to this question was incredibly helpful. It’s one thing to understand all of the definitions and concepts that go into analyzing the data but a completely other thing to put a story behind it that everyone can understand. You should consider doing a post of hypotheticals!
is this different from UPPD? hoping UPPD is Unit per point of distribution
UPPD most likely is units per point of distribution. Both the Nielsen and IRI databases indicate if a measure is in units, dollars or equivalized units (EQ) and the abbreviation for units is U.For example, in a Nielsen database I am currently working with, the description is Sales per Point of Distribution (U) and the tag name is SPPD.U. If you are not in the US then the fact names may not be as explicit.
Rank $ TY $ Vol Abs Chg $ Vol % Chg $ SPPD $ SPPD Chg Avg Price TY Avg TPR Price TY
1 380,636 113,541 43% 5,438 752 $4.28 $4.05
2 239,712 80,509 51% 4,700 817 $4.32 $4.07
3 236,408 74,538 46% 4,298 619 $4.29 $4.02
4 210,839 186,720 774% 5,142 2,128 $4.18 $3.99
5 142,777 44,582 45% 4,462 1,076 $4.15 $3.99
6 121,830 121,830 4,686 4,686 $4.11 $4.00
7 1,562 1,562 1,562 1,562 $4.46 $1.64
8 478 478 478 478 $2.99 $1.70
how do you interpret thsi data>
To learn the basics of data interpretation, here is what I would suggest for you:
1) Take some classes. Excellent classes are available from both CMKG.org and learningevolution.com starting at about $100 per class.
2) Read some of the basic articles on this website. I suggest you start with this article:
Are You Syndicated Data Literate?
This covers the most important terms and concepts and provides links to more detail about each.
Hi all, question I’m hoping you can help me get a better understanding:
An item’s distribution (%ACV) and unit sales are both declining year over year but drastic swings upwards noted for Unit sales per or of Distribuion. Generally, what are some factors that may be causing this spike in USPPD? Promo, pricing, weakening of competitive products?
It would not be uncommon to see Sales per Point of Distribution rise as distribution (% ACV) is lost. When you lose distribution in stores where your sales rate is weaker than average (the very place where you are most likely to lose distribution, right?) then your average sales rate has to go up. Why might your sales be relatively weak in those particular stores? Could be all the usual reasons: less promotional support, less aggressive pricing, stronger competitors, intrinsically weaker consumer demand (maybe the retailer where you lost distribution had a poorer fit between their shopper and your product).
When looking at Customer levels across the US (Total Walmart vs Total Alb/Sfy). Can you use SPPD or should you use Sales/$MMACV?
You need to use Sales/$MM ACV when comparing across geographies. That’s because a point of distribution is worth more at Walmart than it is at Alb/Sfy. $MM ACV is the same regardless of the size of the geography.
What are the best measures to use in SPINS to evaluate if sales growth is being driven by new distribution or velocity? SPINS advised I used Max % ACV for distribution and SPM (Sales per Million) for velocity. But, some of the comments I am seeing here suggest I should use SPPD instead? I am looking at the retail growth for two of our product lines across Channels (Nat/Spec/MULO), as well as, across same store retailers.
I am new to SPINS and to interpreting this type of data, so I am having a tough time analyizing the data that is coming back. For example, our sales are the highest in MULO for our items, but, ACV in MULO for same items is lower than other channels:
Total US MULO: $ Sales – $16,570,308, Max % ACV is 45%, $ SPM is $31
Total US Natural: $ Sales – $3,939,213, Max % ACV is 97%, $ SPM is $232
Total US SG: $ Sales – $1,088,174, Max % ACV is 90%, $ SPM is is $62
Is SPM the best way to look at this? Our items are in greater distribution in both Natural & Specialty channels, given their smaller size compared to MULO. So based on above scenario is it correct for me to glean that velocity for my item in MULO is driving growth?
Thank you!
You have a lot of questions here and it’s complicated! I’ll try to address some of them:
1) If you want to know what’s driving growth, you need to look at changes in the measures, not the absolute measures. In other words, to know whether sales growth is driven by new distribution, you need to look at change in distribution. The numbers you gave above are just one point in time. So no, you cannot conclude that “velocity in MULO is driving growth”. You can’t conclude anything about what is driving growth because you haven’t looked at what is and isn’t growing. Hope that makes sense?
2) What time period are you looking at? If it’s 52 weeks, I personally would not use a measure like Max % ACV. I think that’s too long a time span to look at a Max measure. I would discuss that further with your SPINS contacts. Make sure they understand what your business question is so they can advise you on the best measure to address that particular question.
3) Sales per million is a good measure of velocity. If you are comparing across channels, you can’t look at SPPD. It’s only valid within a particular channel or retailer or geography.
4) MULO is a huge channel – it’s way way bigger than Natural and Specialty Grocery. That’s why your absolute sales are higher there, despite lower % ACV and lower velocity.
5) Another thing you need to consider is how many items are in the line and is the % ACV for the line as a whole? If you are looking at distribution growth, that could come through breadth (getting any item from your brand into more stores) or depth (getting more varieties into retailers where you already have a presence). If you are looking at distribution growth, you need to look at both and the way to do that in one measure is TDP (Total Distribution Points). You could ask SPINS about that measure.
Hope this helps you get started down the right path! If you want to read through our articles in a logical order to try to fill in the gaps in your knowledge, I would suggest starting here:
Hope this helps!
Sally brings up lots of good things for you to think about, Kelly. To expand a little in a couple of things…
2) You don’t want to use Max %ACV in a velocity calculation for 52 weeks especially if your brand is getting into more stores over time. Using Max instead of Avg % ACV as the denominator in velocity will lower the velocity.
4) MULO is a much bigger channel than Natural and Specialty so even though the brand is in fewer stores (45% vs. 90+%), sales are much higher. It seem like there’s an error in the velocity you state for MULO. I have always seen velocity in MULO be much higher than in the other channels just because those stores have a lot more shoppers than Specialty or Natural stores.
Thank you both so much for this information. The data I pulled was for a 12wk total – We typically do not look at 52wks for the reasons you both provided.
For velocity, my company also prefers to look at Average Units Per Store/Week measure.
They also want to look at total ACV by Channel, as well as, a consolidated total of all channels for our items. Since you cannot add ACV, what is the best calculation/measure to use to get a both a channel total, as well as, a consolidated channel total?
I
I think it will be easier if we have a quick phone call to discuss this. Please use the Contact Us form to send a message and I will get back to you.
Hi, I understood what SPPD and TDPs are, but I stil don’t understand what are the reasons for rising or declining of those KPIs.this week I saw the SPPD of one brand increased vs YA instead its TDPs decreased vs YA.
TDPs change vs Year ago is -16% and SPPD change vs YA is + 14%. how can interpret these data ?
Hi there, The pattern you are observing it not surprising. SPPD is measuring sales rate- in other words, how strong your sales are where you have distribution. If your distribution declines, you are likely losing distribution in your weaker markets. Therefore, average sales rate will go up because sales are stronger than average in the locations where you were able to retain distribution. You might want to review our article on velocity to get a better understanding of SPPD and other velocity measures.
Velocity: How Well Your Product REALLY Sells
Hello!
Is it possible to forecast unit sales and ACV Reach for a new item launch, using Units/ACV%, Est ACV Selling, and Market ACV?
For a new item launch, you don’t have any of those numbers until you are in market. You can make a forecast based on historical new item launches (yours or competitors). If you expect to get a different level of distribution, you can take your forcasted/expected new item Units/ACV% and apply that to your expected Est ACV Selling and that will give you a units estimate. If you expect units/%ACV to be 100,000 and you expect %ACV to be 50%, then expected units = 100,000*50 = 5,000,000.
If you are using Units/ACV metric, you don’t need to factor in Market ACV. That number would be needed if your sales rate was units/$MMACV rather than units/ACV%. That brings up the point, though, that you can only use Units/ACV% to do forecasts or expected unit calculations if the market is the same for both the previous new items and the current new item. For example, if your historical items were launched regionally but the new item is being launched nationally, then you can’t do the simple calculation described above. Now you do need to adjust for the change in market size as well as your expectations about the %ACV you will achieve.
I’m not sure where ACV Reach comes into the picture. Maybe you mean the max ACV your new item will achieve? If that is something you wanted to forecast, you could likely use past new items, assumptions about the relative appeal of this new item, and assumptions about the strength of your marketing plan.