Coke and Sprite, Batman and Catwoman, Abercrombie and Hollister. Two similar ways of addressing a need, each with its own twist. In the Nielsen/IRI world, that duo is Store Data and Panel Data.
Store data is your basic, bread and butter sales tracking tool. It is gathered from retailers through their scanner/POS systems. It’s the first thing most companies purchase from Nielsen or IRI. Volume, distribution, price and promotion are the focus. Store data tells you what’s happening at the total store level. It’s sometimes referred to as “scanner data”.
Panel data is tied to individual buying households. With panel data, you can look at the consumer dynamics underlying your sales patterns. Panel data can tell you the demographics of your buyers, how often they buy, how loyal they are, and what other products they purchase. It’s sometimes referred to as “household data”.
Panel data has its own measures and vocabulary. Penetration and Buying Rate are the first two terms you need to learn. Next step: study up on Purchase Frequency and Purchase Size.
Nielsen and IRI both have similar household panel data services. IRI calls theirs Consumer Network and Nielsen’s is branded HomeScan. The underlying data actually comes from the same source – funny to think of those two companies collaborating on anything! They have a joint venture called the National Consumer Panel (NCP). The NCP is made up of about 120,000 US households, balanced to demographically reflect the most recent US Census. These panelists electronically record all of their UPC-based purchases. So the NCP panel captures data from all outlets and retailers. Panelists of course receive incentives for their careful work.
Retailer loyalty card data is also a type of panel data. Some retailers look at their information using a panel data approach, decomposing sales into penetration and buying rate and assessing customer loyalty. So getting familiar with panel data vocabulary can be crucial for communicating with some retailers. For example, Kroger is famous for their deep commitment to understanding shopper dynamics. They have a big staff dedicated to data analysis as well as a long-standing collaboration with loyalty consulting company dunnhumby, now known as “84.51.”
Want to see panel data in action? Here’s a common analysis used by manufacturers and retailers alike.
For more on panel data, check out this related post.
Got questions about panel data? Contact us or leave a comment below.
Want to get new posts delivered to your inbox? Subscribe via email. We won’t share your email address with anyone and you can subscribe at any time.
Costs
-How much does it cost a supplier / vendor for syndicated data?
-How much does it cost a retailer, or does a retailer sell to IRI/ACN?
-How much is HH Panel data?
Jamie, Thanks for your questions. Here’s what we’ve seen regarding costs:
1) How much does it cost a supplier / vendor for syndicated data?
Assuming you mean syndicated store scanner (not panel) data, it can cost anywhere from about $2,000 for a one time report delivered in Excel to millions of dollars annually for multiple databases and custom tools to access and organize the data.
The cost is a function of many things including: number of product categories (as defined by IRI or Nielsen), number of markets and accounts, what facts you need, and the time periods you require.
Ongoing data access usually requires a minimum annual budget in the tens of thousands of dollars. If that’s not feasible, individual excel-based
reports can be a good place to start.
Is it worth it? Consider the value you can get from the information and then compare that to the cost. If the data helps you sell in a new
product that ultimately contributes millions of dollars in sales, or if it
helps you to make a strategic pricing decision that increases profits on an
ongoing basis, then you’ll clearly get a great return.
2) How much does it cost a retailer, or does a retailer sell to IRI/ACN?
Retailers do not pay! They are providing data and their incentive for doing
so varies a lot. In many cases, the retailers get clean data and analysis
in return.
3) How much is HH Panel data?
Household panel data is quite a bit more expensive than scanner/store data. Like store/scanner data, panel data can be obtained as one time reports or
as an ongoing subscription. However, because the appropriate analysis period is usually longer (often one year or more), it is more common for manufacturers to get reports than ongoing access. Reports are priced based on specific analyses requested. You usually receive a discount from Nielsen or IRI off the “full price” for panel reports if you already buy scanner data from them. I can’t give you a typical price, because report prices vary so widely by topic.
In addition to providing valuable information on existing products (where the basic language is penetration and buy rate), panel data can provide particularly important insights for new products. The basic language in this case in trial and repeat. Trial refers to a household’s first purchase occasion for the new product and repeat refers to the second (or more) purchase occasion(s). In fact, after 4-6 months it is possible to accurately forecast the success (or failure) of the new product based on trial and repeat dynamics.
Here is a question we received from one of our readers:
“I am The CEO of 2 Natural Food stores. Together they form a monopoly due
to it’s demographic location being Kauai, Hawaii. I am pondering if
Nielsen or SymphonyIRI. has any source of data that can help my business.
I realize that in this post you are talking about either Syndicated Store
Scanner Data or HH Panel Data. My stores gross in the vicinity of 7
Million Dollars/Year. Given this knowledge do you have any products that
are specific to The Natural Food or Health Food Industry that you believe
could help me to increase sales? Thank You for any time you put into answering my question!”
Thanks for contacting us! Unfortunately, neither Nielsen nor IRI provides
data for Natural Food Stores. There is however another data supplier,
SPINS, that provides retail sales data specifically for Natural Food
stores (and also Specialty Gourmet stores). I don’t know if they include
any stores in Hawaii in their sample, though. You may want to check with
them on that.
Even if SPINS does not cover Hawaii, it may be helpful to use their data to look at which products/brands are driving growth in the Natural channel in the mainland US. Looking at sales for conventional/traditional grocery stores could also help you increase sales. Even if you have a monopoly for natural food STORES, you might not have a monopoly on natural foods SOLD. So finding out business size, items carried and pricing within traditional grocery could help you enhance your own business. I believe SPINS can provide information on conventional/traditional stores (they have partnerships with Nielsen and IRI) or you could purchase that information from Nielsen/IRI directly.
Hope this helps!
I am wondering which retail grocery-selling accounts do not report scanner data for Nielsen? Whole foods, H-E-B, gourmet stores? Isn’t Walmart now included in scanner data, nationally? Thanks!
Hi Kari,
For the food channel, Nielsen and IRI are focusing on the universe of stores with at least $2MM in ACV sales annually. So most gourmet stores wouldn’t make that cut and would not be included by design (as opposed to not being included because they won’t cooperate). The major food chains that do not cooperate are Aldi, Trader Joe’s and Whole Foods. There are others but those are the big ones. Sales for non-cooperating retailers are estimated by Nielsen and IRI using other sources.
Some chains cooperate and send data but do not allow Nielsen and IRI to release data for their individual trading area. HEB falls into that category. Their data is included in aggregate geographic markets (like total US or census regions) but you cannot purchase HEB account level data.
Walmart does cooperate. Their data is included in totals and you can purchase various Walmart and Sam’s account level markets. However, Sam’s is considered Club and not Grocery. Therefore Sam’s would NOT be included in the Nielsen and IRI total us grocery projections.
Hope that helps!
Can you explain Trial and Repeat and does distribution effect this measure.
Trial and repeat are terms generally associated with new products. They are often expressed as cumulative numbers, getting larger (hopefully) as more time passes since the product was introduced.
Trial is similar to penetration. It is the % of households/shoppers buying your product for the very first time.
Trial is highly correlated with distribution. If you don’t get distribution, there is not possible way someone can try your products. Other factors are packaging, advertising and promotion that gets the attention of consumers and incents them to try something new.
Repeat can be expressed as different metrics. One of the most common is % of triers repeating. This is less influenced by distribution. Bigger factors are product quality and perceived value.
Sally, could you explain the difference in panel between Total US Grocery for a particular retailer (let’s say HyVee) and HyVee Corp RMA? I thought they were the same but the numbers turn out to be different.
Oh gosh, there are so many nuances here! I will give you some basic info that may help but I suggest you get definitive answers from your data vendor (IRI?). I am not personally familiar with Hy-Vee data so that makes it tougher for me to help.
1) First off, are you just looking Hy-Vee in both cases or is one or both of these markets including Hy-Vee + competitors?
2) If it’s Hy-Vee + competitors, any market with “grocery” in the name will only include grocery stores. However, there may be a multi channel RMA that includes all competitors in all channels. Some grocery retailers have this and some do not.
3) The last potential area of confusion is which banners are included in the market. I don’t think Hy-Vee has multiple banners so likely not the issue but something to be aware of.
Best, Sally
Who and why use household panel data.
Could you please explain how exactly it works?
I have already covered your general questions as best as I can in this post. If you have a more specific question, you can feel free to ask it. You could also take a look at the other articles we’ve written about panel data:
https://www.cpgdatainsights.com/tag/panel-data/
The Promotional Allowance process — identification, negotiating, execution and reconciliation is slow, rigid and error prone. With online marketplaces and DTC momentum are there initiatives/effort to revamp this process for agility and velocity that you are aware of is being worked by any CPGs/Vendors.
Nothing that I know of, specifically, but of course it’s a perennial issue so I’m sure there are always industry groups and individual companies working to improve the process.
Surprised to see that the panel is a non-probability panel?
I assume you mean a non-random panel? That would certainly be ideal and I’m sure Nielsen/IRI would love to be able to promise that. However, because of the type of people who are willing to go through the effort required, Nielsen/IRI can’t get enough respondents from certain demographic groups. That’s why they need to boost/balance to get to a nationally representative demographic.
This brings up an important point about panel data in general. You always need to take care to ensure a panel adequately covers the population you are most interested in. Make sure to ask the panel vendor about how they recruit people/households of interest to you and the sample size available.