What You’ll Learn:
- How to conduct an effective post-event analysis AND apply key learnings
- The importance of cost effective promotional strategies for you and your retail partners
- Which factors are responsible for your most lucrative results
Did you know that according to the 2019 POI State of the Industry Report…
- CPG companies spend about 23% of their revenues on trade promotions – which is one of the largest expenses on your P&L.
- Within CPG, 30% state that revenue management teams are the owners of post-event ROI analysis (what we refer to as PEA). ROI incremental profit was the most common KPI that they measured.
- What’s interesting – is that 25.3% of organizations out there have not established an RGM team.
- 18.7% are taking a non-RGM approach.
- 45.3% will be advancing their RGM practices in the future.
- Almost 35% will be advancing their technical resources with tools and advanced capabilities to be able to do more PEA.
Let’s take a poll 🖊
Post-event analysis (PEA) is very important. But many CPGs aren’t sure how to go about doing it, or where to start.
How Do You Conduct Post-Event Analysis?
To start, every post-event analysis needs to have clean source data. I’m sure you’re all over ran with data — data is everywhere!
How clean is it? How easily accessible is it? There are so many sources of data… in multiple locations. It can take you hours, days, or even weeks to bring all the data together before you can glean a speck of insights.
Many CPGs are finding value in software systems that provide trade promotion optimization (TPO) to assist with these needs. TPO provides a place for all of your data to sit – in one location – combined and visible to easily decipher what’s going on with your trade events. Plus, you’ll start understanding business mysteries like:
📊 What was your base volume?
📊 What are the lift coefficients?
In addition to data, one of the biggest hurdles is change management. Doing things the way they’ve always been done is definitely the easier route to go. But is it the most productive and the most cost efficient? Just because we understand what will happen if we run the same promotion year after year, does that mean it’s the best promotion to continue running? Could we do better if we mix it up a bit? What if we plan for something completely different?
And lastly, having a clear and consistent approach to your organization’s trade spending ensures that the most efficient promotional plans are being executed and money is being spent wisely.
Your Guide to Post-Event Analysis…
(video time: 5:44)
What we’re looking at: The promotional event has been completed and consumption data now exists in Blacksmith TPO master calendar. The system is showing us that a historical view of the most current two years for a specific brand, at a specific retailer.
Zeroing in on the bottom portion of the calendar, we can see:
- When we promoted and when we did not.
- What our average pricing was week in and week out.
- When national consumer marketing events were being held.
- What our competition was doing.
- If EDLPs going on.
- Retailer POS data.
Here is where the beauty of taking multiple data sources and combining them into one view really helps.
Looking at the last 2 years, we can see that there were a total of 5 promotions that took place.
(video time: 6:50)
Some were more successful than others. Each contained specific promotional tactics that drove incremental volume during that event. We can see how price changed during each event as well.
Out of the 5 promotions the one directly in the center of the calendar appears to have driven the most volume.
We’ll conduct this PEA on this specific event – and learn what performed well (and what didn’t).
(video time: 7:30)
When I hover my mouse over that particular event, I’m able to see a quick and easy read of PEA for this event.
🔎 The event was held during the weeks of August 6 – August 19 2017 — so it was a two week event.
🔎 In total, it generated almost 52,000 units or around 4,300 cases.
🔎 Syndicated data states the base unit volume for both weeks combined was just around 7,300 units or a little over 600 cases. (This is volume that syndicated says we would have sold regardless of having a promotion – basically, our everyday volume.)
🔎 The promotional tactics executed during this promotion were an ad, a display, and a TPR. We can see the amount of volume that each generated by looking at the colors in the chart as well as the breakdown of units.
🔎 Display drove the majority of the volume with 52% of the total volume.
🔎 Even though an ad was present, it doesn’t appear this retailer had much execution of it, thus driving very poor results for that tactic alone.
🔎 The average price during this promotion is $1.65 – a discount to the consumer of 49%, which is a tremendous decrease to the average regular price of $3.24. It still didn’t drive the volume that the display did.
Results: The Good, The Bad, The Ugly
We can see that total volume exceeded any other week in the past two years.
(video time: 10:15)
- Display drove the majority of the volume.
- The ad did not drive much volume.
- Did the ad not execute? How much money did we spend on that ad?
- The TPR generated a decent amount of volume but at 50% off the regular price, I would have expected more volume to have been generated.
❓ So that leads me to the question❓
Is there a need to drop the price so drastically?
Without such a drastic discount, less could have been spent with a retailer to buy a price down. And we know that in a lot of retailers those half price / BOGO events – they require them – but maybe we can work them to not have as many of them – especially if both parties are not gaining what they had hoped to.
Competition didn’t appear to be promoting the same week as us. We can see that in the calendar. But could the competitor’s promotions the week prior to ours cause our promotion to be less effective? Was our planned and approved trade spend for this event (within our TPM system) aligned with the actual retail event? By bringing in that approved and planned TPO programs from within TPM we can ensure that when we are planning to execute promotions that that’s exactly when they’re taking place at retail level.
Finally, we see the overall results are that this event generated a total ROI of -25% and an incremental per case spend a $14.40 per case.
(video time: 12:20)
These figures are eye opening.
Let’s take a look at one more interesting observation from our master calendar. (video time: 12:38)
Notice that a few weeks later, we had a second promotion – that although not as large as the promotion we just reviewed – didn’t perform poorly.
Interesting enough, this promotion was executed in the very same manner as the promotion we just analyzed… but with very different results.
Not as much volume sold. The promotional price was the same / $1.65. We generated display activity, but we didn’t have a strong display activity in the second promotion.
There’s only one key component we might be missing if we didn’t have multiple data sources viewable in one spot.
What was the competition doing?
(video time: 13:31)
Notice a competitor was promoting the exact same time that we were. So could that have possibly been the driving factor behind our volume being lower in this promotion versus the previous one? What type of promotion was the competition doing and how successful was it? Can your organization answer these questions? These are critical parts of PEA and they cannot be ignored.
Should We Repeat the Past?
Do we ignore best practices and just go with what’s comfortable and repeat those same promotions again next year. If we want to strive to achieve better trade spending than we must apply best practices in situations such as these. Change it up. Think outside the box.
Within TPO, the ability to perform optimization exists to help you determine what to do, when to do it, and how to do it.
Utilizing TPO constraint based modeling, we can optimize the plan to generate more profit for both the retailer and the manufacturer.
What-if event scenarios and optimizing provides users with the ability to think outside the box and develops fact-based promotional strategies that can mean the difference between a successful promotion and money that’s wasted.
Using Blacksmith TPO what-if event scenario model, I apply the same tactics as the promotions we just analyzed to see how they would perform the following year. Pretty much the same situation…
(video time: 16:03)
👎 Negative ROI for the manufacturer
👎 Subpar margins for the retailer
👎 Repeating the same mistakes again
How do these two promotions differ? Starting with the same BOGO and then applying our new optimized event scenario.
Comparing it side by side with our optimized event, we can see how they differ and what benefits exist going with the new optimized plan.
We definitely see that there are less cases sold in the new promotion, but higher incremental revenues achieved.
There is a decrease in total spending of almost $42,000 for the manufacturer, which those funds could be used to add maybe an additional event or two throughout the year or provide some other type of marketing support.
We can see there’s improved ROIs for both sides and an increase in an incremental profit for both sides.
(video time: 18:00)
Key Learnings
📓 PEA is critical to ensure that previous ineffective promotions are not repeated (and you’ll understand why).
📓 The analysis provides the retailer with fact-based data of how and why promotions executed as they did.
📓 Visibility into the past provides much needed “guardrails” to stop bad practices and begin implementing best practices.
PEA Success Stories
CPGs can build out the best library of cost-effective promotional strategies to save weeks of head scratching on “what do we do this year” and ensure a positive promotional experience.
(video time: 22:20)
🏆 In the first year of using Blacksmith TPO, this customer saw a 16-basis point increase incremental profit of promotional plans. That’s more than 12x return on the initial cost of the software.
🏆 In 3 consecutive years, a TPO customer experienced YoY growth. Most recently, showing a 2% increase in net revenue and a 6% drop in trade spending. That’s more than 21x return on Blacksmith TPO software.
🏆 This CPG made a 20% improvement in the accuracy of their demand forecast.
🏆 One team of analysts reduced time spent on PEA by 50%. They went from analyzing only their top 5 customers quarterly to being able to analyze all of their customers.
Q&A
Question: What type of data is best to do proper post-event analysis?
Answer: Sources of data for PEA include syndicated data, national/consumer marketing data, retailer POS, shipment data and TPM trade spend plans.
Syndicated data provides us those data points on the merchandizing conditions that were employed during that promotion. And not just your own syndicated data; we want to bring over the entire category so that we’re seeing your competitors. That competitor information has a direct bearing on your products and your promotions.
Bring over national/consumer marketing events. That dats is very, very helpful. In our master calendar, we can see instances of where if you did not have that information overlaid, maybe there was something going on. For example, we may see that for a couple of weeks in a row our base volume is growing but we don’t have any promotions going on, and we’re questioning why would base volumes start to grow for these two weeks. Well, as it happened, we had a national FSI coupon that dropped that week — that was what was driving volume to go up. Not having the visibility into that data overlaid on top of our syndicated data, that would have been a big question mark – a big gap for us.
Question: How often should organizations be conducting post-event analysis?
Answer: I think on a quarterly basis is sufficient.
Once your users start to adapt and learn how to do effective PEA, as soon as that data gets posted and the promotion is over with, they can create a habit of going in, doing that analysis. and making those notes.
So I would say at the very least, I would want to be doing it on a quarterly basis so that you can stay on top of what activities are working and what aren’t so that you have time to adjust for future promotions.
Question: Can you tell us more about the ability to cleanse the data we sometimes see the wrong SKU, the wrong feature dates, or shipment dates, which impact the volume.
Answer: The biggest thing out is data but unfortunately, lots of times that data is not necessarily clean. Our team at Blacksmith will work with you to bring in certain data points, we will direct you with extract samples and let you know what fields of data that we’re needing to come over from either your syndicated data, your retailer POS, shipment data all of those data sources and we’ll work with you to get those data sources aligned so that they are all talking about and aligning with the correct dates, the correct products, etc. That in itself can be a big task for an organization to try to take on alone if they’re merging all of that into some kind of homegrown system or even into a big Excel sheet or pivot sheet. Our team will work with you to ensure that all of your data is aligned and cleaned so you can do proper work within the system – and conduct proper PEA.
Question: Can Blacksmith TPO correct that data?
Answer: Yes, you do have the ability to correct that data. So for example, let’s say that your approved trade spending plan that comes in from your TPM system that was supposed to take place the first two weeks of May but when you go to do your PEA analysis, you can see that the actual event occurred maybe the second and third week of May instead of the first two weeks may as you had planned. You have the ability from within the user interface to go in and align that event manually, it would not require development resources from behind.
So that’s just one example.
And we can definitely talk with you more about other ways that you would be able to clean up your data if it’s needed if it’s not been done already when the data is integrated.
Question: From beginning to go live, how much time do we need?
Answer: Well, we can get you up and running in Blacksmith TPO in as little as 8 week period. There are always things that can come up, such as how quickly you can get the data sources to us, as well as what amount of time is involved in cleaning up the data. But we have implemented lots of clients in as little as 8 weeks. You can be up and running doing your post-event analysis and what-if planning.
If you’re entertaining the thought of bringing in a system let’s talk more.
