TABS Analytics | CPG Sales Analytics | TABS Analytics Blog

What is Trade Promotion Optimization and the Problems it Faces?

Written by TABS Analytics | August, 11 2014

Editor's Note: this post was originally published in August 2014 and was updated in December 2015 to include addditional trade promotion optimization content links. 

The concept behind a trade promotion is simple: spend money with a retailer with the expectation of deriving a tangible benefit such as increased sales, increased profits or higher market share. The mechanics, however, are more involved: in order to ensure that you're taking the right steps to drive more sales, optimizing your promotional plans is key. And, with trade promotion spending accounting for 15-20% of sales revenues[1], it’s even more critical to get this process right.

Trade Promotion Management versus Trade Promotion Optimization

The mindset that many manufacturers approach trade promotions with doesn't always make optimization easy. Typically the industry has just divided the process into two steps — Trade Promotion Management and Trade promotion Optimization. This basic paradigm can get in the way of making the most of any trade promotion plan you're considering because it will likely lead to your overlooking vital components of the process.

Taking a more granular approach will produce a better payoff. Breaking the process down differently can help you focus your efforts where they'll do the most good, and will guarantee that all stakeholders are speaking the same language. Consider this approach to trade promotions instead:

What is currently called Trade Promotion Management (TPM) is actually:

  • Execution – promotional information is entered into some type of software system
  • Tracking – users can follow the utilization of the promotional funds over time
  • Reconciliation – payments to retailers for promotional performance with the appropriate accounting entries

There is a notion among companies that they need to iron out every detail of TPM before they even consider Trade Promotion Optimization (TPO), but in fact, this is erroneous. TPO is really the learning stage of how you change the way you are doing things (and how you make more money) – and you can start doing this now.

What do we mean by Trade Promotion Optimization?

What is currently called Trade Promotion Optimization is actually five distinct and very different steps

  • Data harmonization – getting all of your data sources (sales data, spending, performance dates and qualitative information) aligned so that the results can be analyzed.
  • Measurement – a formal process to measure the results of specific promotional events
  • Strategic analysis – an aggregated view of the individual events analyzed in the measurement stage
  • Planning – a process to plan for future promotional activity using the learning from the strategic analysis stage.
  • Optimization – a mathematical model that provides the ability to generate the best promotional calendars based on a series of constraints identified by a user.

What Problems Prevent You From Optimizing Your Promotions?

Breaking your Trade Promotions Optimization down in this manner significantly improves your ability to act on the data you're collecting in an ever-improving way. After all, you're never going to run only a single trade promotion program over the entire life of a brand so you will continue to learn as you go.

Seeing the trade promotion process as a long-term, multi-year opportunity to optimize is the first step to improve your ROI.

But there are problems that face the Trade Promotion Optimization process, and these problems are what has kept the vast majority of large CPG players from having even a minimal TPO capability. Specifically these are the hurdles that you will need to overcome:

  • Good Sales Data: If you are you using shipment data to gauge the efficacy of your trade spend you are off on the wrong foot already. In fact, throw it out entirely. There's one reason: DIVERSION! Even if you don't care to click on the Diversion blog post link that explains the rationale, accept it as a truism that weekly retail point of sale data is the only way to accurately measure trade promotion effectiveness. Also accept it as a truism that any supplier that claims shipment data is essential to the TPO process is not an expert in data, analytics or the CPG industry. Retail sales data is easily attainable either through the syndicated data suppliers like Nielsen or IRI or directly from retail customers such as Walmart, Target, Home Depot, Babies ‘R Us, Staples and dozens of other retailers.
  • Misguided obsession with getting good spending data: Can’t get to anything close to good spending data? No problem, you can still measure promotional effectiveness and model future promotional lifts without spending data. In fact, I’ve often stated to clients that their initial Strategic Review should be done without spending data. The goal in this initial review is to accurately estimate the effects of various promotional options. Once we can do that we can then back into the maximum amount I should be willing to spend to get that level of sales lift. As a managerial matter ongoing, we will want to get accurate spending data and promotional ROI’s to confirm that promotions are being run correctly. The first promotional review should be a “no judgement zone” where we there are no negative consequences to bad promotional execution. Assume everyone has done things poorly, and we are starting from scratch right after there is agreement on the Strategic Approach going forward.
  • Lack of Skill in Data Harmonization: Most software vendors are just that, software vendors. They can program exactly what you tell them to do, and most of them have really slick front-end tools to visualize the data. Unfortunately you never get to that slick data with these lumbering software vendors (and you know who they are) because they are not experts in harmonizing disparate sets of data. Why? They usually lack the two core elements required to harmonize data: lack of experience in harmonizing data and lack of understanding of the business issues that need to be properly coded into the data. Here’s an example: a software vendor that agrees to build item-level promotional measurement capabilities for 2.6oz Degree Antiperspirant doesn’t understand CPG retail. Retailers rarely promote single items of Degree; they typically promote all items of the same price point, size or even the entire brand. Build the database around your level of analysis instead of placing an unnecessary level of analytical and data processing burden on everyone. So who’s left with cleaning up your database on product, time, markets and measures? Usually the client. It’s similar to a plumber coming in to tell you what’s wrong with your sink and then leaving it to you to fix it. With most of us in that situation, chaos will ensue. (I can just see some Demand Planning schnook shaking his head violently disagreeing with the premise that item-level detail into the promotion measurement process is unnecessary. Schnook, there are other ways to get to accurate item-level forecasts without clogging up the Promo Measurement tools with superfluous data).
  • Organizational Micromanagement: Most people at big companies would just call this “getting alignment” when 20+ people are on the steering committee to create the TPO specs. I don’t have an easy answer on how to overcome that except for an autocratic CEO laying the hammer down on 3-4 smart people to figure it out and telling everyone else to get out of their way. It’s no coincidence that companies $500MM-$1B have a much easier time at implementing TPO; they have less backbenchers chiming in on what to do. The last three TABS Group TPO implementations have taken 13, 6 and 4 weeks, respectively (notice a trend?). I’ve heard three speeches from Mega CPG on their “journey” to TPO. All of them took 3 or more years to implement.
  • Good Complementary Data: Maybe this is misclassified as a hurdle because some obvious flaws in syndicated baseline sales estimates and tracking of causal data such as display levels don’t seem to deter companies from charging ahead on their TPO mission. In the off chance the results actually transpire, however, the credibility of the analysis is often scuttled from members in the field who immediately identify odd baseline spikes that underrepresent the true incremental sales or understated Display ACV that don’t give them due credit for promotional execution. Once the credibility of the analysis is undermined at the field level there is almost no chance that the promotional measurement tools will be utilized. Here’s a partial list of where you will experience flawed data: syndicated baseline sales, syndicated Display ACV’s, Weekly or 4wk ACV, Percent Store Scanning in Retailer POS (understates real distribution), and equivalized volume (retailers don’t track ounces, they track units…and that’s what consumers buy).

In 2015, we did another series on trade promotion, covering the following areas:

These articles will help you harmonize, measure and analyze data so you can trust that the results are enabling you to make the right decisions and understand which future strategies will maximize profits.

[1]Gartner Report: Vendor Panorama for Trade Promotion Management in Consumer Goods, 2014 (Published 28 March 2014)