In any type of organization, Finance (often FP&A) has always had healthy conflict being at opposite ends of the spectrum with partner teams:
- FP&A wants to grow the bottom-line operating profit (i.e. EBITDA on the P&L)
- Marketing wants to spend more to grow its customer base
- Account managers want lower top-line volume targets to overachieve and make a larger bonus, etc…
Getting all the teams to buy into a shared goal of profitability has never been more challenging for an e-commerce FP&A team given the numerous stakeholders involved. How do you determine the optimal efficient marketing and direct expense spend to drive volume growth when building a forecast? Starting with overall top-line volumes to cascade to partner teams will allow the FP&A team to drive to an optimal bottom-line target.
The balance that needs to be managed by FP&A
Often, FP&A is tasked as a business partner for various groups within an e-commerce organization – business operations, online marketing, account/supply management, customer service, etc. Managing these partner relationships is a delicate balance for FP&A because each business function has fundamentally competing interests with one another, while at the same time everyone wants more resources to deliver. If these interests are not appropriately balanced, then the P&L will be lopsided in one of the function’s favor – which may be good and/or bad, i.e. too much marketing spend creating too much volume, which also means the call centers and fraud management are understaffed.
Let's grow! Let's be profitable! …. Uhhhh can we do both???
Depending on how large the gaps in expectations are between the various business partners, a forecast can iterate multiple times. This means that for each iteration, data must be reloaded into models, profitability must be recalculated, and multiple touch points are needed across the organization to validate the new targets.
If the teams are small or processes & systems to manage these recalculations are efficient, then this forecast becomes a quick turnaround (and the FP&A team would in this case be considered an expert at data management and forecasting). However, this is usually not the case due to the number of stakeholders and time constraints involved. What eventually happens is that the cycle of multiple iterations becomes too complex and ultimately no one is happy with the end-result of the forecast.
Setting the forecast up for success
Conflict will always reign, however healthy conflict drives organizational success.It is important to ensure that targets aren’t a 'he-said, she-said'-type of battle and the best way to do so is … to use data! Data minimizes conflict because it can nearly eliminate the surrounding subjectivity. Consequently, the most efficient process that I have found in an e-commerce organization is as follows:
- 1. Project a momentum-based view of product volume sales based on historical seasonality
- 2. Align with the senior leaders on this view
- 3. Combine this top-down overall view with a segmented perspective for each of the business partners and overlay in any expected variations due to new campaigns, new products, etc…
- 4. Apply margins, rates and costs to these volumes
- 5. Assess the risk in each P&L component once you put it all together
This standardized process provides multiple stakeholders the ability to anchor on a shared target while giving them the opportunity to help drive their own projections for their function or business area.
Consumer behavior is surprisingly consistent over time, and this is profoundly true when it comes to travel. For example:
- In Mexico, the busiest time for consumers to book and travel is during the Easter holiday, and they mainly travel domestically at this time
- In Canada, consumers primarily book all-inclusive vacations to the Caribbean in January as the cold winter months will have started to take a toll on the nation’s psyche
Similarly in the Retail domain:
- Consumers in the US make the majority of their purchases in November and December for the Christmas season, and
- In China, Singles Day is the largest single-day driver of sales in that economy each year.
This inherent seasonality is important to understand, especially in regional markets, and is fundamental to the customer behavior in most e-commerce organizations. That’s why this is the best starting point to begin a forecast: use seasonality to project your transactional purchases by product and by region while also baking in some expectation of growth based on history.
In mature e-commerce companies, this is not as complicated but requires a clean view of the historical data. In start-ups, this is much harder as the understanding of seasonality is murky given there is not a lot of historical data for the business. To guide expectations, do some research on similar industries and competitors as a starting point to get an idea of consumer behaviors. A baseline is necessary to drive the overall process. It doesn’t have to be perfect, but it will certainly guide the big-picture expectation of performance.
Can we all just get along?
Getting buy-in on this seasonal perspective is critical as it sets the expectations with the leaders of the organization. It provides a starting point for the team and it anchors everyone on a single metric (i.e. product volumes) that is agreed upon by the leaders. The functional/business leaders cannot debate their own perspective afterwards because the team now has a shared goal (if they do, they may lose credibility themselves). Without this alignment, the business leaders have more of an incentive to skew their targets to meet their own function’s goals.
All business units need to rely on this top-line product view because variations will cause them independently to suffer: from overstaffing call centers due to higher than expected volumes, to not enough spending such that online marketing misses out on opportunities. Think of an efficient supply-chain: every stakeholder group needs to ensure its view of the product volumes are as 'correct' as possible to meet its own input and output needs. However, getting this overall view right is not easy – this is where FP&A earns its money as a business partner: in addition to expected seasonality, the top-line outlook should take into account any significant anomalies that occurred or are reasonably expected to occur. If this does not happen, FP&A may lose its own credibility as a business partner.
Segment, segment, segment …
Once this overall top-line projection by product is set, the volumes are passed on to the other teams to work with FP&A but should be segmented for each business function, and the segment characteristics are usually different. For example in online marketing, these product expectations are broken down by channel, while for the supply managers, the products are similarly parsed by supplier groups. This additional dimension can create complexity quickly depending on the number of channels, suppliers, etc. Tools such as Forecast Pro can help automate the deconstruction of product volume projections into appropriate segments using history.
This partitioning of volumes using different segmentation criteria is a great test to discover the smaller anomalies that were not accounted for in the overall product view. These anomalies should be gathered by the FP&A teams and viewed as either relevant, or dismissed as inconsequential – for example, the sum of the gaps identified can either be applied to adjust the original top-line volumes, or absorbed in other segments within the targets. Assessing the risk of these anomalies in the forecast is important for performance management and should be summarized for the senior leaders.
Flow all the way down the P&L
Once volumes are agreed up on, the corresponding margins, rates and costs need to be applied to these volumes to help build out the P&L. Seasonality is not just for volumes, but also spending and rates. Specifically in both Retail and travel, hoteliers and airlines can increase their prices just before the holidays, meaning that marketing teams want to increase their spending then to ensure a strong market share, while customer service costs consequently increase during those times due to higher call volumes.
One thing to be careful of is that rates may drive product volumes outside of seasonality depending on the business – i.e. promotions and competitive pricing. This needs to be determined by the FP&A team when setting top-line product volumes: the larger the impact of the rate change, the larger the variance in expected volumes.
At Hotwire, we have multiple FP&A teams that anchor off of these initial volumes and apply seasonal rates to drive our targets for the different business functions. This coordination is complex but would have been much more so if the initial anchor of the transactional top-line would have not been set.
Putting it all together
The calculated bottom-line EBITDA based on this flow-down from top-line volumes is now the expected profitability based on input from multiple stakeholders. Put another way, it’s a bottom’s-up view of profit based on shared volume, rates, spend and margin guidance (outside of indirect operating expenses). However at this point, this is where FP&A’s financial stewardship and curiosity must come through to question the teams:
- Is the EBITDA target reflecting the proper P&L expectation? (rolling forecasts should help align expectations to prior forecasts and point out appropriate variances)
- Are the business teams applying direct spending efficiently based on the volumes projected?
- Are the rates and margins following the current trends or is there risk baked in?
- Do we need to adjust the forecast for new products or campaigns?
Ensuring that the right metrics are available to prod the teams on their accuracy is critical for forecast success. In addition, knowing where to put the risk in the P&L requires the FP&A team to be close to the business and understand how the metrics fluctuate compared to expectations.
It’s not up for debate … but rather to strive for!
In the end, the forecasted EBITDA target compiled should not really be debatable because it was built using a standardized process with alignment at the top driving expectations all the way down to the bottom-line. Conflict should be minimized because fact-based analysis sets the targets and not senior leaders trying independently to drive their own agendas. 'Sand-bagging' should be able to be caught due to the functional interdependencies and alignment on the initial top-line target.
An unbiased calculated view also helps set the longer term financial strategy – where do we need to invest more, or less? Are there P&L gaps which we can address with new products, or campaign timing? A systematic top-down forecast process helps answer these questions, as opposed to setting a profitability target without understanding the drivers that get you there.