SkuBrain Blog

Here at SkuBrain, customers often ask us, “Why does my forecast look like this?” The answer is (almost) always, “Because of your data”. SkuBrain uses historical time series data to predict future performance. Although it runs through lots of statistical algorithms to choose the best one, and does all the heavy lifting to reconcile forecasts at your chosen hierarchies, ultimately the reason why a forecast looks the way it does is because that’s what the data says.

However, in some cases the future is not captured in the historic data, and management action can often explain these disconnects between historical data that builds a forecast, and the performance of that forecast going forward. So first we will look at common examples of where forecasts fail, and why, and then turn to the strength of SkuBrain to get the forecast right under conditions where the future can be predicted, at scale, customized to the SKU level, on your sales and inventory data. The value of this forecasting is improved when accurate versus inaccurate results are segmented, reviewed by management, and new effectiveness trends are identified.

As examples of where management review and forecast accuracy segmentation can add value, here are some cases of where SkuBrain forecasts are inaccurate, with explanations and strong hypotheses as to why.

DISCLAIMER: Some of the data shown in the following examples have been time-shifted into the future to demonstrate the effects under discussion. That’s why you’ll see some “actuals” which are in 2017.

Forecast 1

Here is a case where a SkuBrain forecast was inaccurate, and the visualization shows why. With almost no sales in the two years preceding the forecast period (Oct 2016), but very strong sales in the three years prior, SkuBrain predicts modest performance. Instead, performance goes off the charts. Clearly SkuBrain was trying to strike a happy medium, and something else in the business process – or market demand that was not apparent for two years – must have changed for the sales to both increase and show repetitive three-month cycles.

Forecast 2

This forecast shows activity of an unpredictable drop in product demand. There appear to be two distinct patterns in the training data: a peak every three months that is trending upward forcefully; and the “normal” sales that are trending upward, but more moderately. SkuBrain’s best guess is a happy median, and as more data are acquired over time the stable cyclical trend in the data will be reflected in the forecast. But early on, twelve months of training data is not enough to tell one way or another, so SkuBrain picks a flat average forecast. This turns out to be really off, since sales for the product actually tank. Rather than ask “why is the forecast off?” the real question is why did the product tank, and whether that is due to issues that management can control.

Forecast 3

Nothing in the historical sales suggests that the recent, sustained, six-month upsurge of demand starting in July 2016 was going to continue, or stop. Management would hopefully know. With the data it does have, SkuBrain predicts a decline in near-term sales. (SkuBrain has produced a base forecast – the green shaded area – and a final forecast, indicated by the fine green line – to lean more about forecast hierarchies, click here.)

Forecast 4

With a single data point of sales, no algorithm in the world will give you an accurate forecast. In fact, SkuBrain’s internal algorithm would more than likely consider this to be a data error, or a one-off event, and so predict (correctly, in our view) no future sales. Management would need to identify which products should be excluded from forecasting so that the general results of the forecasting process are not biased negative because product segments that no forecast will predict accurately are included in the evaluation of the forecasting method.

Now that we’ve reviewed these examples, let’s look at some examples from the 90-95% of SkuBrain cases that show acceptable accuracy

This product has five years of data, with a lot of activity in the first year, followed by a slump, and stabilizing somewhat in the next three years (’14 to ‘16). As such, Skubrain bases its 12-month forecast on the more recent performance, and gets it right for the month of October 2016.

A Classic forecast

There is some seasonality (a dip every January) but no strong trend upwards or downwards. With three years of data, SkuBrain has produced a classic forecast – predicting September sales to within 1.39%. Few forecasts are this good!

Seasonal forecast

Very seasonal behaviour – perhaps, if you were a brewery, this would show behaviour by a summer beer (or a winter ale, if you happen to be in the southern hemisphere). The peaks are not obviously trending up or down, but July, August and September show reliable demand. October 2015’s forecast was highly accurate and we expect the forecast to be highly accurate in the future, because the underlying data trends are exactly what SkuBrain works best with.

Base vs. Final

Here is another seasonal product, but the sales trend is more nuanced. Again, SkuBrain has made two forecasts here – the Base Forecast at the SKU level (green shaded area) predicts a slight dip, but the Final Forecast retains the previous years’ performance, due to strong performance at the category level (not shown). To understand the difference between the base and the final forecast, have a look here.


Here, the data shows some cyclicality (every six to seven months), as well as seasonality (a clear peak around March or February). However, the surge in Feb ‘16 is not repeated historically, so SkuBrain errs on the conservative side. Even so, the October 2016 forecast is within 3%, and over time the cyclical nature of the data will be reflected in the forward-looking forecast.

Phased out forecast

This product’s best days are behind it. Despite good years in ’14 and ’15, sales have clearly diminished in ’16, and SkuBrain picks that eventually sales will subside to zero, perhaps punctuated with a half-hearted resurgence or two before petering out. Naturally, management may know this product is already phased out and, if so, then no future forecast is needed and these results can be set aside.

Adjusting forecasts

This shows another one-off product that has seen better days. Of course, SkuBrain could be wrong, and in fact the product peak in February could indicate a seasonal product that requires forecasting adjustments. But nothing suggests that is the case in the (less than one year of) data at hand. In cases like this, the business planner will need to adjust the forecast before creating a plan – something that is easy to do with SkuBrain.

Forecasting with accuracy

This product has decent sales data stretching back three years, but with no recent sales in the last 12 months SkuBrain predicts no further sales in the next 12 months, and the results for October 2016 concur. Note that just as we might segment out difficult-to-predict products in order to assess general forecasting accuracy, one can argue that cases such as this should also be excluded, because product discontinuance is the proximal predictor of no future sales, and attributing accuracy to a forecast in that case will tend to inflate an estimate of future forecasting accuracy.

More data, better results

This product has five years of data, with a lot of activity in the first year, followed by a slump, and stabilizing somewhat in the next three years (’14 to ‘16). As such, Skubrain bases its 12-month forecast on the more recent performance, and gets it right for the month of October 2016.

Some cases (SKUs) will always result in prediction errors. Management action and expert knowledge often explains these cases. The insight is that management should actively segment the products into what is amenable to prediction, and apply expert knowledge and data visualization to improve that segmentation over time. Setting aside those difficult to predict cases, SkuBrain is very accurate on 90-95% of forecasts where data are stable. The algorithm tournament fits the best forecast given the data, and as the data mature the forecast will become more accurate.

Assessing an algorithm’s forecasting accuracy should exclude segments that are either impossible to predict (preferably pre-identified segments based on management input during the data warehouse build) or very easy to predict based on a management decision to discontinue a product last year and where SkuBrain will accurately predict zero sales in the future. What we are saying is that SkuBrain is an aid to improved forecasting, at massive scale, and management review, expert knowledge, and learning from the SkuBrain results over time will improve the value of forecasting.

SkuBrain is free to use for 7 days. Sign up here and start forecasting!

When customers approach us requesting some guidance or feedback around the way they use SkuBrain, we are given the opportunity to observe their unique interactions with SkuBrain and draw some interesting conclusions.

We’ve noticed that some of our customers are uploading data, running large forecast jobs, and then re-running on a similar but “pruned” data set. This is a good practice, so we want all our users to be thinking about how forecasting data need to be filtered before accepting the results of the forecast.

A common filter is the elimination of data spikes that can throw off a forecast. Those filters can be applied earlier in the process to reduce the time needed to upload a new data set, and we have an illustration of this method using Excel (see below).

Most businesses see occasional spikes in their sales. Sometimes these spikes are data errors, and sometimes the spikes reflect real sales on a cyclical basis (holiday season, back to school, etc.). Unfortunately, error spikes tend to ‘pull’ the demand distribution in the direction of the spike, potentially skewing inventory planning. And if you do not look at your data before accepting a forecast, the error spikes could ruin the forecast and cost you money. So this is important, and it requires some judgement on the part of the business user.

So always look at your data, even if it is just a sample of SKUs or points-of-sale (POS). Look for spikes in the data that do not make sense, and do not hesitate to remove a record, or a set of monthly records, that seems to not match your more recent sales experience. You can always use SkuBrain to re-run a forecast next month, incorporating your most recent sales and inventory experience, so dropping some old data makes good sense given that new data arrive all the time, and the most recent data in general are more relevant to your forecasting needs.

From a forecaster’s perspective, spikes in the data should be researched separately from interpreting the forecast, because once the forecast is available most people tend to focus on the forecast and assume the earlier data filtering is sufficient. But consider this situation: a transportation company may operate vehicles from various locations. When repairs or maintenance are done at an alternate facility while a vehicle is in transit (say a freight depot where a breakdown occurred, rather than the maintenance depot where trucks are usually repaired), the purchase of spare parts should be recorded as one-time events so that the freight depot does not begin to stock parts for future repairs. Although this is a simple example, we have seen similar issues occur when running massive forecasts that incorporate information on geographic location. The solution is simple: filter out locations where the demand forecast is irrelevant -and do this with confidence, because you can always revisit the issue with new data, a new forecast, and knew insights as they develop.

Now, back to SkuBrain. In a forecasting system, we generally want to eliminate these types of irregular spikes from our forecasting. SkuBrain manages data spikes inside the algorithm tournaments that it runs and it will choose the algorithm with the lowest mean average percentage error (or MAPE). These algorithms, (like ETS, or exponential smoothing), tend to ‘smooth out’ spikes that might be considered statistically insignificant, and are suitable for forecasting data with no trends or seasonal patterns.

The way in which SkuBrain manages spikes over pre-determined time periods can be best illustrated in the following forecasts. Studying these graphs will help you to identify situations where a data spike should be pruned. Graph A shows a basic forecast devoid of any significant one-off spiking in the data:

Graph B shows how SkuBrain has forecast against a spike four years prior:

Graph C illustrates a single data spike in 2015:

As you can see, single spikes cause the forecast to flatten and diverge from the smooth upward trend over time. So if you see a spike, you’ll likely also see a forecast that looks a bit odd. If you are an Excel ninja, you could return to your data at this point, aggregate it by month and then identify where the spike is being generated and massage or remove the offending data. In fact, this is what yielded the results in Graph A.

If the data cleaning tasks exceeds your Excel capabilities, please keep an eye out for a SkuBrain blog post in the coming weeks that will guide you through this process, step by step.

Graph D shows how SkuBrain has incorporated a recent data spike into the forecast. Note that a recent spike is more likely to be incorporated into the forecast as a seasonal or cyclical event. This is where expert knowledge will help you to either trust the data (for example, if a promotional event explains the one-time spike) or remove the spike. Investigation is needed to determine what caused this spike and whether it was indeed erroneous or potentially recurring. This is where your expert knowledge of your specific business and use of the forecast is important because the forecast is an aid to making better decisions, but never a substitute for expert knowledge.

As mentioned earlier, in a future post we’ll show you a few simple ways to remove your data spikes with Excel before creating forecasts in SkuBrain. So stay tuned, and in the meantime, always look at your data, at the resulting forecast, and let us know if you have questions. We want you to get value from SkuBrain and to become a “Power Forecaster”.

SkuBrain is free to use for 7 days. Sign up here and start forecasting!

At SkuBrain we’ve built integrations to the Order Management Systems (OMS) which are popular with our client base – these include Vend, Shopify and Brightpearl. With the integrations, users of these platforms can easily import their sales history into SkuBrain without resorting to data extraction and upload using CSV files. However, what if you’re using a platform but we haven’t added support for it? No worries – Zapier probably has.

You can think of Zapier as an online ’glue’ that helps you mash your cloud apps so that they work better together. Below, we outline how you can use Zapier* to connect BigCommerce, a popular e-commerce platform, to SkuBrain. The concepts described here are not limited to BigCommerce – Zapier supports over 500 different connectors to systems such as Xero, Magento, Unleashed, and many more.

Setting up a Zap (basically, a blueprint for a task you want to do repeatedly) will take less than half an hour, and will save you many hours or even days of work in the future.

The overall concept is simple – push your data from BigCommerce into the SkuBrain staging area via Zapier.

Get your own dedicated staging area

Assuming you’ve already got a SkuBrain project, the first step is to get your own dedicated staging area by contacting us at

The SkuBrain staging area is hosted in the Microsoft Azure SQL platform, and you’ll get connection details, like this:

We’ve provisioned a SkuBrain staging area for you:
Database: halosource01
Username: bigcommerce01
Password: **********
Your Table Identifier (GUID): zapier-231
Sales Table: [zapier-231].[Sales]
Stock Table: [zapier-231].[Stock]
Connection String:,1433;Database=halosource01;
User ID=bigcommerce01@skubrainsql;Password=**********;Encrypt=True;
TrustServerCertificate=False;Connection Timeout=30;

Create a zap

The next thing to do is to sign up to Zapier (if you haven’t already) and create a zap that pushes your BigCommerce data into Azure SQL:

In the zap above, Step 1 is a trigger that responds to a new order being created in BigCommerce. When this happens, Step 2 inserts a record in the SkuBrain staging area:

Set up SkuBrain to import from Azure SQL

The last step is to connect your SkuBrain account to the staging area where your BigCommerce orders are flowing. You can do this by using the Azure SQL connector in SkuBrain.

That’s it – instant connectivity to your Order Management System without the hassle of CSV. Once your zap is activated, it runs every five minutes, so your BigCommerce orders will be flowing into SkuBrain by themselves – you can just relax and put it out of your mind.

Whenever you need to create a replenishment plan, simply log in to SkuBrain and use the “Import from Azure” feature to refresh data in your SkuBrain project and forecast your way to retail domination!

If you’d like to set this up but need assistance, don’t hesitate to reach out to – we’d love to help.

SkuBrain is free to use for 7 days. Sign up here and start forecasting!

*SkuBrain is not affiliated to Zapier. Use of Zapier will incur additional charges.

Yes, another blog post that begins by acknowledging the terrifying landscape of advanced analytic technologies on offer to businesses trying to make strategic choices, invest in key tools to gain new insight, cut costs, drive growth, remain competitive, improve the bottom line… Here’s the thing – you can read and research until the cows come home. The fact is, you need to take the first step, try something new, break out of the ‘how do I know where to start?’ or ‘my way is best’ mindset, and just get cracking.

Here’s a great article from Supply Chain Brief by Supply Chain Insights founder Lora Cecere that quantifies the difference advanced inventory planning tools* are making to successful companies. Supply Chain Insights’ research conducted in 2015 compared ‘company maturity on meeting inventory targets to choices on inventory technologies’. Those businesses using advanced demand planning tools – ‘advanced’ meaning tools specifically designed for this purpose, and not spreadsheets or bolt-on solutions to existing ERPs – were almost twice as satisfied with the advanced software than those using basic programs, and were seeing 63% ROI (over a time period that allowed for complete implementation and intelligent analysis). The research also illustrated that adhering to recommendations made by inventory planning software was key to improving margins and driving business growth.

While you’re there, check out Supply Chain Insights’ 2016 Supply Chains to Admire for more inspiration to take the first step. An advanced planning and inventory optimization solution like SkuBrain will have you creating demand forecasts and stock replenishment plans in a matter of minutes. The time is now.

SkuBrain is free to use for 7 days. Sign up here and start forecasting!

Muses from The Lazy Statistician: How to use general analytic methods to solve specific analytical problems

The lazy statistician loves to adapt general frameworks that have broad application to a specific analytical problem. Why, you ask? Simple: a general framework is one where the assumptions are well understood, the tolerances for messy data are understood, and the framework usually delivers 90-95% of the value that a newer or more exotic analytical framework can provide. But the general framework can be set up much faster – call it the 90/10 rule – get 90 percent of the value with 10 percent of the effort when you use a general analytical method with broad application.

My current favorite “general framework with broad application” is SkuBrain, a supply chain management forecasting tool that rocks when it comes to creating SKU-level forecasts and order plans. We at Halo have “pivoted” SkuBrain to perform general ledger budget forecasting.

This “pivot” was a weekend exercise that required no mods or development – it was a simple matter of identifying a few constraints, working through those constraints by way of data prep, and finding that data prep was the fastest route to using a general framework with broad application. BTW, that is a core truth – when using a general framework with broad application, you may need to transpose your data. Fortunately at Halo, that’s easy.

Any other lazy statisticians out there who want to adapt SkuBrain for life sciences research and forecast outcomes at the level of the genotype? Simply array your data at the level of analysis that fits your needs. This could be down to patient-level forecasting of dose response curves based on day-over-day blood assays. Or discovering disease sub-types by grouping cases on other dimensions such as family history or responsivity to previous treatment protocols, and building sub-type forecasts that can be compared to actual outcomes. Pharmacosurveillance with SkuBrain – I like that.

How about financial services and creating branch-specific forecasts for bank deposit growth trends? Why not drill deeper, into branch representative-level forecasting and analysis of sales and customer satisfaction trends? Create an objective, data-driven assessment of where client-facing sales associates are exceeding todays goals, controlling for past trends. Then roll that report up to the branch and region, or dis-aggregate to the level of product – this is what SkuBrain plus a little data munging can do. Get a free trial of SkuBrain and see how easy to use it is, and get lazy.

SkuBrain is free to use for 7 days. Sign up here and start forecasting!

It’s that time again – the financial year is drawing to a close and departmental budgets need to be revised for the next year. Expenses need to be forecasted, salaries adjusted by a few percent, revenue predicted and everything mashed together into something sensible that the CFO will hopefully approve. If only we had a forecasting tool. Wait a minute – We do! SkuBrain! Yes, that’s right. You may be mistaken thinking that SkuBrain is only for inventory management. It can forecast just about any time series! In this blog I’ll share with you exactly how to do that.

First, you’ll need to get a CSV extract from your financial or online accounting package. Basically we want all income and expenses, by date. The fields you will need are:

Account Code The GL code
Account The meaning of the code
Amount The amount of spend or income
Date The date of the transaction
Department The department responsible

If you have other field that you’d like to group the result by, you can add those too – e.g. Department, Cost Center, Region, and so on. Next, we’ll need to do some Excel-based data prep to get it into a form that SkuBrain can work with. The first two steps are easy:

  1. Create a new field called “SKU” that is composed of the GL and GL Code.

  2. Rename the Date to “OrderDate”
  3. Rename the Amount to “Quantity”

Before we do the next data prep steps, some explanation is in order: SkuBrain cannot forecast negative quantities, but the output of the ledger systems are often positive for expenses and negative for income (or vice versa, depending on which side of the ledger you are looking at). So we’ll need a workaround for SkuBrain. Fortunately, there is a simple one – the UnitPrice field. Remember that in SkuBrain, your Sales are calculated using the formula : Sales = Quantity x UnitPrice. Therefore:

  1. Create the UnitPrice column which has 1.0 for income items, and -1.0 for expense items. (In my data, expenses were negative. If your data has it the other way around, just flip the logic)
  2. Then, reverse the sign of expense items.

BEFORE (expense item has negative quantity)

Quantity UnitPrice

AFTER (expense item has positive quantity, but negative Unit Price)

Quantity UnitPrice
42.53 -1
30.89 1

Your input data is almost ready. The last thing to do is add a couple of required columns: LineReference and OrderReference. They don’t need to contain any values - SkuBrain will just fill them in automatically if they are blank. When you’re done, your data will look like this (some extra fields are shown for me, but they are not mandatory).

Now, export your XLSX file into a CSV file and upload that to SkuBrain as usual. Mine had 55,000 rows and when imported into SkuBrain, looked like this:

Now, its time to do some Expense and Income forecasting! Since I have a “whole-of-company” extract, and I’m interested in departmental performance, I start with a forecast that has a two-level hierarchy “Department > SKU”. Here are my forecast settings:

I start the job, and a sip of coffee later, we have a forecast! Here’s what expenses of my Distribution department look like (The “$” value is negative because expenses have a UnitPrice of -1.0 in my data.)

In contrast, my Sales Department forecast is positive (since revenue is recorded there with UnitPrice = +1.0)

And the SUM of everything (“All sales”) is essentially my EBIT, forecasted for the next year (we’re not looking too bad)!

I also know that we’re planning a major digital campaign in the coming year, so we’ll just make sure our Marketing expense forecast reflects that.

Indeed, I could go further and create a forecast that breaks down my finances by cost center as well. We’ll need to let Head Office know that their IT team needs to be reined in:

So, there you have it – a financial forecast, with adjustments, to back up my budget for next year, done on a per-GL code basis that’ll have my CFO wanting a SkuBrain account for herself. To feed the forecasted numbers back into Excel, all I have to do is download the data (More → Download) as a CSV file. I’m inspired now. Where can I get some decent historical data for index funds, I wonder…

SkuBrain is free to use for 7 days. Sign up here and start forecasting!

Izze’s Amp Warehouse

Several years ago, I started a small retail business providing guitar amplifiers and effects pedals to musicians. Izze’s Amp Warehouse starred my English Bulldog who bullied suppliers for the best prices. After some initial success, we established a demand plan to support our e-commerce site and three locations in California. We forecasted our sales at a company level then allocated targets to each location. Can you guess what happened?

Our sales forecast was actually pretty decent, statistically speaking. But, we ran into problems with stock levels. The fact was, our model predicted total future sales for Amp Warehouse – but did nothing to account for current stock levels or store lead times. I will finish this chapter of the story by noting that by the time we should have started on a different demand planning approach, Amazon caused us to rethink our commitment to the business.

But, for me it’s a good place to kick-off a discussion about how to choose a demand planning approach. In fact, you have two basic options:

  • 1. Top Down

  • 2. Bottom Up

Most experts consider the Bottom Up approach to be best today – but, the vast majority of retailers still practice Top Down forecasting. Why? Let’s consider the options.

Top Down Forecasting

The basic idea of Top Down Forecasting is to generate a total sales volume forecast over a time horizon for the business – then “disaggregate” the volume by sharing it out to each store. The top down approach is OK when you have little or no sales history and you want your corporate headquarters to control the prediction of your growth.

Assuming you have some experience with sales at each location, you can also make some good estimates about the percentage to allocate to each store. So, its not a bad approach for an executive strategic plan. But, Top Down, doesn’t deal with stock levels or lead times. Thus, you can forecast sales but its much harder to optimize inventory at a local level. That was the core problem at Izze’s Amp Warehouse.

Top Down forecasting is more in-grained in business processes and there is a traditional approach to applying it. Basically, local stores take the “target” number and then apply “Min/Max” rules that indicate when to reduce inventory and when to re-order. The problem with these approaches is they tend to be reactive. Introducing new products, dealing with seasonality and local demographic preferences across store trade areas make it hard to optimize inventory.

Bottom Up

The bottom up approach creates a forecast at the lowest product level (e.g., SKU or prduct), then “rolls” that estimate along with every other product to the next level in the product hierarchy. Thus, you are assured that your top level forecast will align with your results. Bottom up forecasting gathers individual forecasts from each store and applies those predictions to drive replenishment choices.

As such, bottom up has more tactical value. It can be used to understand and improve customer profitability, to improve customer satisfaction at the store level, and to do so while minimizing local inventory.

By managing inventory at store level, you are gaining a level of granularity that will help you optimize your working capital and ensure you meet your desired service levels. More specifically, you will be better prepared with address varying sales patterns, such as caused by seasonality in demand or promotional activities. A quick increase in sales can lead to “stock outs” while a sudden drop will may mean you are over-stocked. And both situations will reduce the profitability of the business.

Once you’ve implemented a bottom up approach, you will no longer need to forecast at the distribution center level – its simply the summation of your retail locations. And, the challenges of breaking down a Top Down, corporate forecast to be used by your production team for capacity planning is reduced. Both sales and production are now looking at product level unit forecasts.

To Summarize

SkuBrain allows both top down and bottom up forecasts. We recommend:

  • A top down forecast for new businesses or where you lack at least 2 years of sales history
  • A bottom up approach when you are focused on reducing inventory and working capital – and have a few years of data to work from.

SkuBrain is free to use for 7 days. Sign up here and start forecasting!

Keith Peterson

Keith Peterson is the President and CEO of Halo. Keith assists clients in developing business strategies that are enhanced through data, analytics, and technology.
Follow Keith on Twitter for the latest on Supply Chain Analytics and Business Intelligence at @KPeteHalo

While SkuBrain may have started in the retail space, it has quickly gained popularity in beverage manufacturing and distribution forecasting. Beverage suppliers and distributors are using SkuBrain to help them apply the five Koan of beverage forecasting:

  1. Simplify your models – trade a little predictive power for a better long-term view of your business (and have an easier time explaining your forecasts to management)
  2. More data is not always the best idea – use categories wisely and you’ll understand your data better
  3. Forecasting is not Time Series Analysis - Forecasting is what you do with time series analysis to make the best decisions
  4. ‘Forecastability’, not accuracy – your accuracy is only as good as the predictability in your data. Remember, forecasts are just a decision support tool.
  5. The machine can’t do all the work – SkuBrain is very good at forecasting, but add your judgement and you’ll have a winner

To read the full article, visit

Acquisition Extends Company’s Reach and Bolsters Presence in Retail

San Diego, CA (PR WEB) February 4, 2015. Halo, provider of self-service supply chain intelligence geared to the mid-market and with a focus primarily on manufacturing, distribution and retail, today announced the acquisition of all assets of New Zealand-based SkuBrain. SkuBrain offers Software-as-a-Service (SaaS) for supply chain planning for small to medium-sized retailers. The acquisition extends Halo’s reach and also bolsters Halo’s presence in retail. Halo will migrate all technical and marketing assets of SkuBrain into the Halo environment to create a single platform that can be scaled to the needs of companies with revenue from US$50M to US$5B.

“Adding SkuBrain to our product portfolio is an excellent complement to the Halo Supply Chain Intelligence platform.” Keith Peterson, President & CEO, Halo.

The SkuBrain software was designed specifically to handle retail demand forecasting, inventory optimization, and replenishment planning. The founders, veterans of retail industry, built SkuBrain to address the significant gap in planning capabilities, and thus competitiveness, between small and large businesses. In particular, smaller businesses need simplified tools that favor ease of use and straightforward reporting. Halo’s acquisition of SkuBrain comes with a portfolio of customers who will be maintained on the existing platform for the near-term.

Keith Peterson, President and CEO of Halo, stated, “Adding SkuBrain to our product portfolio is an excellent complement to the Halo Supply Chain Intelligence platform. Smaller firms are often too small to leverage the enterprise capabilities of existing planning platforms. The SkuBrain team saw the gap and has filled it with a highly targeted solution. Customers can now start with a subscription-based SkuBrain solution and seamlessly transition to the Halo platform as their needs grow. We don’t know of another solution in the market that makes this possible.”

About Halo Halo delivers self-service supply chain intelligence solutions to hundreds of enterprise customers and service provider partners in North America, Europe, and Asia-Pacific. Our solutions provide enterprise grade products without the drag and cost of major platforms. Halo replaces the complexity of siloed data sources and disparate applications with a single purpose platform that lets firms analyze, decide and plan faster than ever before. Halo is headquartered in San Diego, California, and can be reached via the web (, Twitter (@Halo_BI) or email at

If you don’t have any sales history then it can be pretty tricky to plan inventory. However new businesses, businesses that have just migrated to a new ERP system or existing businesses that introduce new product lines will, inevitably, have to create forecasts at some point for products with little to no sales history.

SkuBrain, of course, allows you to manually adjust forecasts for any of the products that you sell. So if the statistical forecast that SkuBrain comes up with is 0 units (which would be pretty typical for products without any sales history) you can always manually override this with some estimates that you’ve come up with after talking to suppliers or some other kind of market research.

Previously, such manual adjustments would have been required for any product that had less than one month of sales history since, until a month is fully completed, SkuBrain considers it part of the “forecast horizon” rather than the “sales history” and most statistical forecasting algorithms only use sales history in order to generate forecasts.

However now SkuBrain includes a handy feature called extrapolated forecasts, which lets it create ballpark forecasts for products, even when you only have a couple of days worth of sales history for these.

Extrapolated Forecasts

Extrapolation is super simple. Essentially, if you create a forecast on October 13th for a product that you’ve only been selling since October 1st, SkuBrain extrapolates those early sales figures out so that the estimated sales for the full month of October (for that SKU) will be:

    Forecasted Sales  = Sales to Date * Days in Month / Days Elapsed
                      = Sales to Date * 31 / 13

The result is non-zero forecasts even for products with no “sales history” (in strict forecasting terms).

extrapolated forecast

The above forecast was generated on October 13th… so roughly one third of the way through the month of October. The projected sales for the full month are therefore roughly 3 times (31/13 times to be precise) sales to date. This is probably pretty much the process you’d use if you were going to generate these forecasts manually.

Hopefully this helps make things smoother for folks that are just starting out or anyone who often needs to forecast new products!