How can you improve the accuracy of your corporate FP&A forecasts?

1. Understanding of current & historical P&L including budget. 2. Understanding picture of business from business folks (sales forecast, cost improvements etc) 3. Use any industry data available on Drivers of sales or cost. 4. Multiple scenario analysis along with analysing sensetivity of factors considered. 5. Understanding of cost structure (fixed , variable) if variable, the factor on which it's variable

Driver-based forecasts are a must today if you want to create something that’s accurate and flexible when you want to make changes.

I had once the opportunity to work (before it was mainstream) on a machine learning statistical model of forecast. Engineers were so convinced that the more drivers you used, the less accurate your model was, that they finally built that only on revenues, no volumes. Guess were that model is today (Rest in Peace, model)

When I work with startups, I often remind them of the importance of truly understanding their fundamental business drivers like revenue and costs. It's easy to get lost in the details, but grounding forecasts in these core elements can make all the difference.

Taking a 'business-first' approach to building forecasts can transform the quality of your forecasts. By identifying the key business drivers and activities that influence the financials, you can build a financial forecast that can can be tied back to a business reality or context that those around the business can understand. Powerful stuff when you get it right!

Actually this article fully describes the responsibility of an FP&A role, let it be full company scope or part of the financials. In case the above is not met, the forecast will unfortunately be inaccurate. The forecast is not just a trendline, should be a vision composed after the business insights.

Historical data serves as the foundation, but with SaaS businesses, the rapid pace of change means relying solely on the past can be misleading. Leveraging advanced analytics and predictive modeling techniques can harness this data to fine-tune forecasts. Additionally, integrating feedback loops into the forecasting process allows for continual refinement. Regularly comparing forecasts to actuals and understanding the deviations can provide invaluable learning. Collaboration across departments, especially those with client-facing roles, offers insights into potential market shifts or customer behavior changes. Ultimately, the fusion of data-driven analysis with cross-functional input can significantly bolster the accuracy and reliability.

Frequent Interaction with business leads and teams, better understanding of the business, benchmarking with historical data and competitor data to uncover any trends, strategy shifts, opportunities helps in reassessing the changes in assumptions and drivers used in the forecasts.

Identifying the right business drivers is key for an efficient and effective forecasting. This driver changes frequently, thereby adapting a change management mindset is pivotal for projecting the growth of the business.

The forecasting model should be granular as much as possible, in order to accommodate as much possible seasonality and variability of the business. Secondly to the possible extent the most latest information available should be used as the basis. Thirdly the key drivers, ratios, allocations, proportions should be visited and updated regularly

The only thing you can be certain about your forecast is that it’ll be wrong. Consider different scenarios and have plans in place to adjust your course of action when needed.

Utilizing multiple scenarios and sensitivity analysis indeed offers a holistic view of potential outcomes. By considering the best, base, and worst-case scenarios, businesses can strategically anticipate and mitigate risks. Sensitivity analysis, in particular, sheds light on which drivers have the most profound impact, enabling better-informed decision-making and robust strategic planning.

B2B SaaS companies, with their unique sales cycles, churn rates, and revenue recognition patterns, require specialized attention in forecasting. To bolster accuracy, consider the recurring nature of SaaS revenues. Understand the nuances of customer acquisition costs (CAC) in relation to the lifetime value (LTV) of clients. Integrate your CRM and financial systems to tap into sales pipeline data, using it to forecast new business. Incorporating a rolling forecast approach, where predictions adjust regularly, keeps your forecast aligned with recent trends. Engage closely with sales and customer success teams; they provide valuable front-line insights on customer behavior and market dynamics.

Using multiple scenarios can be helpful to a certain point, but it also can be very time consuming. Always have a hedge and use the 80-20 rule, while also keeping the bird view in mind.

A forecast is uncertain by its nature, logical by data and result driven over a time. Integrity and flexibility of the analysis is important for a review mechanism.

Enhance forecasting by using multiple scenarios and sensitivity analysis. Scenarios, like best, worst, and base cases, uncover risks and opportunities. Sensitivity analysis quantifies the impact of changing assumptions. This fosters robust, adaptable forecasts, aids in contingency planning, and optimizes resource allocation. Improved communication, enhanced decision-making, and adaptability in a dynamic business environment are key benefits.

Getting it right with scenario analysis can make it such an important part of the decision-making toolkit! But don't build it in isolation - include the end-users of the information and understand how they need to use the information to make decisions. From here, FP&A can really advise on the best approach to scenario analysis for the business!

Participation of people from different teams within the company helps to challenge the assumptions and arrive to a more realistic model.

Feedback is always welcomed - either it be good or bad. And most of the times, when there is constructive feedback is received that is better than a simple "this looks great" - as it can really drive improvements.

A great forecast is also a collaborative one yet too often it happens in silos. Hence, involve all relevant departments to make it a good one.

By inviting diverse perspectives, we may discover overlooked insights and build a stronger sense of community and trust within the organization.

I agree, inclusivity is key as the way to Involve key stakeholders, such as sales, marketing, and operations, in the forecasting process. This will help to ensure that your forecasts are realistic and achievable. It is also important to get buy-in from stakeholders early on in the forecasting process. This will help to ensure that everyone understands the goals and objectives of the forecast, and that everyone is on the same page.

A good forecast needs a legit team contribution. Having the right information and frequency of reviews will lead to an accurate forecasting path.

Feedback Loop: Continuously compare forecasts to actual results, learn from errors, and refine assumptions. Collaboration: Engage cross-functional teams and experts for diverse insights, holistic views, and alignment with organizational goals. Learning: Apply lessons from successful and failed forecasts to enhance future predictions. Adaptability: Stay flexible by incorporating real-time data and insights for agile decision-making.

Creating a collaborative feedback process with all the key stakeholders can scale up the quality and impact that scenario planning has in the business. Ultimately, well designed scenario planning can help improve the quality of decision-making, so it is vital to work with those who will be relying on the information to make decisions!

Forecasts serve as ongoing assessments of your progress and provide guidance for course corrections throughout the year. A regularly updated forecast helps address key questions such as: - Where are we currently in terms of our goals? - Where do we need to be to meet our objectives? - What strategies must we employ to reach our targets? It is essential to address these questions effectively and to ensure that the business remains on the right trajectory.

If you have a good driver based forecast you can update it as frequently as you want. Shouldn’t take more than a week max to create an updated forecast.

In the B2B SaaS field, understanding contract renewals, churn rates, and expansion opportunities becomes paramount. Regularly revisiting your forecast allows you to adjust for evolving customer behavior and market conditions. Consider automating data flows between your CRM, billing system, and forecasting tools. This ensures real-time data reflecting customer acquisition, renewals, upsells, and attritions is consistently fed into your forecasting models. Engage sales and customer success teams, as their direct interactions with clients provide invaluable feedback on market sentiment and potential risks. Furthermore, by maintaining an ongoing dialogue with other departments, you can integrate operational insights.

Regularly revising forecasts keeps us grounded, adaptable, and ensures we're making choices that resonate with the current realities. It's about staying responsive and connected to the evolving story of our business.

In marketplace we understand there is no such a things like certainty. FPA should update their forecast regularly to revisit their mathematical model and ensure the relevancies of the linear relationship.

Macro and micro environments in which the company operates are must to consider. Reassessment of leaks/assumptions that are constantly failing the forecast (i.e. SBUs/users/departments/activities which are trouble) will help us in ensuring the integrity of the forecast.

You can also use forecasting error analysis to identify the sources of forecasting errors and to develop strategies for improving your forecasting accuracy.

Continuous Updates: Regularly revise forecasts to capture changing business dynamics, ensuring they remain current and reliable. Data Refresh: Keep datasets up-to-date for more accurate projections. Scenario Adaptation: Adjust scenarios to align with evolving circumstances and be better prepared for contingencies. Strategic Agility: Embrace adaptability to respond quickly to market shifts and opportunities. Stakeholder Confidence: Build trust and confidence by demonstrating a commitment to transparency and accuracy through regular updates. Alignment with Goals: Ensure forecasts support evolving strategic objectives and resource allocation decisions.

Yes. For instance, in industries such as Real Estate and Hospitality, we have seen that forecasting can be particularly sensitive to macroeconomic and operational factors : Macroeconomic shifts, geopolitical events, and changes in consumer sentiment can significantly impact the performance of businesses. As such, updating and revising forecasts regularly is crucial in these industries.

Having good tools make it a lot easier but you can create a driver based model in Excel even. Not advisable to run your company on Excel but you can build there first and then implement a tool afterwards.

ERP systems are the boring, but essential backbone of every business. They're like the plumbing in your house. You don't think about them much, but when they're not working, you're in a shitload of trouble. BI tools are the sexy new things on the block. They're like the Instagram filters of the business world. They can make even the most mundane data look pretty and engaging. Imagine you're a CEO and you're about to give a presentation to your board of directors. You need to explain how your company is performing and what your plans are for the future. You could just give them a bunch of boring financial reports from your ERP system. But that's not going to capture their attention.

In my practice, Forecasting software can help you to automate the forecasting process and to improve the accuracy of your forecasts. Forecasting software can also help you to track your forecasting performance over time and adjust accordingly & according to your business needs as well.

From my point of view data visualization is essential to make financials understandable to stakeholders outside the finance room. I've never seen a stakeholder saying that give me a report or a large spreadsheet instead of a chart. The beauty is to choose and work with right visual that concentrates on the problem and represents the data compactly and comprehensively at the same time. Not less important is to use visualization at your own work - it gives fast insights to historic context, recognize trends and drives your attention to relevant. First thing I do to analyze large unfamiliar data is to generate some fast charts out of it to get an idea what's going on, what is the context.

Even the most advanced tools and methods can lead to inaccurate forecasts if not used properly. It's crucial that teams are well-trained and understand the intricacies of the tools and methods they're employing. Additionally, continuous training ensures that as tools and methodologies evolve, the personnel remains adept at utilizing them to their full potential.

Within the B2B SaaS industry, where MRR and ARR are vital metrics, selecting the right tools and methods becomes critical. Leveraging dedicated SaaS forecasting platforms can streamline data from various channels, such as sales funnels, churn rates, and upsells. Additionally, consider advanced statistical methods or machine learning algorithms that can capture patterns in your customer behaviors more accurately than traditional methods. As SaaS business models are highly scalable yet equally volatile, employing tools with real-time analytics and predictive capabilities can give you the edge. Engaging in periodic tool & method training sessions also ensures that your team remains up-to-date.

Building an auto or semi-auto tool based (with AI and machine learning if possible) on historical data from ERPs to supplement the forecasts and to make the forecast foolproof is the need of the hour.

Documenting steps during the setup of the process would help in identifying the right tools and methods. A good understanding of interpretation of laws and regulations will be key for audit compliance.

Efficiency and Effectiveness: Select tools and methods that streamline data handling, reduce errors, and improve efficiency. Transparency and Consistency: Opt for tools and methods that enhance transparency, communication, and consistency in the forecasting process. Customization to Context: Tailor your choice to your organization's unique context and needs. Scalability and Integration: Ensure tools can scale with your organization's growth and integrate diverse data sources seamlessly. Cost-Effectiveness: Evaluate the cost of tools and methods in relation to the benefits they offer.

Einstein said “If you can’t explain it, you don’t understand it well enough.” - which basically is also the gold rule on creating any report or analysis. Create it having in mind someone that does not know what you did there, and still manages to understand it. This will also help create a visible audit trail for the future.

Once you have a clear understanding of the key business drivers and the different scenarios were the company can operate, next important task is to summarize the findings in a dashboard that helps to understand the current situation and the different choices the company has to make an informed decision.

Infographics for jargon have helped in my current role. At times finding an innovative way to simplify a step or a process is within it, look for clues from existing tasks.

Business knowledge and understanding the operating model of the company are critical for a finance professional to be able to build an accurate future financial plan. Engaging different stakeholders in the process and cooperating on a daily basis keeps you informed about forecast updates needed. Variance analysis between actual performance and forecast is also important for improving the accuracy and understanding of the drivers behind. Data quality of the historical information is also important to build a correct run rate.

One of the biggest challenges I hear from business leaders on scenario planning is that it 'takes too long' to run scenarios. By the time the information gets updated in models, it is out of date. Model design plays a critical role in getting this right, and the value gain from getting right can be hugely significant. It can really give FP&A teams a seat at the table.

Start with understanding how your accounting works. Ultimately, you will measure and refine your forecast against actuals. Analyze the components of the FS lines that you'll be forecasting. Develop indicators of their behavior. Research industry trends, understand corporate plans, then apply that information to your forecast. Separately, there is a budget process with it's own assumptions that you can compare your forecast against. A forecast is what you expect will happen, rather than a limitation of what you are authorized to spend (budget) or actuals (accounting). Use that perspective to stay focused on the outcome.