What are the key challenges and best practices of revenue forecasting in a subscription-based service?

Best practice for revenue recognition would be aligning RevOps and Finance. RevOps can own the systems, the playbooks, the tools but finance typically is the one reporting the recognized revenue. Commission can be supported by either team but is something else to consider in this topic. I found it critical that RevOps considers the needs of Finance when developing the operations plans to ensure there is alignment with one source of truth. For example, sit down with Finance on a quarterly or yearly basis to review assumptions about how various scenarios are being handled - annual deals, multi-year deals, ramp deals, so on. Ensure the playbook works for both teams and review any underlying assumptions.

Revenue forecasting and revenue recognition are distinct concepts, yet both require consideration of the business model within which an organization operates. Understanding the revenue drivers is crucial for both processes. Thus, revenue recognition and revenue forecasting should both be based on revenue drivers.

Developing a strong revenue model is crucial for subscription-based services. Below a few pitfalls I've witnessed over the years: 1. The finance model not being aligned with bottom-up forecast for both new and existing accounts. 2. Stage conversion rates and win rates being wrong as the sales process isn't strictly followed by the sales team. 3. Poor visibility into pipeline health as CRM data and deal signals are missing to understand likelihood to close. 4. Poor understanding of churn as account health isn't based on product-usage data, and the service model isn't deep enough. 5. KPIs across Finance and RevOps not being aligned, leading to differences in assumptions for the revenue model.

Here's how to model out your subscription business. In your initial model, have a tab for all assumptions. Include the following categories: - Pricing plan - Operating expenses ** include ALL operating expenses that impact subscriber lifecycle in this table** - Churn % schedule - Subscriber ramp schedule - Refund/ chargeback % - Capital investment Those assumptions will drive the forecast model. Your forecast should also include a 3-year projected cash flow. It should also include a tab for the operating P&L. Most importantly, the model should highlight what your estimated KPI's are, ex: - LTV - LTV: CAC - Payback - IRR - Net revenue per subscriber - # of subscribers by month/ year

In subscription-based services, developing a robust inventory model is crucial. This model should focus on managing growth and churn, while also accounting for startup costs. Analyzing historical customer behavior is vital for forecasting growth accurately. It's essential to prioritize active users over total subscriber count, as active users are more likely to contribute to sustained revenue. The model should aim for a balanced end-of-month inventory, considering both new acquisitions and customer attrition. This balance is critical for making informed pricing decisions, ensuring the service remains competitive and profitable.

Net-revenue-retention (NRR) and churn rate are key factors to drive growth. It's best practice to track churn rate on a monthly basis and it's a KPI investors care deeply about. While it's fairly easy to track historic churn rates, prediction of churn rates is a lot harder. For many SaaS companies, churn rates increase steeply in 2023 given the focus on efficiency and profitability. Benchmarks for net-revenue retention (NRR) are between 110-160% in SaaS.

Finance should drive the methodology and Rev Ops should own the reporting and communication. It is important to align on the methodology and not change/tweak or expand it at the request of the business or the sales organization. In addition to additional work cycles, this results in confusion and a "pick the metric that best suits you" environment. Keep it simple and straightforward as possible so it is easily understood by all stakeholders.

- Churn can be actively who decide to cancel their subscriptions due to dissatisfaction, cost concerns, which is hard to avoid at some times. Another type of churn such as expired credit cards, failed payments, or changes in personal circumstances should be taken care of easily. -Customer Lifetime Value (CLV) would plays important role here. by estimating CLV, can prioritize retention efforts on high-value customers who contribute the most to long-term revenue.

Key things to understand about churn: 1. Churn isn’t linear. Customers are most likely to churn earlier in the subscriber lifecycle, ex: during the trial period, or first month. 2. B2C subscription businesses have higher customer churn than B2B businesses.  3. Voluntary vs. Involuntary churn. Voluntary churn indicates customer dissatisfaction, while involuntary churn points to payment issues. Solutions - voluntary churn: * Focus on strengthening value proposition
 * Get clarity on billing practices * Offer more consumer controls
 Solutions - involuntary churn: * Automatic cc updater tools 
 * Retry transactions post failure
 * Notify subscriber when their card declines (dunning) * Offer alternative payment methods (ex: Paypal)

Customer Lifetime Value (CLV) should be treated more as a guiding output in revenue forecasting rather than a primary input. CLV, as a reflective measure, gauges the effectiveness of a business's strategies in customer acquisition, retention, and value maximization. Using CLV as an output rather than an input encourages businesses to focus on current operational efficiencies and customer engagement, rather than overly relying on projected long-term values. Essentially, CLV as an output offers a reality check, keeping business strategies aligned with actual performance and market trends.

I'm a big fan of base, best-, and worst-case scenarios. A few more notes: 1. 12-months revenue plan should run different scenarios that should align with the hiring/ cost plan. It's crucial to start early to ensure the revenue capacity plan for pipeline generation and conversation can be met given the time-to-hire, ramp-time, and quota attainment assumptions. 2. The quota goal should be larger than your attainment plan which should yield your best case revenue plan. That ways, you have multiple layers to against your revenue goal. 3. Running a current and next quarter bottom-up forecast process for both new and existing business should yield insights into target achievement and allow for hiring / investment decisions to be adjusted.

In revenue forecasting for subscription services, using Monte Carlo simulations with thousands of iterations can be overly complex and processing-intensive. I believe a simpler approach is more effective: focusing on best, base, and worst-case scenarios with a few key variables. This method provides clear, actionable insights without overcomplicating the analysis. It aligns with the basic revenue principle of price multiplied by quantity and is more accessible for stakeholders. Simplified scenario planning offers the necessary foresight for strategic decisions without the burden of processing-heavy simulations, making it a pragmatic choice for effective revenue forecasting.

I would add the challenges we face : The Impact of marketing & sales initiatives: Often witnessing the New company or the New products gets implemented too frequently, Hence it's getting too difficult to consider or introduce uncertainty into the forecast due to the New marketing campaigns or pricing changes... Most important Regulatory changes: like what just happened with Paytm in India, New regulations or industry shifts can disrupt the market and impact growth expectations.

I would recommend looking into Cohort Analysis: Cohort analysis is a powerful tool for understanding customer behavior, which can significantly impact revenue forecasting. For example, if your data shows that users who haven't signed in to their accounts for three months are likely to leave your account, you can create a cohort with those users. This allows you to predict your probability of churn so you can adjust your forecast accordingly.