How to Build a Sales Forecast to Supercharge Your Startup’s Growth [free template]

There’s power in forecasting. It can supercharge your business. Help you to plan. Expect the unexpected. And most important, accelerate your startup’s growth. In this article, I’ll give you some of the basics, walk you through an example, and give you a free template to get you started. It’s a serious read, but trust me, it’ll be worth it.

Play along: get a copy of the model here.



Wait, what is forecasting?

Forecasting is looking ahead. It’s the activity of estimating future performance indicators in your business. At its heart should be your one driving metric: usually sales or users. From this, you can predict other important factors like revenue, cost and the cash flow you’ll need to survive.

I’m going to show you how to use ‘top down’ and ‘bottom up’ methods to produce a solid plan for growth.

A little scenario.


You’re the proud CEO of LegWork. It’s the Uber of dog walking: customers can simply open up the app, and request a walk from your favourite walkers.

You’ve already found some traction. You’re running a beta trial in London, where you have 500 users, who pay £15 for a ½ hour walk — that’s £5 more than the average basic service around town. You’ve decided that your primary market is the UK, but you’ll look to expand later on. You’ve raise a seed round of £250,000 that’s allowed you to take on a small team, mainly to develop the app and provide customer service.

You’ve got the basics down, as your investors required them, but feel that there might be opportunity in really understanding the market.

How to size a market using a top down analysis


Top down forecasting produces a rough estimate. You start this by looking for a total addressable market. Narrow it to the market that you could really service. Then decide on the percentage of that market that you can actually obtain, considering your constraints and competition. Together this produces what’s known as a ‘TAM, SAM, SOM’.

This is great for a pitch competition, a hackathon, and for bouncing ideas around, but you shouldn’t base a business on this alone, no matter at what stage you are. A top-down sizing relies on good data. If you’re funded, you’re likely to have this already. It’s really important to do your research and only use reputable sources.

Here’s a quick sizing for LegWork.


The PFMA estimates a UK Dog population of 8.5m. An American Express study on UK dog owners also unveils that the average UK dog owner spends £1252 on their pooch each year. Multiply that out and it gives you a Total Acquirable Market, a TAM of £10.6bn.


That study breaks down the spend too, uncovering that dog owners spend an average of £122.16 on dog walking each year. If an average dog walk costs £10, that’s 12 walks per year. Or one per month, per dog. Multiply that by 8.5m dogs to get a UK serviceable acquirable market (SAM) of £1bn, or 102m dog walks in each year.


But you can’t hope to capture the whole market, even with a product as incredible as LegWork. Research, and you might find some figures on competitor market share. I didn’t, so we’ll assume 5%. This gives you a Serviceable Obtainable Market of £52m, or roughly 5m dog walks in each year.

But is this useful?

Top-down is quick, and great for a back-of-napkin business model. It can give you the headlines that investors look for, but to get to an obtainable market share, you more often than not have to take a really big guess. This is where the top down approach really falls down.

How to size a market and forecast with a bottom-up analysis


Bottom-up forecasting starts with your customers, and your activity. It’s granular. It’s lengthy. But it can produce great results if you do it right.

It’s calculated by estimating how many products you can sell. It forces you to think about where and how you can sell, and creates practical detail and a forecast you can grow from. Here, I’ll run your through a few fundamentals, then get to the nitty-gritty of the model.

The key to forecasting: your customer.

Our example, LegWork, isn’t just an app, it’s a service. We not only need a beautiful UX, you need to recruit owners and walkers. It’s likely by now you’ve identified the best areas to target (a.k.a. market segments and personas) on a limited budget. For the sake of our example, let’s say it’s professionals under 45 in London.

You’ve been working on building traction (here’s a great book if you want to know more about that) and found that the best channels for this audience and your business are organic searches, via your blog, and old-school London Underground advertising.

Build a model founded on good thinking.

A bottom-up good forecast is based on a solid sales pipeline. The exact model you need will depend on the type of business you run — SaaS models will be different to hardware or service led businesses… but the theory is always the same: put prospects into the top of the funnel, get customers out of the bottom. How good your solution and your sales are will influence how effective this is.

The following model uses Dave McLure’s Pirate Metrics. It provides simple, understandable indicators of what’s happening at each stage of the sales funnel. It’s very well suited to subscription-based SaaS businesses, but can be adapted to give value to a great deal more (that’s why I’ve chosen a hybrid service business to show you).

We’re going to make assumptions. You need to make a lot in any model. The key is to start simple. Do a few things well. Use good tools. Don’t get wrapped up in complexity — the more assumptions you pile on top of each other, the bigger the error you could have.

Play along: get a copy of the model here.

1. Acquisition: The top of the pipeline

Back at LegWork, you’ve honed your product and channels through some lean testing. You even have your first blog content up and running, and your advertising agency queued up. Next step is to identify the drivers of success for each channel.

The key driver for organic blog traffic is the number of articles you have on your site. Every article gains you a number of reads every month. Let’s call it 500 per article, every month. We assume that 50% of those readers click through to the main website.

The key driver for above-the-line Tube advertising is views (sometimes called eyeballs). How many people walk past your advert every day? You’ll be able to get an estimate of this from your agency. Let’s call it 100,000 per campaign per month. From that, we assume that 3% of the Tube Commuters visit your website.

We put this data into the forecast, right at the top, and reference it in the rows below where we work out the pipeline for each channel. Doing it this way means that you can change your assumptions at any point and change the model without fuss.

2. Activation: App downloads

You’ve invested time and money into building a pipeline, and the website has good traffic. The next step is converting those visitors into app-downloaders. This is what Dave deems ‘activation’.

Here, you must estimate conversion rates: the percentage of website users who will download it. For each channel, this might be different. In our model, 2 out of 100 blog readers will download the app, but 5 out 100 advert visitors will download (as they’ve visited the site especially). As always, you can make assumptions, but the only way to be sure is to test it.

Again, we put this into the headlines at the top, and reference them below in the pipeline for each channel.

3. Retention

Churn is the drop off of users you experience every month — who uninstall the app, and stops using LegWork’s service on a regular basis. It’s super-important to any business forecast, but especially relevant to subscription and SaaS models. It’s a lagged variable — that means that you should consider last months drop off in this month’s calculation, subtracting churn from the total acquired that month to get a net figure. For LegWork, we’ve given it a healthy 7% churn rate.

You’ll see how it has been worked out in the model if you click a cell — it takes last month’s downloads and applies a churn rate to it from your headline 7% figure.

4. Referral

Referral can be an incredible driver of growth. It’s the reward you get for a great customer experience.

If LegWork customers love the app and service, they’re likely to recommend it to their friends. Those friends will visit the website, and go through the funnel like anyone else — but because of the recommendation from a trusted source, they’re likely to convert better.

If you’re a digital marketer, or you’ve been running a startup for a while, you’re very likely to have spent time developing growth hacking tactics. These referral techniques create exponential growth — via viral or network effects that produce forecasts that investors love. No wonder: not only is it ‘free’, referral can generate the phenomenal growth that’s been the key to success of many products you know (from Hotmail to Uber and Facebook). Learn more about these effects from the excellent Anu Hariharan via the Y Combinator blog here.

In LegWork’s model we already have the data: let’s say it’s a flat 2% referral rate based on the cumulative users from the month before. This is based on the assumption that the user base will continue to refer their friends and family each month as the service gets better. It’s a key driver of the classic revenue ‘j curve’ that investors like to see. Almost every business that a VC will hear pitch will have one of these graphs, so the most important thing here is to real evidence to back up your claims.

5. Revenue

Now this is a tricky one. How much are LegWork customers prepared to pay each month for the service? How many visits do they need?

There’s only one way to do this. Test the service by getting customers to pay. It’s often overlooked — but there’s little more crucial that finding out whether your customers will pay for your offer (the only exceptions to this really lie in pure SaaS and service/subscription models that build value through network (e.g. Facebook and Amazon’s Kindle). Even at an early stage, building a paid-for MVP will help you to test and understand.


In our example, LegWork has already attracted 500 customers, so we have a benchmark. That from the ~£10 per month that an average owner spends on dog walking services, LegWork acquires 50% of each user’s total spend on this service each month — a total of £5 per user per month. By the end of the year, LegWork has reached nearly 40,000 users and £200,000 in revenue.

How a bottom-up forecast can unlock explosive growth.

Put all the factors we’ve discussed together, and you have a very simple, bottom-up forecast. It gives you a much better estimate of sales than you might have shown to an investor in your last round of funding, and much more still.

The value lies in how it can help you to plan growth.

Good forecasts help you to understand revenue

Measure effectiveness at each stage of the funnel. As you move into real users and real revenue, you’ll be able to use this to track performance of each channel and the activities within them.

Uncover routes to growth. Advertising is expensive. Investing in blog content takes time. Figuring out which channels work for you is incredibly important. Using a proper forecast with a pipeline allows you to identify the right channels, test new against existing, and identify problems in the pipeline (e.g. conversion) where extra growth potential can be found.

Good forecasts help you to understand cost

Every activity in a business has a corresponding cost. Write a blog post — that’s one of your people out of the game for a couple of hours. Buy advertising — that’s serious budget spent. There’s huge value to unlock in understanding the effectiveness of each activity, and how that changes over time.

Growth needs resources. If you invest in growing your user base, how long before you need a community manager? A customer success team? Forecasting allows you to anticipate when you’ll need to budget for more people — and when you may need another round of investment to accelerate your growth.

Good forecasts help you to understand strategic decisions.

Growing doesn’t just mean you need a bigger team, it has a huge impact on your strategy too. A good forecast can help you anticipate tough decisions. For LegWork, these big decision could include:

  • How many users is it possible to acquire before you must move to a new territory? (LegWork has a 250k forecast by month 24 — London is unlikely to support that growth).
  • How much will that growth cost?
  • How many users can your current hosting or code base support?
  • At what points will you need to hire a customer service team, or full time marketers?
  • When will you need to raise the next round of funding to fuel growth.

Bottom-up forecasts help you to make assumptions, and test and iterate from them. Understanding these scenarios and the quality of the customers that come from different channels is a basic but effective way to test your messaging, offers and optimise what you offer and how you talk to your customers.

What’s next?

If you don’t have a forecast, build one today. It’ll uncover ways to get better value for money and from your people long into the future. Download our starter template here.

Our advice: start with your customer. Their needs. Their wants. The problem you’re solving for them. Consider what the most important metric is for you to show traction, and the resources and activities you need to grow it.

Let’s chat.

Once you start, you’ll probably have questions. Thoughts and feedback. I’d love to hear it. Mail me here or message me jar4d.

Jared Ruddy