Sales forecasting is about understanding the accuracy of deal timing and commercial terms—not achieving perfection. The biggest driver of forecast accuracy is data quality, not AI or tools. By improving CRM inputs, capturing internal deal conversations, and asking one critical question in every pipeline review, teams can build more realistic and reliable forecasts.
While not a lot has changed in sales since Mesopotamia, one thing most certainly has, and that’s sales forecasting. The last few years have made getting to a more truthy sales forecast easier. However, people still need to realize, even with AI, sales forecasting is never going to be perfect.
What You’ll Learn in This Article.
- The real definition of sales forecasting (beyond tool-based hype)
- Why “garbage in, garbage out” drives forecast accuracy
- How CRM engagement actually impacts forecast reliability
- The hidden data most companies ignore in internal conversations
- How to evaluate forecasting tools and build success metrics
- The single best question to improve sales forecasting accuracy
- Why bringing humanity back into forecasting matters in 2025–2026
The biggest mistake in sales forecasting right now is how much the executives and heads of sales are over-indexing on tools and in particular over-indexing with AI in sales.
We all know garbage in equals garbage out. Which means the more input we have the better our odds are at creating a more accurate sales forecast. So, with all this in mind I wanted to share some best practices in sales forecasting to help create the right balance between the human and the machine. And I will also share the one best question to ask to help ensure you have the most accurate data for improving your sales forecasting.
Sales Forecasting Defined
At the simplest level the goal of sales forecasting is to understand the accuracy of the commercial terms along with the accuracy of the timeline for each deal in your sales pipeline.
A fun way to think about this is deciding if our sales forecast is going to be based on accuracy or are we just “smoking hope-i-um” based on promises made by executives and the board of directors? My friends at people.ai like to call this “revenue reality”.
We should not assume our sales forecast is right. We should actually assume our sales forecast is wrong. And then from there we work to improve its accuracy. And the ultimate goal is to have it improve week over week, month over month or quarter over quarter. And yes, year over year.
Why Sales Forecasting is Important
By nature, humans are creatures of comfort. We prefer things to be easy, not hard. Getting out of our way or getting out of our comfort zone is easier said than done. This mindset is present in our personal as well as our professional lives. I mean, who wants to look at a bunch of data and see you are not doing well, right?
In sales, we have to do this to determine the true health and sometimes viability of our business. We use sales forecasting and actual sales numbers to set new milestones, goals, and growth of the organization. The better our sales forecasting is, the better we can predict our growth. With this information we can budget for new tools and initiatives, budget for additional headcount, budget for improving the execution of our product roadmap. Yes, it’s scary sometimes. As I always like to say, the truth will set you free, it just might sting a little at first.
Sales Forecasting Accuracy Best Practices
Not everyone is at the same stage of sales forecasting. So this is merely a list of some of the best practices for sales forecasting I’ve collected over the years. Please use, ignore, or add anything you think will help your sales forecasting accordingly.
And if you are new to sales forecasting let this be a starting point to help you establish your own sales forecasting best practices.
Garbage In – Garbage Out – As I mentioned above, GIGO, is often the most important criteria for improving sales forecasting. It’s also a place where people over-index on AI in sales. They think the AI will eliminate the errors. AI is just like a human; it can only be interpreted based on the data it has. So, if it’s got bad data, its not going to make some magical fix. In fact, it will simply give you the bad information faster than if you did it in an google sheet. Maybe in time AI could fix this. We aren’t there yet.
CRM Engagement – CRM Engagement is just a fancy way of saying GIGO. Keep in mind there are a few parts of CRM Engagement related to improving sales forecasting accuracy.
Most of the time CRM engagement starts with the inputs from the Sales Rep into the CRM. That’s not true.
The first step of solving GIGO and CRM Engagement is making sure you have the right fields collecting the right information in relation to the deal moving through the sales cycle.
So often field creeps are allowed into your CRM. These are all those fields that other departments and executive leaders love to have added to the CRM because they want some fancy and special report spit out at their convenience. Then they try to justify this by telling the sales team that this information will help them make more commissions. Hey Leaders, news flash! NO IT WON’T. Your sales reps are not your organization’s assistants anymore, than an SDR is an assistant to do the grunt work for an AE who’s just as lazy as you are.
If you ask for too much garbage to go into the CRM, guess what you’re going to get? Do I need to really spell it out for you?
So, the first step to improve sales forecasting accuracy is having a clean and meaningful CRM opportunity tracking page. Additionally this means most fields should be drop downs or check boxes. Diatribes only go into one place. An open field called, “Top Notes”
Now, the second part of CRM Engagement related to sales forecasting is about the infomation input into the CRM by a sales rep or a note taking tool
All the information must be input into the system. This includes information from emails, phone calls, zoom calls, text messages.
If you still expect your sales team to input this information into the CRM manually I have two suggestions for you.
You should be fired if you’re a sales leader or executive, and even you super-duper smart technical founder.
You should update your resume ICs like SDRs, AEs, and CS. Go find a company who values your time in front of customers not in front of a screen.
Automated note taking tools are now table stakes in the modern sales world, they are not “nice to haves”.
The Missing Data in Your Sales Forecast
Even if you do everything, I mentioned above you’re still missing one the most crucial pieces of information in your CRM and therefore your sales forecast will be wrong. Right now, you’re only getting a partial glimpse into each sales opportunity. You’re only getting the external view. You’re only getting the communications that are happening “outside” of your company.
What about the conversations that happen internally at your company about the deals in your sales pipeline? You know, pipeline reviews, big deal meetings, slack messages, internal emails, conversations with product or marketing? How about those discount conversations you have at the end of the month or quarter?
Shouldn’t these also be included? Wouldn’t you like to know the predictability of a deal actually closing if you offer some extension, discount, or inclusion of additional services? Might that information tell you that offering a discount won’t actually make the deal close faster so maybe you are disappointed the deal closes next month or quarter, but you get higher margins?
I know, I know, even if you could predict that some executive is still going to offer the discount anyway because they think they are the smarty-pants in the big deal meeting. And then they are going to blame the rep when it doesn’t close. At least now the rep can just hold up the sales forecasting report showing it was never going to happen in the first place.
This information is truly the missing link in your sales forecast. So simple to capture, yet nobody is doing it. This is a critical part of understanding the forecast and it really isn’t that hard to do. When we say “better, faster, and cheaper”, this is what “better” means.
And allow me to rant a bit more about one of the biggest mistakes executives make when it comes to sales forecasting. It’s quite simple. Stop asking reps for information you can get out of the system. Nothing is a bigger waste of time and creates a terrible sales culture than an executive sitting in a meeting asking for information they can find in the CRM.
If the executive has a specific question they consistently want answered, then they need to build or have revenue ops build a report that shows this information. You keep whining about information not being in the system, then when it is in the system, you won’t take the time to get it.
Second, newsflash executive leader. Your reps’ time is more valuable than your time. Every moment you are asking redundant questions with information already in the CRM, you are wasting the time I should be spending on actually closing the business based on the unrealistic goals you and the board created.
Ok, rant over. For now, at least.
Stop complaining like my 15 year old son who won’t put down the video game controller and asks me to make him a snack when he can easily get up and do it himself.
Improve Sales Forecasting with the Right Tools
The truth will set you free, and it might sting a little at first. As previously hinted, everyone wants Better, Faster, and Cheaper. Unfortunately, most people choose faster and cheaper and create an internal bias that it will be better. That is simply not true. Never has been, never will be. There’s a reason the tortoise beat the rabbit in the race.
Now don’t get me wrong, the world in 2025 and 2026 is moving faster than ever with the addition to AI in sales. That doesn’t mean you have to make mistakes by going too fast though.
One of the smartest best practices for improving sales forecasting aside from having a strong sales process is, of course, having the right sales stack and specifically choosing the right type(s) of sales forecasting tools. Here’s how I coach everyone to look at making any large decision around their sales stack and even their sales training.
We expect to be better at ___, ___, and ___ upon completion (or implementation) of ______.
Now if you use buzzwords like forecasting, negotiation, or closing. You need to get more specific. Forecasting what? Negotiating which part of the deal? Negotiating with a particular title/role like CFO or Procurement? Get my drift?
From there we need to define success.
An increase (or decrease) in ___, ___, and ___ will be our measuring stick of success.
Sometimes a reduction can be an improvement. Meaning, if we reduce the dead deals in our pipeline, the pipeline may decrease but our close rate and accuracy should increase.
And from this point we need to take a hard look in the mirror.
True/False – Our baseline numbers supporting the metrics we are wanting an increase / decrease are currently available.
If they do, exist (true), fabulous!
If they do not exist (false), we have to get real, real fast by answering this question. “Since we do not have baseline metrics it will take us ___ weeks/ months to create a trusted baseline.”
And sadly, if your sales cycle is 60+ days, a two-week baseline is probably not going to work for you. Sorry, not sorry. This would just be another example of faster and cheaper not equaling better.
The Best Sales Forecasting Question
Ok, so, you want to know the best question to ask to help ensure sales forecasting accuracy? It’s quite simple. For every opportunity in your pipeline always ask this question first.
“What’s the one thing preventing this deal from closing by the end of (month/quarter)?
Look, we all know there is typically more than one thing. However the human brain can get overwhelmed when forced to make a decision. So, let’s make it easier.
When the rep gives an answer, then simply respond, “Great, what else?”.
This now allows for a much more comfortable dialogue to get to a more accurate understanding of the opportunity. And help determine how comfortable we are in including this opportunity in the forecast.
This one question is better than the following:
So, what’s it going to take to get this over the finish line / goal line?
As you can see this is far to ambiguous to help us really determine anything. And does not allow for a singular focus. And a singular focus is critical when it comes to navigating and forecasting sales opportunities and sales pipelines.
Ask a vague question, get a vague answer. Which is the same as GIGO.
Conclusion: Bringing Humanity Back Into Sales Forecasting
Now lets step back for a moment. We’ve talked about a lot of things here. And here’s where I hope to leave you. It’s about humanity. Specifically bringing humanity back into sales. While the machines are here, we need to remember they cannot do it all for us. AI in sales can enhance our capabilities. It cannot fully automate and replace humanity. It’s the human touch that helps create a greater sense of reality. Especially when it comes to sales forecasting.
