Most founders overestimate their monthly revenue
Something I've noticed talking to founders:
If pipeline looks like $200k, people assume they'll close $200k.
But when the month ends, maybe $40k–$60k actually closes.
Deals slip, timelines move, buyers disappear.
How do you adjust for that when forecasting revenue?
Is there a rule you follow?
blog
This is painfully true, Manish. Early on with FontPreview.online , I'd get excited about a "promising" conversation and mentally add it to the month's revenue. Then... nothing. Or a "definitely next month" that never came.
The gap between pipeline and reality is humbling.
These days I use a simple rule: take the pipeline, cut it by at least half, then cut again if it's early-stage. Still get it wrong sometimes, but less painfully.
Curious — does RevPredict have a rule of thumb for different stages of deals? Like, how much should you discount "just met" vs "sent proposal"?
That “cut the pipeline in half… then cut again” rule sounds painfully familiar 😅 I think a lot of founders end up discovering something like that the hard way.
What I'm experimenting with in RevPredict right now is stage-based probabilities instead of a single discount. Rough idea looks something like:
• Just met / discovery → ~15–20%
• Demo / deeper conversation → ~35–45%
• Proposal sent → ~60–70%
• Negotiation / final discussion → ~80–90%
It’s not perfect of course, but it tends to produce a much more realistic expected revenue number than counting the whole pipeline.
One thing I’m still trying to understand better is time-in-stage deals that sit too long often end up slipping even if they’re technically in a late stage.
Out of curiosity, with FontPreview do you mostly see deals stall at a particular stage, or is it usually the timeline slipping?
That stage-based probability model makes a lot of sense, Manish. A single discount across the whole pipeline hides too much - a "just met" deal and a "negotiation" deal are completely different animals. Your breakdown feels much closer to reality.
The time-in-stage point is sharp. A deal that's been in "proposal sent" for three weeks is probably not the same as one that got there yesterday. Do you factor time into your model at all, or is that something you're still figuring out?
For FontPreview, most of our "deals" (if you can call them that - it's free, so conversion is softer) stall at the awareness stage. Someone finds the tool, tests it, maybe even saves a few fonts... then just never comes back. It's less about timeline slipping and more about not converting curiosity into habit.
For the paid side (licensing reports, brand kits), the stall usually happens after a question. They'll ask "can I use this font for a logo?" and then... nothing. No follow-up, no purchase. I'm still learning what happens in that gap.
This is super insightful .. especially the distinction between “time slipping” vs “no habit formed.”
What you’re describing on the free side feels less like a pipeline problem and more like a behavior problem people reach value once, but don’t cross into repeat usage.
On the paid side though, that gap after a question is really interesting.
Feels like a classic “intent without urgency” situation .. they’re evaluating, but there’s no strong trigger to act.
I haven’t built time-in-stage into the model yet, but conversations like this are pushing me in that direction.
Out of curiosity ... when someone asks a question like “can I use this font for a logo?”, do you usually respond quickly? And do you see any difference when you follow up vs when you don’t?
I try to respond as quickly as possible, and it definitely helps. But even then, some conversations just go silent. I'm still experimenting with different follow-ups to understand what actually turns curiosity into action. Thanks for the insight!
Very true, revenue numbers can be misleading without considering costs and churn.
Yeah, one pattern I keep noticing is that people overestimate deals that are early in the conversation.
Anything thatfeels promising (great call, interested prospect, positive feedback) often gets mentally counted as future revenue even though the real buying signals aren’t there yet.
Another place I see overestimation is around timelines .. founders assume deals will close this month when in reality they slip to next month or later.
Curious if you've noticed something similar in your experience, or if there’s a different pattern you see more often.
True lol
This is very true.
A lot of founders confuse pipeline with actual revenue. In reality, only a small percentage of deals usually close.
Do you personally apply a specific “close rate” when forecasting? For example assuming only 20–30% of the pipeline will actually convert.
Yeah exactly ..that's the trap a lot of founders fall into.
Seeing a $200k pipeline and mentally counting it as future revenue rarely works because deals slip, stall, or disappear.
What I've been thinking about is applying stage-based probabilities instead of a single close rate. For example:
• Discovery → ~20%
• Demo → ~40%
• Proposal → ~70%
• Negotiation → ~85–90%
That tends to produce a more realistic forecast than just saying “30% of pipeline will close”.
But I'm curious .. do you usually apply one global close rate across the pipeline, or do you track conversion by stage?
yes . i think so but not always true
I run a commission-based ed-tech business — trainers, students, parent inquiries. Same problem, different shape: 80 inquiries in a month sounds like 80 potential students. You close maybe 18.
Took us an embarrassingly long time to stop forecasting from the top of the funnel.
What actually helped: we started reverse-engineering from our closed deals instead of our open ones. What did every conversion have in common? What question did the parent always ask right before they paid? How many touchpoints on average? Once you know what a closing conversation looks like, you can tell within five minutes whether a new lead is going there or not.
The rule I follow now: if I can't point to the specific friction that's been resolved, I don't count it. No friction resolved = not closing this month. Simple, brutal, usually right.
This is a great way to look at it.
Starting from closed deals instead of open pipeline is probably the most honest way to understand what actually converts. The “friction resolved” idea resonates a lot .. in many cases the deal only really becomes real once that key objection or concern gets cleared.
In B2B sales it's often things like budget approval, internal buyin, or timing. Until that moment happens, the deal can sit in the pipeline for weeks but isn't really progressing.
Your point about recognizing a closing conversation within a few minutes is interesting too. Do you feel like those signals show up consistently (same questions, same objections), or does it vary a lot depending on the parent/student?
That’s a really interesting way to think about it.
Reverse-engineering from closed deals instead of open pipeline makes a lot of sense. Looking for the patterns right before the payment moment is probably where the real signal is.
I'm curious — once you identified those common questions parents ask before paying, did you start proactively addressing them earlier in the funnel?
I think it's the same as why project timelines slip. If there's no honest calculation made, just an optimistic "guess" based on "we want it to be done by then", it'll be wrong. Each of these cases (I mean deals slipping, timelines missed etc) should be taken into account based on previous data and prediction calculated taking them into account. Most of the time the guess is instead based on "we have 200k in the pipeline! we'll be rich!" and the less convenient parts of previous data are ignored.
Otherwise the calculation would be more like % of closed deals based on previous month pipeline * current pipeline. In a bootstrapped business this might actually happen, but most VC funded businesses I've seen would prefer to ignore this calculation to show higher predicted value so that board won't complain quite as much.
Edit: Just saw your tool - pretty much it. The problem is often that there's no will to even perform this type of analysis.
Yeah, this resonates a lot.
I think the biggest issue is exactly what you mentioned .. the forecast often starts with the desired outcome rather than the historical data. Once people see a big pipeline number, it's very tempting to anchor on it and ignore the inconvenient parts of the data.
Looking at historical conversion rates tends to bring things back to reality pretty quickly. Even a simple calculation like “what percentage of last month's pipeline actually closed?” can already make forecasts much more grounded.
And you're right that the challenge is sometimes cultural rather than technical. The data might exist, but teams still prefer the optimistic version because it feels better in the short term.
I'm curious .. in companies where you've seen this done well, was it usually driven by the founder, finance, or the sales team?
I haven't seen it done well anywhere at all. At least nowhere where I've had access to the predictions. I think it starts from the founder, board or investors and if any of them shows any disappointment when estimates are made, they'll get massaged into something more palatable.
Bootstrapped companies are probably more honest, although I don't have experience with any other bootstrapped company beyond my own, because their survival depends on the estimates matching the financial reality whereas VC funded can ask for more money, if their estimates are higher.
That's a really interesting point, and it does feel like a cultural problem as much as a technical one.
If forecasts are used more as a signal of confidence to investors or the board, there's a strong incentive to make them look optimistic rather than accurate.
Bootstrapped companies probably feel the pain faster because the numbers immediately affect cash flow and survival, so there's less room for "optimistic accounting".
Ideally forecasts should be more like a diagnostic tool — something that tells you early that you're going to miss the target so you can adjust pipeline or effort.
Out of curiosity, when you ran your own forecasts in your bootstrapped company, did you mostly rely on historical close rates, or more on intuition from the conversations you were having with prospects?
I don't do forecasts for the predicted income side at all at the moment, because there's no reason to. Unless I need to apply for a loan, or my expenses grow significantly, I don't think it's too relevant and I can just skip it.
That makes sense, especially in a bootstrapped setup.
If expenses are predictable and you're not relying on outside capital, the pressure to forecast aggressively probably isn't there. In that case actual cash flow matters more than projections.
I’ve noticed some founders only start caring about forecasts when they’re making bigger decisions like hiring, increasing marketing spend, or committing to longer contracts.
Out of curiosity, do you mostly track things like monthly revenue trends instead or is it more of a “watch the bank balance and react” approach?
For now it's just watching the bank balance and being ascetic with expenses. Makes no sense with only a couple of users to try to do more. I'm at the very beginning stages. Hopefully revenue trends will start mattering at some point.