Why a spreadsheet isn’t an update
I’ve sent investor and stakeholder updates for small operations where I was the only person who touched the numbers all month. The temptation every time is to just paste a table: MRR, churn, new signups, done. It feels rigorous. It isn’t communication.
A table tells someone what happened. It doesn’t tell them why it happened, whether it was expected, or what you’re going to do about it. Data storytelling, in this context, just means adding the connective tissue between the numbers and the decisions behind them. It’s not about making the numbers sound better than they are. It’s about making them legible to someone who wasn’t in the business every day watching them move.
If you run a one-person shop and you’re sending updates to an angel investor, a co-founder who isn’t hands-on, or even just keeping a monthly record for yourself, the same discipline applies. The reader needs three things from you: what the number is, what changed it, and what you think that means. Leave any one of those out and you’ve handed over data, not a story.
Start with the metric, not the narrative
It’s tempting to open an update with a story (“this was a tough month because…”) and back it up with numbers later. I’d flip that. Lead with the actual figure, then explain it. If MRR moved from $4,000 to $4,300, say that first, plainly, before you explain the “why.” Readers who are used to reviewing numbers will scan for the figure before they read your explanation, and if they have to hunt for it inside a paragraph, they’ll assume you’re burying something.
This matters more when the number is bad. If churn ticked up, put the churn number in the first sentence of that section. Then explain it. Softening bad numbers by wrapping them in throat-clearing paragraphs reads as evasive even when that’s not the intent.
Explain what the metric can and can’t tell you
This is the part that’s easiest to skip and most important to include. Every metric in a business update is a proxy for something you actually care about, and proxies have blind spots.
MRR tells you recurring revenue booked this month. It doesn’t tell you whether that revenue is fragile (a handful of large accounts that could churn together) or durable (spread across many small accounts). If your update just says “MRR grew,” an investor has no way to know which kind of growth it was unless you tell them.
Churn tells you how many customers or how much revenue left. It doesn’t tell you why, and it conflates very different situations, a customer who churned because your product failed them looks identical in the number to one who churned because their own business closed. If you know the reason, say it. If you don’t know the reason for most of your churn that month, say that too. “We don’t have clear attribution on this month’s churn yet” is a more honest and more useful sentence than silence.
CLV (customer lifetime value) is a projection built on assumptions about how long customers stick around and what they spend. It’s a modeled estimate, not a fact you observed. If your CLV number moved a lot in one update, it’s worth telling the reader whether that’s because customer behavior actually changed or because you adjusted the assumptions in the model. Those are very different explanations and they lead to different conclusions about the business.
Conversion rate tells you what fraction of a defined group did the thing you were measuring. It says nothing about the size of that group or whether the group even represents your target market. A conversion rate can go up because your funnel improved or because a low-intent traffic source dried up, leaving only your best-qualified visitors in the denominator. Those look the same in the top-line number.
None of this means the metrics are useless. It means the number by itself is not the finding. The context around it is.
Use cohorts and A/B results as evidence, not verdicts
If you’re citing a cohort analysis or an A/B test result in an update, be precise about what it actually showed. A cohort comparison tells you how one group of customers who started at the same time behaved over time, compared to another cohort. It’s useful for spotting whether retention is trending better or worse across signup periods. It is not proof of what caused the difference unless you can rule out other things that changed at the same time, like a pricing change, a different acquisition channel, or a seasonal effect.
Same with A/B tests. A test that shows one version outperforming another is evidence, weighted by how much traffic ran through it and how long it ran. A test with a small sample or a short runtime can show a difference that doesn’t hold up. If you’re going to report a test result in an investor update, it’s worth including enough detail (roughly how much traffic, over what period) for the reader to judge how much weight to put on it themselves, rather than asserting the conclusion as settled.
I try to write these sections the way I’d want to read them if I were the one reading someone else’s update: tell me the result, tell me how confident you are in it, and tell me what would change your mind.
Attribution deserves a caveat, always
If you’re reporting where signups or revenue came from, say plainly how you’re attributing it, first-touch, last-touch, or something else, because the same customer journey can produce very different-looking numbers depending on the model. If two channels get double credit or a channel gets none because of how tracking is set up, an investor reading “channel X drove most of our growth” is being given a more confident claim than your data actually supports. A short caveat like “this is last-touch attribution, so channels earlier in the journey are undercounted” costs you one sentence and saves the reader from over-trusting a number that has a known blind spot.
Data quality is part of the story, not a footnote
If a number in your update is affected by a tracking gap, a migration, a change in how you define a metric, or missing data for part of the period, that belongs in the main text, not a footnote at the bottom nobody reads. “MRR growth this month includes a one-time correction for a billing error found in March” is a normal, honest sentence. Investors who’ve been burned by a founder quietly restating numbers later care a lot more about a track record of flagging this stuff upfront than about the update looking clean.
Where AI tools actually help here
For a one-person business, AI tools are genuinely useful for the mechanical parts of this: pulling last month’s numbers into a consistent format, drafting a first pass of the narrative sections from your notes, and checking that you’ve used the same metric definitions as last time so you’re not accidentally comparing apples to oranges. What they can’t do is know which of your caveats matter, or catch that your churn number this month is inflated because of a billing glitch you haven’t mentioned to it. Treat AI-drafted sections as a starting point you fact-check against your own source data, not a finished update. The judgment calls, what to flag, what to caveat, what you’re actually unsure about, still have to come from you.
Keep the format boring and consistent
The best investor updates I’ve seen (and the ones I try to write) use the same structure every time: the headline metrics up top, a short explanation under each one, a section for what’s uncertain or unresolved, and a short list of what you’re doing next. Consistency lets a reader compare this month to last month without having to re-learn your format. It also makes it much harder to quietly bury a bad number in a reorganized layout, which is its own kind of honesty.
None of this replaces your own judgment about your business, and it isn’t a substitute for a real conversation with the people reading your update if something significant is happening. It’s just a way to make sure the numbers you send are doing an honest job of representing what you actually know.
For more on how we think about metrics, methods, and running a one-person business on the numbers, visit the home page.
Get new guides and videos first — join the Telegram channel.