When you build something that handles other people's workloads, there is a strong pull to keep your reliability numbers private until they look good. The logic feels rational: why advertise imperfection? Why give a prospect a reason to hesitate?
We felt that pull. And then we published our uptime numbers anyway — months before they were where we wanted them to be.
This post explains why.
Founders evaluating infrastructure tend to be builders themselves. They have shipped code at 2 a.m. They have watched a deploy go sideways. They know that perfect uptime numbers from a young company mean either a short track record or a lie.
A status page showing 100% availability for twelve months tells a seasoned buyer one of two things: you haven't been tested yet, or you're not measuring honestly. Neither inspires confidence.
What does inspire confidence is a company that says: "Here is where we are. Here is what went wrong last month. Here is what we changed." That pattern — number, context, action — is the actual signal buyers look for. Not the number alone.
There is a gap between what you know about your own system and what your prospect knows. The wider that gap, the more a buyer has to trust you on faith. Faith is expensive. It does not survive procurement reviews or security questionnaires.
When you withhold reliability data, the gap fills with assumptions — and those assumptions skew negative. A prospect who cannot find your uptime numbers will assume you are hiding them. A prospect who finds a polished page with no incident history will assume you are performing.
Publishing honest numbers narrows the gap. It replaces assumption with evidence. Imperfect evidence beats none, because it gives the buyer something to evaluate instead of something to suspect.
We shared monthly availability percentages, a list of incidents with durations and root causes, and a plain-language summary of what we did after each one.
Some of those months were not flattering. We had stretches where our numbers would not have survived a comparison chart against a larger competitor. We published them anyway.
The response surprised us. Prospects who saw the data asked better questions during sales calls. They asked about specific incidents, our response process, how we prioritize fixes. Those conversations moved faster and closed at higher rates than calls where reliability never came up because the data was hidden.
Transparency did not scare buyers away. It attracted the ones who were serious.
Publishing early is not the same as publishing carelessly.
Every number we share comes with three things: what happened, why it happened, and what changed as a result. The number without context is a liability. The number with context is a story about a team that fixes things.
We also set expectations clearly. We never claimed our uptime was where we wanted it to be. We said we were working toward higher targets and showed the trendline. A buyer can look at a chart that moves up and to the right and make their own judgment about trajectory. That judgment, because they made it themselves, carries more weight than any claim we could make.
Here is the thing about publishing early: you establish a pattern. By the time your numbers are genuinely good, you have a track record of sharing them. The good months land differently when they follow a history of honest bad months.
If you wait until everything is polished to start publishing, you are asking the market to take your first data point on faith. If you have been publishing all along, the latest number is just the newest entry in a series the buyer already trusts.
Silence compounds too, in the other direction. Every month you do not publish is a month where the gap between what you know and what your buyer knows gets wider.
"Won't competitors use the data against you?"
Maybe. A competitor could screenshot your worst month and put it in a battle card. But a buyer swayed by a single monthly number pulled out of context is probably not the buyer you want. The buyer you want will look at the full record, see the trajectory, and ask you about the incidents. That conversation is one you will win, because you lived through the fixes and your competitor did not.
The real risk is not that you publish too early. The real risk is that you stay silent long enough for buyers to build a story about you that you never get to correct.
Transparency about what you are still fixing earns more trust than silence about what is already working. We learned this the way most startups learn things — by trying both approaches and watching what happened.
Publish your numbers. Add the context. Show the trendline. Let serious buyers decide for themselves. They will, whether you give them the data or not. You might as well make sure the data is yours.
Be the first to comment.
0 comments
Loading comments...