KPIs, OKRs, & Dashboards: How MSPs Can Know When Their Numbers Are Wrong

Whether you’re drowning in metrics or doing your best to ignore them, neither approach helps your clients or your MSP improve. While you might be inclined to think that tracking the wrong metrics causes the most issues, the reality is that trusting the right-looking ones too much is responsible for just as much, if not more, damage. Keep reading to discover why MSP KPIs sometimes lie, what counterbalance metrics are and why you need them, and the four ways MSP operators accidentally game their own numbers without realizing it. Good numbers only mean something if they’re accurate; it’s your job to know when they’re not.

What KPIs, OKRs, and dashboards actually mean for MSPs

As an MSP, it can be easy to overlook certain business metrics in favor of others, categorizing them as too advanced or only necessary for larger organizations. Metrics such as KPIs or OKRs, while familiar, may not be as top of mind as OMLs. The truth is, not only are metrics like these important for your MSP’s growth, they’re likely metrics you’re already measuring, but just calling something else. The following is a brief overview of what each tracks, as well as why each matters:

KPIs (Key Performance Indicators): 

KPIs are your “are we healthy?” numbers. For MSPs, these typically include average ticket resolution time, gross margin, technician utilization rate, mean time to resolution (MTTR), and first-time fix rate (FTFR). They answer the core question of: “Are we able to manage our current volume of work with the team and tools we have?”

OKRs (Objectives and Key Results): 

OKRs track goals and progress. An MSP’s OKR might be broken down into an objective, “improve onboarding experience for new clients,” and a key result of “reduce average onboarding time from 45 days to 30 days.” OKRs should be tied to departmental and company goals that are situational, based on the overarching strategic path forward. Good OKR’s are reverse-engineered by deciding the outcome you want and figuring out what milestones it will take to get there.

Dashboards: 

Dashboards are where KPIs and OKRs come together. They pull your KPIs and OKRs into one view so you can see how you’re tracking. Good dashboards should exist both at the department level and the individual level. Everyone should know what success looks like in their role, and how it ladders up to the larger business goals they’re a part of.

How to know if your MSP metrics are telling the whole truth

Here’s the unfortunate truth: numbers only tell part of the story. The difference between a metric that helps you and one that misleads you usually comes down to counterbalancing metrics. A counterbalance metric is a second data point that validates or challenges your primary metric. Without a counterbalance in place, a single number can be taken at face value even though there is another metric pointing to a larger issue below the surface. 

Take the common MSP metric of CSAT as an example. A 100% customer satisfaction score sounds incredible, that is, until you find out only 6 out of 100 clients actually filled out your survey. Once you take into account your response rate, that initial customer satisfaction number actually tells you very little. You need to track both CSAT score and CSAT response rate as a counterbalance metric to ensure accuracy. Low response rates like these are worth investigating. Why are the other 94 clients silent? The silence alone is its own data point. Metrics, or lack thereof, can become the trigger you need to make better business decisions. 

Here’s a list of common MSP counterbalance metric pairs worth building:

• Ticket volume and client-reported issue rate: The question being answered: Did volume drop because things got better, or because clients stopped calling?
• CSAT score and CSAT response rate: The question being answered: How many clients are being counted vs actually being heard?
• Average resolution time and first-time fix rate (FTFR): To note: Fast closure doesn’t mean good closure.
• MRR growth and net revenue retention (NRR): The question being answered: Are you growing new business, or churning your existing customer base to fund it?
• Technician utilization and billable efficiency rate: To note: High utilization doesn’t automatically translate to profitable utilization.

Building metrics is easy enough. But making sure your numbers actually reflect reality takes context, counterbalance metrics, and a willingness to look at the story behind the numbers; rather than just the numbers themselves.

Top 4 ways MSPs cheat on their KPIs and OKRs, without realizing

Although most MSPs don’t set out to cheat on their performance metrics, the phenomenon is common enough that most MSPs fall into one of four categories: 

1. The goalpost shuffle:

You set a bold target. Then reality hits, and “increase MRR by 20%” quietly transforms into “work toward improving sales pipeline health.” It sounds strategic, but it’s actually moving the goalpost when you’re already mid-game. OKRs exist to stretch you, not to be rewritten to guarantee you can claim a win. If your key results are changing before the cycle ends, they’ve become aspirations you’re reaching for, not commitments you’re responsible for.

How to fix this issue: Lock OKR language at the start of each cycle. Conduct post-mortems after the period ends, instead of during it.

2. The vanity metric victory lap:

Everyone loves a pretty chart. The problem is, charts can lie. Ticket volume down? Could mean your team got more efficient. Or maybe your clients stopped calling because they couldn’t get through. CSAT looks perfect? Maybe only three customers actually filled it out. Vanity metrics make us feel good, but they rarely tell the whole truth. 

How to fix this issue: Every primary metric needs a counterbalance metric that can challenge it. If a metric consistently looks too good to be true, it’s probably a vanity metric.

3. The dashboard of dreams:

If your dashboard looks too clean, it’s probably because someone filtered out the messy parts. Cutting out red flags doesn’t make them go away; it just makes them harder and more expensive to fix. A good MSP dashboard should surface exceptions and anomalies automatically, not just when you want it to.

How to fix this issue: Build your dashboards to show what’s wrong first, not just what will make you feel good. Design for the outlier first, then the average.

4. The “blame the data” move: 

When the numbers aren’t what we hoped, it’s easy to say, “The report must be wrong.” Data quality can certainly be an issue worth investigating. But more often than not, it’s the reality we don’t want to face. If you’re spending more time looking for data errors than insights, it’s time to check your process, not your math.

How to fix this issue: Keep your data quality reviews separate from your performance reviews. This way, one won’t turn into a convenient excuse to avoid the other.

If you think your MSP falls into any of these categories, now is the time to take action. The longer you wait to address your KPI and OKR issues, the more difficult and the more expensive they will be to fix.

How Auvik helps you keep your MSP metrics honest

Data quality problems extend even deeper than your spreadsheets. If your network visibility is lacking, every KPI that is built on top of it can be called into question. From ticket resolution times to uptime figures and alert accuracy, every metric you rely on needs to be investigated. Auvik solves the network visibility problem for MSPs. Auvik provides MSPs with real-time, automated visibility into client network infrastructure, so that the numbers feeding your dashboards actually reflect what’s happening within the network. Learn how Auvik supports MSP performance tracking today by signing up for a free demo.