Most SaaS dashboards are built to look complete rather than to actually surface what matters, and the difference between someone reading it for the first time and someone who’s seen two hundred of them is mostly about which three or four numbers get checked first, not how many numbers get read overall. 

The first thing worth checking, before anything else, is whether growth is coming from new logos, expansion within existing accounts, or a mix, and how that mix has shifted over the last few quarters. 

A business growing almost entirely through new logo acquisition with flat or shrinking expansion revenue has a very different risk profile from one where expansion is doing real work, even at the same topline growth rate, because new-logo-dependent growth is far more exposed to a slowdown in the sales pipeline. 

The second thing worth checking is the shape of the retention curve by cohort, not the blended average. A blended NRR of 110% can hide a cohort that’s actually deteriorating sharply, propped up by an unusually strong recent cohort that hasn’t had time to churn yet. Cohort-level retention tells you whether the business is getting structurally healthier or whether the average is being carried by recency. 

The third is the relationship between CAC payback and contract length. A payback period that looks fine in isolation can be genuinely concerning if contract terms are shortening, because a shorter contract means less time to recoup that acquisition cost before the renewal risk shows up. 

None of these require exotic data. 

They require knowing which three questions to ask of a dashboard that’s usually built to answer twenty different, less important ones. 

Once you know to look for these specifically, a dashboard stops being forty metrics and starts being three or four numbers with everything else as supporting context.

Most board packs are full of benchmarks nobody asked for and short of the one benchmark that would actually change the conversation. 

Revenue growth versus sector average, gross margin versus comparable companies, the usual suspects. All reasonable, all somewhat expected, and all things board members have seen enough times that they barely register anymore. 

The benchmark that actually shifts a board conversation is almost always narrower and less obvious than the standard set. Not “how do we compare on NRR” but “how do our sales cycle lengths compare for deals of this specific size and this specific buyer persona.” Not “how’s our churn” but “how does our churn in the first 90 days after onboarding compare to companies with a similar onboarding motion.” 

These are harder to source, because they require genuinely segmented data rather than a broad sector pull, but they’re the ones that actually tell a board something they didn’t already suspect. 

The reason these narrower benchmarks matter more is that board members already have a rough intuition about the standard metrics, built from sitting on other boards. 

What moves the conversation is a number that’s specific enough to be surprising, because a surprising number is the one that actually changes what gets decided in the room, rather than confirming what everyone already assumed going in. 

If you’re preparing a board pack and want one benchmark to genuinely earn its place on the agenda, pick the narrowest, most structurally comparable slice you can actually source data for, even if it means one benchmark instead of five broad ones. 

A single well-targeted comparison beats a wall of loosely comparable sector averages every time, because it’s the one nobody’s seen before and the one that actually prompts a decision rather than a nod.

Most benchmarking decks get built the same way. 

Someone pulls a dataset, filters by broad sector, and produces a chart with your company as a dot against a cloud of “comparable” businesses. It looks rigorous. It usually isn’t. 

The issue is what “comparable” actually means in most of these datasets. Sector tags are blunt instruments. “B2B SaaS” covers a company selling £50-a-month subscriptions to small businesses and a company selling seven-figure annual contracts to enterprise procurement teams. Both get the same tag. Neither one’s metrics mean anything next to the other’s, but they end up in the same benchmark cloud because nobody built the dataset with contract structure or customer size in mind. 

The second problem is survivorship. A lot of benchmarking data is built from whichever companies were willing to share numbers, or whichever companies are large enough and public enough to have numbers available to scrape. That’s a specific population (generally healthier, generally further along, generally more visible) and it skews every median upward in ways that make an otherwise perfectly healthy company look like it’s underperforming. 

What you actually want, and what almost nobody builds properly, is a benchmark that segments on the things that structurally drive the metric you’re looking at. For NRR, that’s contract length, customer concentration, and expansion motion. For CAC payback, it’s channel mix and average contract value. For gross margin, it’s whether the business carries meaningful services revenue alongside software. Compare on sector alone and you’re comparing apples to a fruit bowl. 

The practical test before you trust any benchmark: ask what’s actually driving the segmentation. If the answer is “industry” and nothing else, treat the number as a rough orientation, not a target. If the segmentation includes structural variables like contract length, customer size, and revenue mix, you’re looking at something closer to a genuine comparison, and it’s worth building your board narrative around it. 

This is exactly the distinction we built the benchmarking tool around; segment first, then compare, rather than compare first and hope the segment was close enough. 

There’s an important difference between a management team that’s coasting, capable but not pushing hard because nobody’s holding them to a higher bar, and a management team that’s genuinely stuck, working hard but blocked by something structural they can’t solve alone. 

Both can look similar from a board pack, flat or underwhelming numbers, but they call for completely different responses, and misdiagnosing one as the other tends to make things worse rather than better. 

The tell that usually distinguishes them isn’t in the numbers, it’s in how the management team talks about the numbers. A coasting team tends to describe underperformance in comfortable, low-specificity language, “the market’s been tough,” “we’re tracking roughly where we expected,” without much evident frustration or urgency behind it. 

A genuinely stuck team tends to be considerably more specific and considerably more visibly frustrated, because they know exactly what’s blocking them and have usually already tried several things to fix it without success. The stuck team’s frustration is a good sign, oddly, because it indicates real engagement with the problem. The coasting team’s calm is the more concerning signal, because calm in the face of underperformance usually means the bar has quietly lowered without anyone deciding to lower it. 

The practical way to test this directly: ask the management team, specifically, what they’d do differently if they had one additional resource, whether that’s budget, headcount, or a specific capability, to solve the current performance gap. 

A stuck team will usually have an immediate, specific, well-reasoned answer, because they’ve been thinking about exactly this. A coasting team will often struggle to answer specifically, because the honest answer is that more resource wouldn’t obviously change much, since the constraint isn’t really resource, it’s effort or ambition. 

Coasting calls for raising expectations directly and specifically. Being stuck calls for removing a specific blocker. Applying the wrong response to either situation, more resource for a coasting team, or higher pressure for a genuinely blocked one, tends to make both problems worse rather than better. 

Integration playbooks look impressively thorough in the deal process. 

Detailed workstreams, clear milestones, named owners for every task. What they consistently underweight is the difference between a task being theoretically assigned and a task actually happening on an ordinary Tuesday, when the person nominally responsible for it is also dealing with three other urgent things that have nothing to do with the integration plan. 

The pattern that plays out most often: the integration playbook assigns a specific workstream, say systems consolidation, to a specific senior person, and that person genuinely intends to drive it forward. 

But that same person is also running a piece of the core business day to day, and integration work, however important strategically, rarely feels as urgent on any given Tuesday as whatever operational fire is currently burning. 

Six weeks in, the integration workstream owner hasn’t actually made much progress, not because the plan was wrong, but because nobody built in the actual time and headspace required to execute it alongside an already full role. 

The playbooks that actually work tend to share one specific feature that’s easy to skip when building the plan under deal timeline pressure: dedicated integration capacity that isn’t also running a piece of the existing business. 

Whether that’s a dedicated integration lead, a temporary consultant, or explicitly carved-out time for an existing senior person with their other responsibilities genuinely reduced rather than just nominally deprioritised, integration work that competes for the same hours as running the business tends to lose that competition, quietly, week after week, until the plan is meaningfully behind and nobody can point to exactly when that happened. 

There’s a specific number that shows up in almost no standard board pack, and it’s arguably more predictive of a portfolio company’s health over the following twelve months than most of what does show up: the trend in time-to-resolution for customer-reported issues, tracked at the individual issue level rather than as an average. 

The reason this matters more than it gets credit for is that it’s usually the earliest visible signal of organisational strain, well before that strain shows up in retention or revenue numbers. 

A support or customer success function that’s quietly falling behind on resolution times is often the first sign that headcount hasn’t kept pace with customer base growth, or that a product issue is generating more inbound than the team can absorb, or that internal prioritisation has drifted away from customer-facing problems toward internal projects. 

None of that shows up in NRR yet, because customers who are having a slower-than-usual support experience don’t typically churn immediately, they tolerate it for a while first, sometimes for several quarters, before it eventually shows up as a retention problem that then gets treated as a surprise. 

Most reporting structures don’t track this at the individual issue level, they track average resolution time, which smooths over exactly the kind of tail-end deterioration that matters most, a growing number of issues taking meaningfully longer than they used to, hidden inside an average that hasn’t moved much yet. 

The practical fix is straightforward and cheap to implement: ask for the distribution of resolution times, not just the average, and specifically ask how the tail, the slowest 10 or 20 percent of issues, has moved over the last few quarters. 

A worsening tail with a stable average is one of the more reliable early warning signs available, and it’s almost never in the standard board pack unless someone specifically asks for it. 

Buy-and-build strategies get modelled on synergy and revenue growth, and both of those things are real and achievable. What rarely makes it into the IC memo with any specificity is the actual cost, in management attention and organisational disruption, of executing the integration itself, and that cost is usually the thing that determines whether year one goes well or badly far more than the strategic logic does. 

The specific cost that gets underweighted almost every time is senior management bandwidth. Integrating an acquisition properly, aligning systems, aligning culture, aligning go-to-market, requires serious time from the same small group of senior people who are also meant to be running the existing business and hitting the plan that was already in motion before the acquisition happened. That bandwidth isn’t free just because it doesn’t show up as a line item, and a management team stretched thin across running the core business and integrating an acquisition simultaneously tends to do both worse than either done well on its own. 

The second underweighted cost is the disruption to the acquired company’s own momentum during integration. A business that was performing well pre-acquisition, which is usually why it was an attractive target in the first place, often slows down during the integration period itself, as key people deal with uncertainty about their role, systems get disrupted, and customer-facing processes get temporarily degraded while things get combined. This dip is common enough that it should be expected and planned for, but it’s rarely built into the year-one model with any honesty, because acknowledging it upfront makes the deal look less immediately accretive. 

The practical fix, before any add-on gets approved: model a genuine integration cost in management bandwidth terms, not just financial terms, and be honest in the plan about a likely temporary dip in the acquired business’s own performance during the integration window. Deals that account for this upfront tend to manage through year one calmly.

Deals that assume synergy arrives smoothly and immediately tend to spend year one explaining to the board why it hasn’t.

At some point in most holds, the plan the CEO signed up for at close stops being the right plan, and someone has to have the conversation about changing it. This is one of the more delicate moments in the operating partner relationship, because how it’s handled determines whether the CEO experiences it as a genuine reset or as a quiet vote of no confidence in them personally. 

The mistake I see most often is treating the renegotiation as primarily an analytical exercise. Build the case, show the data, present the revised numbers, expect the CEO to simply update their view because the evidence is sound. Sometimes that works. More often, it doesn’t, because the CEO isn’t purely evaluating the analysis, they’re also evaluating what the request to change the plan says about how the board sees their performance so far, and that’s a much more personal read than the deck accounts for. 

The better approach starts before the numbers, not with them. Have an honest, private conversation with the CEO about what’s changed and why, framed around shared circumstances rather than CEO performance specifically. Markets shifted, an assumption in the original model turned out to be wrong, a competitor moved faster than anticipated, whatever the actual driver is. The goal isn’t to avoid accountability where it’s genuinely warranted, it’s to make sure the CEO experiences the renegotiation as something being done with them, based on new information, rather than something being done to them, based on a verdict about their performance. 

Once that framing is genuinely shared, the numbers conversation becomes much easier, because you’re now jointly solving a problem rather than negotiating across a table. CEOs who feel the plan revision was collaborative tend to execute the new plan with real conviction. CEOs who feel it was imposed tend to execute it adequately and no more, and adequate execution of a revised plan is rarely enough to hit a revised target. 

Every deal team has an instinct about whether a portfolio company needs a cost-out story or a growth story, and that instinct usually forms during diligence, before anyone’s actually run the business for a day. Sometimes it’s right. Often it’s a reasonable first guess that quietly hardens into the plan before it’s been properly tested against reality. 

The tell that a business needs cost discipline first isn’t really the margin number itself, plenty of low-margin businesses are healthily low-margin by design. The tell is whether the cost base has grown faster than the complexity of the business justifies, which usually shows up as headcount added ahead of clear need, tooling proliferation nobody’s consolidated, or vendor relationships that have never been renegotiated because nobody’s had the bandwidth to look at them properly. 

The tell that a business needs growth investment first, even if the margin looks a bit thin right now, is usually a genuinely underexploited customer base or channel. A company sitting on strong retention and an obvious expansion opportunity it hasn’t pursued because it’s been resource-constrained is a very different animal from a company that’s just accumulated cost without accumulated value, even if the two look similar on a topline P&L. 

The practical test, before committing to either story as the primary plan: look separately at cost growth versus revenue complexity growth over the last two to three years. If costs have outpaced the complexity that would justify them, you likely have a cost-out situation regardless of what the original deal thesis assumed. 

If the customer base and retention numbers suggest real unexploited demand, you likely have a growth situation even if the current margin looks unglamorous. 

Reading this correctly in the first month saves you from six months of pulling the wrong lever with real conviction.

Every operating partner walks into their first portfolio company with the same three levers already loaded. Pricing. Cost base. Sales headcount. It’s practically muscle memory at this point, and I get why. They’re visible, they’re measurable, and you can show movement on them inside a quarter. 

Here’s the uncomfortable bit. 

Two of those three are usually the wrong first move, and I’ve watched enough 100-day plans go sideways to say that with a straight face. 

Start with pricing, because it’s the one people reach for first and defend the hardest. Pricing is genuinely powerful. A well-run pricing exercise can add more EBITDA than almost anything else you’ll touch in year one. But it only works if the underlying customer base is stable enough to absorb it, and half the time nobody’s checked that before the price letter goes out. I’ve seen a portfolio company push a 6% increase across the board three weeks after acquisition, before anyone had properly segmented churn risk by cohort. Six months later they were explaining a logo attrition spike to the board that had nothing to do with the market and everything to do with a spreadsheet nobody stress-tested. 

Cost base is the second one, and it’s tempting because it’s the fastest win on paper. Renegotiate vendor contracts, consolidate tools, tighten headcount. All fine in isolation. The problem is sequencing. Cut before you understand what’s actually driving the number and you end up cutting the thing that was quietly compensating for a weakness somewhere else. I’ve seen support headcount cut right before a product migration that needed exactly the support capacity that just walked out the door. The saving showed up in the model. The cost showed up three months later in churn, and by then it’s not a line item anymore, it’s a narrative problem with the board. 

So what’s the lever that actually deserves to go first? Management team clarity. Not process, not tooling, not a new dashboard. Whether the people running the business day to day actually agree on what the next 18 months are supposed to look like. 

This sounds soft next to pricing and cost base, I know. But almost every operating partner I respect tells some version of the same story. They walk into a portfolio company, sit through a few management meetings, and discover the CEO and the CFO have two entirely different mental models of what the growth plan actually is. Not different opinions on execution. Different plans. One thinks the story is expansion into a new vertical. The other thinks the story is deepening penetration in the existing base. Both of them have been nodding along in board meetings for months. 

You can’t fix pricing or cost base cleanly on top of that kind of misalignment, because you don’t actually know which plan you’re optimising for. Every other lever you pull is downstream of this one, whether anyone admits it or not. 

The practical version of this, if you want something you can actually do in week one: sit down with the CEO and the CFO separately, not together, and ask each of them to describe the next 18 months in their own words. Not the board deck version. The version they’d give a friend over a drink. If those two descriptions don’t roughly rhyme, you’ve found your actual first project, and it’s not the one anyone put in the 100-day plan. 

Pricing and cost base still matter. They’re just second and third, not first. Get the alignment right and the other two levers move faster and stick better, because you’re no longer fighting an argument the management team hasn’t had yet.