How Do I Score My SDRs on the Activities That Actually Book Meetings?

Signals you actually need this

The clearest signal is a leaderboard that has stopped correlating with the calendar. Your top dialer places 140 calls a day and books two meetings a month; your quietest rep places 55 and books nine. If you rank the team by activity and rank them again by held meetings and the two lists barely overlap, your current scorecard is measuring effort instead of outcome, and every coaching conversation you have is aimed at the wrong variable.

A second signal: SDRs who can recite their dial number to the decimal but cannot tell you their connect rate, their conversation-to-meeting ratio, or how many of last month's booked meetings actually held. That asymmetry is not a memory problem. People memorize the number they are judged on. If dials are the only line with teeth, dials are the only line anyone tracks.

Watch for gaming patterns, which are the loudest signal of all. Sixty-second calls placed at 7:50 p.m. to numbers that were never going to answer. Sequences fired at a list nobody scrubbed. Meetings booked with a junior analyst who has no budget authority, because the comp plan pays on booked and says nothing about held. None of this is a character defect — it is rational behavior under a badly specified metric. Reps optimize what you count, and they will find the cheapest path to the number every time.

The no-show rate is the fourth tell. If booked-to-held sits below roughly 60 percent, the problem is almost never the calendar tool. It is qualification: your SDRs are booking anyone who says "sure, send an invite" because that click is what pays. A team scoring on held meetings usually runs 70 to 85 percent show rates, because reps start confirming, re-confirming, and disqualifying the soft yes before it wastes an AE's hour.

Then there is the ramp signal. If new SDRs take five or six months to hit quota and nobody can articulate why one ramps in nine weeks and another in twenty-two, you have no diagnostic instrument. A multi-line scorecard is exactly that instrument — it tells you the new rep is fine on volume and personalization but dies at the objection, which is a two-week coaching fix rather than a six-month mystery.

Finally: forecast noise. When RevOps cannot predict next month's meeting count from this month's activity within a reasonable band, it is usually because the activity being counted has no causal relationship to the outcome. A weighted matrix built on the real drivers restores that predictive link, and suddenly top-of-funnel becomes a plannable input instead of a monthly surprise.

What good looks like versus what bad looks like

Bad is one number with a threshold: 80 dials a day, hit it or explain yourself. It is easy to administer and easy to defeat. It also produces a specific failure mode where the best relationship-builder on the team looks like the worst performer, because she spends nine minutes on a call that turns into a six-figure opportunity while her teammate racks up ninety twenty-second voicemails.

Good is a matrix of six to nine lines, each weighted, each scored 1-to-5, rolling into a composite. A workable starting distribution for an outbound SaaS team: meetings held 30 percent, opportunities created or accepted 20 percent, live conversations 15 percent, connect rate 10 percent, personalized emails or sequence quality 10 percent, quality outbound calls 10 percent, and 5 percent for CRM hygiene and pipeline notes. The weights should sum to 100 so the composite reads as a percentage-like score everyone understands intuitively.

The level definitions matter more than the weights, and most teams skip them. A level 3 must be written down as a specific, observable standard — "connect rate between 8 and 12 percent," not "meets expectations." Level 5 is the top decile of your own team's actual historical performance, not an aspiration copied from a benchmark report. Level 1 is a documented gap that triggers a coaching plan. Without written definitions, managers score by vibe, the numbers drift by manager, and the whole thing loses legitimacy inside a quarter.

Bad scorecards are also invisible. If the rep sees the score for the first time in a monthly review, it is an evaluation, not a system. Good ones are published — every SDR can pull up the matrix any day, see their level on every line, see the team's distribution, and know precisely which single line moves their composite most. That visibility is what converts a scorecard from an HR artifact into a behavior engine.

The other split is leading versus lagging balance. A matrix made only of lagging indicators (held meetings, accepted opps) gives a rep nothing to do on a Tuesday morning. One made only of leading indicators is the dial-count trap in a fancier outfit. The working ratio is roughly half the weight on outcomes and half on the controllable inputs that produce them, so the rep always has both a destination and a next action.

Real cost and ROI ranges

The cheapest version is free and takes an afternoon. A spreadsheet with KPI rows, a weight column, a level column per rep, and a SUMPRODUCT formula produces a legitimate composite score. The real cost is maintenance: someone has to pull connect rates and held-meeting counts out of the CRM and dialer every week, and that job is roughly two to four hours weekly for a team of eight. Spreadsheets die from neglect, not from bad math — the sheet that nobody updated after the sequence change is the most common failure in this whole category.

Tooling costs sort into two buckets, and it is worth being clear about which one you are buying. Visibility tools — scorecard, leaderboard, and gamification platforms like Ambition, Spinify, or SalesScreen — put the matrix on TVs and in Slack and pull the inputs off the CRM automatically. Pricing in this category is commonly quoted per user per month and often lands somewhere in the low-to-mid tens of dollars, though most vendors quote rather than publish. Compensation tools — QuotaPath, Xactly, Spiff — put teeth in the score by paying against multi-component plans; QuotaPath publishes a free tier and paid plans starting in the low tens of dollars per user, while the enterprise comp platforms are quote-only. Engagement platforms like Salesloft or Outreach are the data source rather than the scorecard: they capture every call, email, connect, and booked meeting, and you build the composite on top of what they log.

There is also a conversation-intelligence layer — Gong being the best-known — that scores call quality rather than call count. It answers the question a volume metric structurally cannot: is the rep asking for the meeting, and how are they handling the second objection? That signal is expensive relative to the rest of the stack, but it is the only automated way to grade the *quality* line on your matrix instead of having a manager listen to recordings.

The ROI math is straightforward enough to do on a napkin, and you should do it before you buy anything. Take an SDR fully loaded at, say, $75,000 a year. If the matrix moves held meetings from 12 to 15 a month for that rep, that is 36 additional held meetings a year. At a 25 percent meeting-to-opportunity rate and a 20 percent opportunity win rate, that is roughly 1.8 additional closed deals annually per rep from one person's improvement. Run that across eight SDRs and the arithmetic almost always dwarfs a per-seat software cost. Substitute your own conversion rates — the point is that the calculation is available to you today, and any vendor claim you cannot reproduce with your own numbers is not worth paying for.

Watch the hidden costs too. Re-weighting a comp plan mid-quarter has legal and morale implications; most teams run the matrix as a *coaching and visibility* instrument for a quarter before wiring dollars to it. Data quality is the other tax: a composite built on activity that reps log manually will be wrong, and cleaning up CRM logging discipline before you launch is unglamorous work that determines whether anyone trusts the number.

How it plugs into your workflow

Start with a definition session, not a tool purchase. Get the SDR manager, RevOps, and whoever owns comp in a room and list every behavior that precedes a held meeting in *your* motion — not a generic list. Cap it at six lines for the first version. Teams that launch with eleven KPIs almost always retreat to six by month two, having burned credibility in between.

Next, instrument before you score. For each line, name the exact system of record and the exact field: connect rate comes from the dialer, meetings held comes from the calendar or CRM meeting object with a disposition field, opportunities accepted comes from the AE's acceptance stage. If a line has no automated source, either accept that a manager will grade it subjectively — which is fine for call quality, with written level definitions — or cut it. Do not put a line on the matrix that nobody can compute.

Then run a shadow quarter. Score the team weekly, publish the composite, coach off it, and pay nothing against it. This is where you discover the weights are wrong — that connect rate is 15 percent of the score but explains almost none of the variance in held meetings, while personalization depth explains most of it. Adjust, and only then consider attaching money.

The weekly rhythm is where the system earns its keep. In each one-on-one, pull up the rep's matrix, identify the single lowest-weighted-impact line, and agree on one behavioral change for the week. One line, not four. The composite gives the conversation a spine: instead of "you need to work harder," it is "your conversations-per-connect is a 2 and it is 15 percent of your score — here are three call recordings where the transition to the ask fell apart."

The re-weighting mechanism is the part most teams underuse. When you launch a new sequence, move upmarket, or shift from inbound follow-up to pure cold outbound, the behaviors that produce meetings change — and because the weights are yours, you change them overnight and the team re-aims the next morning. Announce the change with the reasoning, not just the new numbers. A matrix that changes without explanation reads as moving the goalposts; one that changes with a stated rationale reads as leadership actually paying attention.

Downstream, the same structure extends past the SDR seat. AEs get a matrix on discovery quality, multithreading depth, and next-step-set rate. Customer success gets one on health-check completion and expansion conversations. The same weighted-level logic works in a car dealership BDC scoring appointment-set and appointment-shown, in an insurance agency scoring producer quotes issued versus bound, and in an agency scoring proposal-sent versus signed. The mechanic is domain-agnostic: name the behaviors that cause the outcome, weight them, define the levels, publish the score.

Upstream, marketing should see this matrix. When connect rate collapses on a specific segment, that is a list-quality signal, not a rep-effort signal, and the matrix is the artifact that proves it. This is the RevOps payoff — one shared picture of what good top-of-funnel work looks like, with the same numbers in front of sales, marketing, and finance, so the argument about whether the leads are bad or the reps are lazy becomes an evidence question instead of a volume contest.

Related questions

How many KPIs should be on an SDR scorecard?

Six is the practical sweet spot; nine is the ceiling. Below five, reps game the remaining lines. Above nine, no one remembers their weights and managers stop scoring consistently. Launch with the five or six that most clearly precede a held meeting, then add lines once the rhythm is established.

Should the composite score drive compensation directly?

Eventually, but not on day one. Run one quarter as coaching-only to validate the weights against real outcomes. Then wire the variable portion — commonly the bonus above a base activity floor — to the composite. Paying against untested weights creates disputes you cannot easily unwind mid-quarter.

How do I score call quality without listening to every call?

Use a written rubric a manager applies to a small sample — three to five calls per rep per week — with level definitions tied to observable moments: did the rep state a reason for the call, surface a problem, handle the first objection, ask for a specific time. Conversation-intelligence tools automate this at higher volume.

What if two managers score the same rep differently?

That is a level-definition problem, not a manager problem. Run a calibration session: both managers score the same three reps independently, compare, and rewrite any level definition where they diverged. Repeat quarterly. Written, observable criteria eliminate most drift.

Does this work for a two-person SDR team?

Yes, with fewer lines. Use four or five — personalized emails, connect rate, conversations, meetings held — and score them in a shared sheet. Small teams get more value from the coaching structure than from the ranking, since there is no meaningful leaderboard at that size.

FAQ

What is the difference between a weighted scorecard and just tracking call volume?

A weighted scorecard assigns points across several behaviors — quality conversations, personalized outreach, opportunities created, meetings held — instead of counting raw dials. High call volume alone rarely produces booked meetings, because the cheapest way to hit a dial target is to place calls that were never going to connect. The composite reveals who is genuinely moving prospects forward and who is generating noise.

How often should I update the weights on my SDR scorecard?

Whenever the motion changes: a new sequence, a new segment, a shift from warm follow-up to cold outbound. Outside of those events, a quarterly review is enough. Because the matrix is published, the team re-aims the day after a change. The one caveat is comp — if dollars are attached, change weights at period boundaries, not mid-quarter.

What happens if an SDR scores high on dials but low on conversations?

Their composite lands low, and the matrix makes the reason unmissable. Instead of a vague "improve your quality" note, the rep sees that conversations-per-connect is their weakest line and that it carries real weight. That turns an argument into a coaching plan with a specific target, and it removes any payoff from inflating the easiest number.

How do I decide which KPIs belong on the scorecard?

Work backward from a held meeting in your actual process and list every step that precedes it. Then test each candidate against your own data: does variation in this metric correlate with variation in held meetings across your reps? Keep the ones that do. Drop the ones that everyone scores identically on — a line with no spread carries no information.

What if my SDRs push back on a new scoring system?

Involve them in setting the initial weights. Reps almost always know which activities actually produce meetings, and asking makes the matrix theirs rather than something imposed. Then publish it. Most resistance is fear of an opaque evaluation; a visible scorecard with written level definitions and a clear path from level 3 to level 4 usually converts skeptics within a few weeks.

Can this scoring approach work outside of SDR teams?

Yes. The logic — name the behaviors that cause the outcome, weight them, define levels, publish the score — transfers to AEs, customer success, dealership BDCs, insurance producers, and agency business development. Only the line items change. Any role with a multi-step process leading to a countable outcome is a candidate.

Sources

• https://blog.hubspot.com/sales/sales-metrics — HubSpot on selecting and interpreting sales metrics.
• https://www.salesforce.com/sales/analytics/sales-metrics/ — Salesforce overview of sales performance metrics and KPIs.
• https://hbr.org/2015/09/the-problem-with-financial-incentives-and-what-to-do-about-it — Harvard Business Review on incentive design and unintended behavior.
• https://www.gartner.com/en/sales — Gartner sales research on pipeline and seller performance.
• https://www.salesloft.com/ — Sales engagement platform documenting activity capture across calls and emails.
• https://www.gong.io/ — Conversation intelligence for scoring call quality and objection handling.
• https://www.quotapath.com/ — Commission and quota-attainment tracking, including published pricing tiers.
• https://sloanreview.mit.edu/article/dont-let-metrics-undermine-your-business/ — MIT Sloan Management Review on metric fixation and surrogation.
• https://www.saleshacker.com/ — Practitioner writing on SDR process and outbound benchmarks.
• https://support.google.com/docs/answer/3093502 — SUMPRODUCT documentation for building a weighted composite in a sheet.

Related on PULSE

• [How Many Sales Reps Do I Need to Hire for My Logistics Company?](/knowledge/tl0058)
• [How Many Salespeople Do I Need to Hire for My Car Dealership?](/knowledge/tl0052)
• [How Many Producers Do I Need to Hire for My Insurance Agency to Grow My Book?](/knowledge/tl0015)
• [How Do I Figure Out How Many People to Schedule Each Day and at What Times for My Single Store?](/knowledge/tl0002)
• [How Many Attendants Should I Schedule Each Day at My Car Wash?](/knowledge/tl0067)