How to Evaluate New Sales Reps in Their First 30 Days

How to evaluate new sales reps in their first 30 days: the answer is not deal count. It is three leading indicators — response speed, script adoption rate, and knowledge base usage. Cross-border orders often span months. A deal in the first 30 days depends heavily on luck and lead assignment. Judging a person by outcome metrics at that stage means making decisions on noise. Below, we break down each indicator's measurement rules, thresholds, data sources, and how to run the day-30 conversation.

Why deal count fails as a 30-day KPI

A hardware parts team had a new rep who, on day 28, landed a Middle East customer who urgently needed to restock. The deal closed at $3,000. The manager made him the benchmark and gave him the best inquiry pool in month two. By day 60, he had 17 customers, and 9 of them had not even had their needs clarified. His follow-up notes said only "quote sent." Another new rep was let go on day 30 with zero deals. But the German customer he was working with placed a trial order on day 45 because their internal certification process had to run its course. That timeline is not something a salesperson can compress.

Both cases point to the same thing: within 30 days, closing is an outcome variable, not a behavior variable. Its variance comes from three places:

• Lead quality: whether the rep gets trade show old contacts or cold outreach lists makes a multiple difference in close probability.
• Product cycle: fast-moving consumer goods can close in two weeks; industrial parts require samples, certifications, and internal approvals.
• Luck: happening to encounter several customers with immediate needs within 30 days.

When the sample size is insufficient, any outcome metric is unreliable. What you should look at are behavioral assets — response speed, script accumulation, and customer profile completeness. These are the leading variables for later deals, and they already have enough data volume by day 30.

Two common misjudgments: firing someone on day 30 with no deals, cutting the person who would have closed on day 45; and treating someone who stumbled into a big customer as a benchmark when their follow-up process is a mess. Both errors come from the same action — substituting outcome metrics for behavior metrics.

Leading indicator one: response speed — a hard threshold under time zones

Break "first response time" into two numbers:

1. Online coverage: how long between the customer sending a message and the sales rep seeing it. This is solved by scheduling, not individual effort.
2. Handling efficiency: how long between the rep seeing it and sending the first effective reply. This is solved by templates and script preparation.

Many managers only track a single total, making it impossible to tell whether it is a scheduling problem or a capability problem. Once you separate them, the coaching direction becomes clear.

Two details in measurement are easy to get wrong. First, look at the median, not the average. A rep replies to 20 messages a day: 19 within 5 minutes, one after 8 hours. The average gets pulled up to half an hour by that one message; the median stays at 5 minutes. The median reflects the norm; the average is held hostage by extremes. Second, track response distribution for night and weekend inquiries separately. Cross-border customers span multiple time zones, and the periods where deals are most often lost are exactly these non-working hours.

A workable threshold:

• Median first response during working hours within 10 minutes.
• First response outside working hours within 30 minutes of the next workday start.
• Two consecutive weeks below standard triggers a coaching process, not immediate termination.

Coaching content is not "be faster." It is specific actions: pre-save reply templates for high-frequency questions, turn on mobile message alerts, and confirm the unread message list during shift handover. Response speed is a habit problem. Habits can be trained, but you need to give concrete handles.

Leading indicator two: script adoption rate — a better signal of learning ability than "how much they said"

Having new reps write their own scripts and then comparing them is too subjective; the manager's preferences contaminate the scoring. A more objective approach is to look at their use of and deviation from the team's validated scripts: how many they adopted, in which scenarios they deviated, and whether the customer continued replying after the deviation.

If the team uses a tool with AI suggestions, adoption rate is a ready-made behavioral data point. For example, Sellenca overlays on WhatsApp Web and drafts replies based on the company's own product knowledge base. The rep confirms before sending — that "confirm" action naturally records adoption or rejection, without the manager manually scrolling through chat logs to count. Under measured conditions, the 365nails team's AI suggestion adoption rate is 97%, with an average of 1,973 AI calls per month. These numbers are generated automatically by the system, not filled in at month-end. To see how this is recorded and presented, check the features page.

High adoption rate does not mean copying. Combine it with "customer reply rate after adoption":

• High adoption, high customer reply rate: scripts match the scenario, the rep has no execution problem.
• High adoption, low customer reply rate: the script itself does not match this scenario. What is needed is knowledge base supplementation, not criticism of the rep.
• Low adoption, low customer reply rate: the rep is writing original scripts with poor results. A script coach needs to step in.
• Low adoption, high customer reply rate: the rep may have a personal style that works. Their replies are worth feeding back into the knowledge base.

These four combinations correspond to four management actions, far more useful than looking at a single percentage.

Leading indicator three: knowledge base usage — judging whether this person can accumulate, not just consume

New reps encounter repeated questions daily: price, MOQ, lead time, certification. Whether they actively look things up and add new answers determines whether team knowledge compounds or becomes repetitive labor.

Three quantifiable actions:

1. Number of Q&As added or corrected per week.
2. Number of knowledge base queries.
3. Number of times their contributed Q&As are referenced by colleagues.

In the first 30 days, do not demand volume — demand existence. For example, require at least 2 valid Q&As per week. "Valid" means a customer actually asked, the rep looked it up or asked someone, and then organized the answer into the base.

The negative signal is typical: when stuck, they ask the manager directly, do not record the answer, and ask the same question the following week. This type of person will likely still be in the same place on day 60 because they treat the knowledge base as someone else's job. Catch it early and coach early, rather than waiting until day 60.

There is also a hidden benefit: new reps with high knowledge base usage often also have high script adoption rates, because they know where to find answers. The two indicators correlate but do not fully overlap — some query a lot but never contribute; some contribute but rarely query. Handle the two cases separately.

Putting the three indicators into a 30-day scorecard

Evaluate in three 10-day segments:

Days 1–10: Look only at response speed and profile completeness. Allow clumsy scripts; do not allow missed replies or missed customer information. Profile completeness means whether fields like customer region, intent, and customer type are filled in — not how beautifully.

Days 11–20: Add script adoption rate and knowledge base contribution. Require at least 2 valid Q&As per week, and stable first-response median. This segment is the watershed: you can see who is actively learning and who is passively waiting for instructions.

Days 21–30: Look at the combined trend of the three indicators — continuous improvement or stagnation. Also observe the execution rate of the follow-up list, meaning whether the rep followed up with the people they should have. Deals in this segment count only as bonus points, not as a passing line.

With this table, the day-30 judgment is based on trend, not a single data point. A rep whose first-response median was 25 minutes in week one and dropped to 8 minutes by week three is improving. Another who stayed steady at 12 minutes is stable. Neither should be fired for "no deals."

Where the evaluation data comes from: stop making managers count chat logs by hand

Manually scrolling through chat logs to count first responses and adoption takes managers hours each week, and the definitions are inconsistent — one person counts "seen to replied," another counts "message arrived to replied." The data cannot be compared across reps. Once you have several new hires, this method collapses.

The workable approach is to let the tool record behavior. Automatic customer profiling with six-dimension segmentation, daily follow-up lists, and AI suggestion adoption or rejection are natural evaluation data sources. The management side also supports team conversation reviews and deal funnels, so managers do not have to count themselves. In Sellenca's production environment, the 365nails team accumulated 960+ customer profiles and 907 knowledge base Q&As. These numbers are themselves byproducts of evaluation data.

The criterion for choosing a tool is simple: whether the rep needs to change numbers, whether they must migrate to the Business API, and whether their chat habits are forced to change. The greater the change, the less trustworthy the first-30-day data, because the behavior itself is distorted by the tool — the rep is busy adapting to a new interface, response speed naturally drops, and what you measure is tool adaptation cost, not sales ability. Sellenca is a Chrome extension that overlays directly on WhatsApp Web. Reps do not change numbers, do not migrate APIs, and their chat habits remain unchanged. Only then does the measured behavioral data approach real performance. To confirm what the data panel looks like, book a demo to see the actual interface.

Execution: how to run the day-30 conversation

Give data first, conclusions second. Put the trend charts of the three indicators on the table and let the new rep say which week they improved and which week they got stuck. The manager listens first, then adds observations. This way the conversation is a check-in, not a verdict.

Three categories of handling:

• All three indicators pass and trend upward: add leads, assign harder customers. This person deserves more weight.
• Indicators stagnate but attitude is engaged: extend the observation period and assign a script coach. Stagnation may be a method problem, not a will problem.
• Response speed consistently below standard: this is a habit problem, usually difficult to reverse within 30 days. You can clearly state it is a red line and give a short improvement window.

Explain the evaluation criteria on day one of onboarding. Then day 30 is not a trial but a check against known rules. New reps know what they are being measured on, and their behavior becomes more focused.

FAQ

Should a new rep with zero deals in 30 days be fired?

Look at the three leading indicators first. If response speed meets standard, script adoption rate is rising, and there is knowledge base contribution, behavioral assets are accumulating and deals are only a matter of time. Do not fire. If all three indicators stagnate or worsen, and response speed is consistently below standard, that is the case to address. Deal count alone is not grounds for termination because the 30-day sample is insufficient to support that judgment.

Will response speed evaluation push reps to send junk replies to pad numbers?

Yes, if you only measure "first response time" without looking at reply quality. The solution is to add a companion metric: whether the customer continues replying after the first response. Sending "I'm here, how can I help?" can indeed lower the first-response median, but if the customer does not engage, that reply is ineffective. Look at first-response time together with customer reply rate after first response, and the room for padding shrinks.

Our team only has three to five people. Do we need such detailed indicators?

With fewer people, you need them more. With three to five people, managers have more room to judge by impression, making it easy to wrongly fire or wrongly keep someone. The three indicators do not require a complex system — a single spreadsheet can track them. The key is consistent definitions. If you consider a tool, Sellenca charges per seat: $19 per seat per month, or $190 per seat annually. For three to five people the cost is manageable. See the pricing page for details. With fewer people, each decision carries more weight. Data is steadier than impression.

What script adoption rate counts as passing?

There is no universal number because team script library maturity varies. New teams have incomplete libraries, so adoption rates are naturally low. Mature teams cover most common scenarios, so adoption rates should be higher. A more practical approach is to look at the trend: compare week one with week three. Is adoption rising? Rising means the rep is learning the team's validated methods. Stagnation means they are either writing their own or have not found the script library entrance.

---

The core of 30-day evaluation is switching the basis of judgment from outcome metrics to behavior metrics, so the day-30 conversation has data to rely on rather than impression. Sellenca automatically records behavioral data such as AI suggestion adoption, customer profiling, and follow-up lists, so managers do not have to count chat logs by hand. A 7-day full-feature free trial is coming soon. You can book a demo to see the data panel, or visit the pricing page to understand seat costs.