Leader Insights

What's at risk this week, and what's it worth?

Which reps need coaching, and on what?

Where are deals getting stuck, and what can I do about it?

Are we going to hit the number?

Persona is structural, not text-based. I figure out who's a leader by reading the explicit leader assignment on each team in Settings → Teams. I don't match on role titles or job titles.

Admins always get the CRO layout. Even an admin who happens to also be a team leader sees the CRO tab order, section order, and team chips. Drilling into a team via the chip selector shifts the data scope only, not the layout.

The team leader counts as part of their team. I include a team leader's own deals, coaching scores, and KPI contribution in their team's roster — even if the leader isn't separately added as a member of the team in Settings → Teams. So a sales manager will see their own deals on their dashboard, and admins drilling into a team see the leader alongside the team's members.

Edge case: multi-team leaders. If you're a non-admin who leads two or more separate teams, I fall back to the CRO/org-wide view, since your scope spans more than one team.

Every tab is team-filterable. Home, Pipeline, Coaching, Impact, and Trends all narrow to the team in scope — for sales managers, automatically; for admins, via the chip selector.

Waiting for enough data. When a team doesn't yet have enough recent activity for me to draw meaningful analysis (the first 24 hours after Leader Insights is enabled, or teams below my minimum activity threshold), I show a friendly "check back tomorrow" placeholder on Home / Coaching / Impact / Trends while more deal data accumulates. Pipeline keeps working in the meantime — it reads live from your deal list rather than waiting on my analysis pass.

Headline summary — a one-line read I write on pipeline health this week (e.g., "Pipeline at $4.8M — 5 deals at risk this week"). When some at-risk deals don't have a synced deal amount, I phrase coverage honestly (e.g., "4 of 6 at-risk deals have values").

KPI tiles — Newly at risk, Pipeline at risk, Closing, Team win rate. A single time-range toggle (Today / Week / Month / Quarter) switches all four at once. Each tile shows a delta vs the previous period.

Team Scoreboard (CRO only, org-wide view only) — one card per team with total pipeline, at-risk dollars, deal count, average health score (1–4 scale), and win-rate delta vs the org average. Hierarchy isn't required — flat-team orgs (no parent/child teams) get a card per team. Cards are non-clickable — drilling into a specific team is done via the header chip selector. I hide the section when an admin has drilled into a team, when the persona is Sales Manager, and when no team has its own data to compare (e.g. a single team that contains every active rep — its card would just mirror the org-wide row).

Top Deals to Escalate — the 3 deals I think are most worth your attention, with a one-line "why", a suggested next step, and (when available) a Leader Play action — a one-sentence prompt for what a sales leader should do on this deal based on historical lift data. Each card carries a win-probability chip (e.g. "Olli 62%") when my ML model has scored the deal.

Leader Plays — up to 3 at-risk deals where historical data shows leader engagement at this stage moves the win rate. Each card shows the deal, an action sentence I generated (≤160 chars), and an expandable "lift" panel comparing close rate when a leader engages vs when they don't, plus the sample size. I render plays only when the org has at least one team with a leader set in Settings → Teams and there are 10+ historical closed deals at the matching stage to draw a defensible signal. I count closed deals from your CRM — any deal whose CRM stage is a closed stage (won or lost) in the last 12 months counts toward the threshold, including deals that closed before Fluint started tracking the org.

Coaching Priority — the 2 reps I think have the biggest coaching opportunity, with the dimension they're weakest on. I only consider reps with enough active pipeline (see Coaching, below). When my ML model can predict the rep's recoverable pipeline, the card shows a "Olli: +$X lift if coached" badge — the dollar amount of currently at-risk pipeline I expect the rep can recover with focused coaching.

Weekly Movers — deals that flipped to at risk, recovered, or closed; reps whose scores improved or declined.

Deals without a pipeline stage — I exclude them everywhere.

Deals assigned to deactivated reps — I exclude them from Pipeline, Top Deals, KPIs, and coaching.

Deals with an unrealistic time in pipeline — very new or very old deals outside sensible bounds get dropped from Home and Pipeline metrics.

Deals without a synced deal amount — I count them, but exclude them from dollar sums (they render "—" / "Not synced" in the UI so you can see which ones).

If a KPI's delta looks off, check how long the org has had Fluint — I need two snapshots at both time points to render a delta.

Clicking a Top Deal card takes you straight to the pipeline with the deal filtered and the detail sheet open.

Forecast tile — total active pipeline, stuck value, and % of pipeline stuck. I exclude outliers (deals stuck in a stage far longer than anything else in that stage) so the numbers stay defensible.

Top 3 stuck deals — longest-time-in-stage deals that need a nudge. When my ML model has a recovery score for the deal, the card adds an "Olli puts recovery at X%" line — the probability I think the deal can still close given its history.

Deal table — every active deal, with:

Click any row to open the deal detail sheet — health rationale, stakeholders, objections, competitors, dimension scores, a 6-snapshot trajectory sparkline, and (when my ML model has a score) my read on the deal described below. I pull this data live, so it always reflects the current state of that deal.

Win probability — my chance-of-winning percentage for the deal (e.g. "62% chance of winning this deal").

What's helping / What's hurting — the top drivers I see, split into positive (green) and negative (red) factors. Each driver shows the feature name and a Strong / Moderate / Mild magnitude bar.

Past deals like this one — up to 3 closed deals from your history that look most similar to this one. Each row shows a Won/Lost badge, the deal name, and how long ago it closed.

The table paginates 25 deals at a time. Use filters and search to narrow down before paging.

The stage progress bar reads your actual pipeline stages from your org settings — the number of segments matches your configured stages.

If a deal shows "—" or "Not synced" instead of a dollar amount, that deal doesn't have a value in the CRM yet. Set it in the CRM (or edit it in Fluint) and the amount will flow in on the next sync.

Sales Managers see only deals owned by reps in their team subtree, plus their own deals as the team leader. Admins drilling into a team via the header chip selector see the same view filtered to that team. The deal table updates immediately when an admin changes the team chip — I don't wait on the nightly refresh for Pipeline scope changes.

Rep rows — each rep's active deal count, a 6-week sparkline, and their score on each of 8 dimensions (Champion, Economic Buyer, Decision Criteria, Decision Process, Identify Pain, Metrics, Competition, Paper Process).

Cell colouring — green = strong, paper = average, sienna = weak. I highlight the weakest cell per rep.

Org average sits at the top of each column so you can see who's above and below the team line.

Expand a rep to see one of three panels I render:

Sparklines need at least 3 of the last 6 weeks of data for me to render. New orgs will see them fill in over time.

I call out specific deals by name in coaching observations — use them as concrete talking points in 1:1s.

If a rep you expect to see isn't there, it's almost always the 3-deal minimum — check their active pipeline in the Pipeline tab.

Sales Managers see only the reps on their team, with their own row appearing on the heatmap alongside their reports. Admins drilling into a team via the header chip selector see the same view — the team's members plus the team leader. I compute the org averages shown on the heatmap against everyone in scope — so a sales manager sees their team's averages (including themselves), while an admin viewing org-wide sees the full org's averages.

My 4-week forecast strip (top of the pipeline-value chart) — my forward look:

Pipeline value over time — total active pipeline week by week

Win rate — rolling 30-day win rate

Health mix — stacked area of strong / moderate / at risk / critical deals per week

Trends work best with at least 4 weeks of snapshot history. Weeks with no data render greyed out rather than as zero.

Trends respects team scope. Sales managers see their team's trends, admins drilling into a team via the chip selector see that team's trends, and the org-wide view shows all-org trends.

Cells with no closed deals in that combination show "Not set".

Cells with fewer than 5 closed deals show the value tagged (n=X) as a small-sample caveat.

Cells with 5 or more closed deals render normally.

Total active pipeline — committed revenue across all open deals

Expected close — probability-weighted close for the quarter (Σ of win rate × current value across cells)

Quota — sum of the team's quarterly quotas

Coverage gap — quota − expected close

HubSpot quota requires Sales Hub Pro+. The HubSpot Goals API is only available on Sales Hub Pro and Enterprise tiers — Starter and Free portals will return no quota even after reconnect.

HubSpot list-price reads coalesce in this order: hs_list_price, then hs_pre_discount_amount, then price. I use whichever is populated first on each line item.

Health signal (strong / moderate / at risk / critical) is my qualitative read — built from deal-digest narratives, stakeholder maps, momentum, and the like. It answers "how does this deal look on the surface?"

Win-probability / recovery rate come from my ML model, trained on historical close patterns and your org's own deal data when available. It answers "based on patterns from similar deals that closed, what do I expect here?"

Admins — I default to org-wide answers. To ask about a specific team, just name it: "show me the East team's pipeline" — I'll scope my response to that team.

Sales managers — I scope my answers to your team automatically. You don't need to specify your team name.

"Summarise this week's movers"

"Which of these deals has a champion problem?"

"Why is Acme flagged at risk?"

"Who should I coach on discovery?"

"Draft talking points for my 1:1 with Sarah on her weakest dimension"

"How is the West team's pipeline trending?" (admin only — I'll scope to the named team)

"What's your win probability on the Acme renewal?"

A small timestamp in the header tells you when I last computed the data, in both relative ("2 hours ago") and absolute form.

If the data is more than ~26 hours old, I'll show a yellow staleness banner. My overnight refresh may have failed — contact support if the banner persists.

The deal detail sheet is the one exception: open a deal and you'll see live, current-state data (health, stakeholders, trajectory) pulled at the moment you open it.

Admins and team leaders only. Reps who aren't admins and aren't set as the leader of any team in Settings → Teams won't see Leader Insights in the menu.

Non-admin multi-team leaders fall back to the org-wide view. If you're listed as the leader of two or more disjoint teams I show CRO/org-wide data, since your scope spans more than one subtree.

Nightly refresh. My numbers can be up to ~24 hours stale. The detail sheet is live.

Sparse history at launch. Some sections need several weeks of history before I can fill them. Expect "insufficient data" or hidden sparklines in your first few weeks — this is expected, not a bug.

Weekly Movers and KPI deltas need at least two snapshots at both time points. Brand-new orgs may see them empty for the first week or two.

Coaching sparklines render once at least 3 of the last 6 weeks have data per rep.

Impact matrix cells with fewer than 5 closed deals show (n=X) as a small-sample caveat.

Forecast tile appears on the Win Rate view only. Switch to Win Rate to see expected close and coverage gap.

Coverage gap only renders when at least one rep has a quarterly quota available via CRM sync.

Cell click-through is Win Rate only. ACV Protection and Cycle Time cells are not clickable in this release.

Salesforce customers need Collaborative Forecasting enabled on their Salesforce org for quota to flow through. Orgs without CF will see an empty forecast tile.

HubSpot customers: ACV Protection and forecast quota stay empty until the HubSpot integration is reconnected to grant the new line_items and goals permission scopes — and the data only fills in on the next nightly refresh after reconnecting. I'll show a reconnect callout on the ACV view when I detect the issue. HubSpot quota additionally requires Sales Hub Pro+ (the Goals API is gated to that tier).

ACV cells show "Not set" for deals without list-price line items in the CRM, or (n=X) when fewer than 5 closed deals exist for that cell.

Deals without a pipeline stage assigned in your CRM are excluded from every tile, matrix, and chart.

Deals whose assigned rep has been deactivated are excluded from Pipeline, Top Deals, KPI tiles, and Coaching.

Deals without a synced deal amount are counted where counts matter but excluded from dollar sums — I render them as "—" / "Not synced" in the deal table.

Deals with an unrealistic time in pipeline (too new or too old) are filtered out of Home and Pipeline aggregations to keep the headline trustworthy.

Reps with fewer than 3 active deals don't appear as rows in the coaching heatmap.

Deals with fewer than 3 scored dimensions are filtered out before I compute coaching averages.

On Track reps (close to the team line) get a short explanatory panel rather than a coaching narrative — this is by design; not every rep needs coaching every week.

Team Scoreboard renders when the org has at least one team in Settings → Teams. Hierarchy isn't required — flat-team orgs see a card per team. The Scoreboard hides only when no team has its own data to compare (e.g. a single team that contains every active rep — its card would mirror the org-wide row).

The team chip selector is admin-only and is hidden if the org has no teams in Settings → Teams. Sales managers don't have a team picker — they're locked to their team subtree.

Leader Plays require team leadership data. I render plays only when at least one team has a leader set in Settings → Teams. Specific stages also need 10+ historical closed deals to compute a defensible win-rate lift; cards skip stages that don't meet that threshold.

Waiting for enough data. When Leader Insights is first turned on, or when a team hasn't seen enough recent activity for me to draw meaningful analysis, I show a "check back tomorrow" placeholder on the team's Home / Coaching / Impact / Trends tabs until the next nightly refresh has more data to work with.

I don't show ML scores for every deal. Brand-new deals, deals with thin field data, or orgs that don't have enough history for me to train on yet may see chips and panels hide. I always fall back to qualitative health signals when an ML score isn't available — nothing breaks.

Org-specific vs generic models. When your org has enough closed-deal history, I use a model trained on your own deals. Until then, I fall back to a generic model trained across orgs. The generic version is still useful but tends to be less specific to your deal mix.

Forecast strip confidence range and accuracy line. Both render when my model has the underlying readings; I render the central pipeline figure unconditionally.

Sales managers don't get a team selector. Managers leading multiple teams fall back to the org-wide view, since their scope spans more than one team.

Confirm you're either an admin (Team Settings → User management) or a team leader (listed as the leader of a team in Settings → Teams).

Leader Insights is enabled per organization. If your org hasn't been turned on yet, contact support.

The first time Leader Insights runs for your team, you'll see a "check back tomorrow" placeholder while I wait for enough activity to draw meaningful analysis. This clears after the next nightly refresh.

If the empty state persists more than a day, your team may not have enough recent active deals to meet my minimum threshold. Check that deals are assigned to reps on your team and have a pipeline stage set.

The Pipeline tab still works in the meantime — it filters live, so you can use it to confirm deals are flowing.

I automatically include the team leader in their team's roster, so your own deals should appear alongside your reps' deals. If they're missing right after Leader Insights was first enabled or right after you were assigned as the team's leader, wait for the next nightly refresh — I rebuild Home / Coaching / Impact on that schedule. The Pipeline tab updates immediately since it filters live.

For admins drilling into a team via the chip selector: I always include the team leader in the team's data, even if they aren't separately listed as a member of the team in Settings → Teams.

I hide the chip selector when your org has no teams in Settings → Teams. Add at least one team to see it. Hierarchy is optional — a flat list of teams works.

I hide the Scoreboard whenever you've drilled into a specific team via the chip selector. Click the All teams chip to bring it back.

I also hide it when no team has its own data to compare against the org. The most common cause: your org has a single team that contains every active rep, so the team card would mirror the org-wide row. Add a second team (or move some reps off the first) to see per-team cards.

Confirm at least one team in Settings → Teams has a leader set. Without team leaders, I have no leader-engagement signal to draw from.

Specific stages need 10+ historical closed deals (won or lost) for me to compute a reliable win-rate lift. I count closed deals from your CRM — historical deals imported from your CRM count toward the threshold even if they closed before Fluint started tracking the org. Cards will appear as your closed-deal history accumulates.

An at-risk deal won't get a Leader Play if its current stage doesn't meet that 10-deal threshold yet — that's by design.

A "leader engaging" means a participant on a meeting tied to the deal who's set as a team leader in Settings --> Teams. If your team leaders haven't joined any meetings on closed deals (or their meeting attendance isn't being attributed to them in Fluint), I can't form the lift signal.

Check the timestamp in the header. Anything under ~24 hours old is expected.

If the yellow staleness banner is showing (>26 hours), my overnight refresh likely failed — contact support with the timestamp.

The deal detail sheet is always live; open a specific deal to confirm its current state.

This is expected at launch. I aggregate historical snapshots, and new orgs only have a handful. Most sections fill in as more snapshots accumulate (daily).

Coaching sparklines need 3+ of the last 6 weeks. Trends charts need 4+ weeks. Impact cells below 5 closed deals are marked (n=X).

The rep needs 3 or more active deals for me to include them as a row. Check the Pipeline tab filtered by that rep's name.

If the rep was recently deactivated, all of their deals are excluded — reactivate them in User Management or reassign the deals.

Deals without a pipeline stage set in the CRM are excluded. Open the deal in your CRM, assign it a stage, and it'll appear after the next nightly refresh.

Deals assigned to a deactivated rep are excluded. Reassign the deal to an active user.

Deals that are extremely new or extremely old (outside sensible pipeline-age bounds) are filtered from Pipeline aggregations — expected for brand-new or long-abandoned deals.

Filters or search on the Pipeline tab may be hiding it — clear filters and try searching by company or rep name.

That deal doesn't have a value in the connected CRM. Set the amount in the CRM (or edit the deal in Fluint) and it will flow in on the next nightly refresh.

Deals without a value still appear in the table, but don't contribute to dollar sums on KPIs, forecasts, or headlines. The headline copy reflects how many at-risk deals have values ("4 of 6 at-risk deals have values").

Make sure you're on the Win Rate view — the forecast tile only appears there.

The coverage gap only renders when at least one rep has a quarterly quota available.

The forecast tile depends on CRM-synced quota data. Check Settings → Integrations to confirm your CRM is connected and the relevant quota fields are populated for your reps.

Salesforce customers need Collaborative Forecasting enabled in Salesforce for ForecastingQuota to flow through.

HubSpot customers need to reconnect the integration to grant the new goals scope, AND need Sales Hub Pro+ (the Goals API isn't available on Starter or Free tiers). After reconnecting, quota only fills in on the next nightly refresh.

If you see a HubSpot reconnect callout on the ACV view, reconnect the integration in Settings → Integrations → HubSpot. ACV cells fill in on the next nightly refresh after reconnecting, not immediately.

Salesforce / Dynamics: list price flows in from OpportunityLineItem / opportunityproducts automatically. Confirm the integration is connected and that your opportunities have line items / products attached in the CRM.

Dynamics specifically: only non-overridden prices count (products where ispriceoverridden = false). If every product on a deal has been manually price-overridden, that deal has no list price for ACV purposes.

Cells showing "Not set" mean no closed deals exist in that (stage, signal) combination yet — not a configuration issue.

That's my small-sample flag — fewer than 5 closed deals in that cell. The value still renders, but treat it as directional rather than reliable. The flag goes away as more deals close in that combination.

Deltas need two snapshots at both time points (this period and the previous period). Wait for the org to accumulate enough history and the arrows will appear.

When both values are zero (e.g., "0 newly at risk last week and this week"), I don't show a percentage — by design, since "-100% vs last week" on a 0-to-0 change is meaningless.

I use the same nightly-refreshed data the tab is rendered from. If the page has a staleness banner, so does my answer.

For the most up-to-date view of any single deal, open its detail sheet from the Pipeline tab — that data is live.

The deal probably doesn't have an ML score from me yet. New deals, deals with thin or missing CRM fields, and deals from orgs that don't have enough history for me to train on yet are the usual causes.

Health signal (strong/moderate/at risk/critical) and my ML score are independent — health signal renders even when ML doesn't, and vice versa.

That's expected, and it's the most useful kind of disagreement to investigate. The two signals look at different things: health signal reads qualitative narrative (champion strength, decision criteria, momentum), while my ML score reads historical close patterns from similar deals. When you spot one of these, it's a deal worth a closer look — start with the deal's detail sheet.

I render the central 4-week pipeline and at-risk values whenever I've scored your active pipeline. The confidence range (low–high) and the "Model accuracy last 30d" line each render only when my model has the underlying readings — they hide rather than showing zeros when not available.