Sign in
Set up for Customer service

AI for insights and analytics

AI analytics for customer service

The team's first contact resolution is 84 percent while the fourth largest account, ten weeks from renewal, is at 45 percent on a surge of tickets in one category. AI analytics reads the ticket export, the contract file and the revenue file, and answers the head of service: which accounts are in trouble, and which have gone silent.

Ask it in your own words

These are the questions service teams ask. Each one maps to a measure our tools compute from your files. The AI model chooses the measure and explains the result; the arithmetic is done by our code, and every total is checked.

The questionThe measure behind itWhat comes back
Which accounts are raising far more tickets than usual?Contact volume against own baselineTickets by account against its own baseline, with the category driving the surge.
Are we meeting each account's own SLA?SLA attainment, account's own SLAAttainment against the response and resolution times in each contract, not a team average.
Which accounts have gone silent?Silence listAccounts whose contacts fell to nothing against their baseline, by revenue, with the renewal date.
Where does effort exceed revenue?Effort per account against revenueTickets and hours per account against revenue, with the ones out of proportion.
What keeps coming back?Recurring causesRoot causes by count and by accounts affected, with the repeat contacts they generate.
Which tickets could have been deflected?Deflectable ticketsTickets matching a documented answer, by category and account.

A conversation, with the figures

Every figure below was computed by a tool from the rows and checked before it was shown. The lines sum; the percentages match; each line opens to the rows that make it.

Which accounts are surging?

121 tickets above baseline were raised this month across 3 teams, by accounts more than double their own normal volume. Enterprise team carries 61 of them, 50% of the surge.

Enterprise team6150% of the total
Mid-market team3831% of the total
Small business team2218% of the total
The 3 lines sum to1210 unexplained

Which enterprise accounts?

3 accounts account for 51 of the 61 excess tickets in the Enterprise team, 84% of it. The largest is ten weeks from renewal, and seventeen of its tickets are in one category.

Halcyon Systems2231 tickets against baseline 9; renewal in 10 weeks
Corvid Retail17SLA met on 71%; contract says four hours
Brightline Health12repeat contact rate 38%
These 3 are5184% of Enterprise team

Who needs to know this week?

The account manager for Halcyon, today, with the category and the SLA figures. The team lead takes the recurring cause to engineering. The surge list runs every Monday.

Account manager, HalcyonCall the customer with the category breakdown and SLA figuresToday
Team leadRaise the recurring cause with engineeringThis week
Head of serviceRun the surge and silence lists every MondayWeekly
Which accounts are raising far more tickets than usual?Are we meeting each account's own SLA?Which accounts have gone silent?Where does effort exceed revenue?

The measures behind the answers

Each measure has one formula, one source and one meaning. They are computed per account and team and in total, and every one carries an identity that must hold before it is shown.

MeasureFormulaFromWhat it tells you
Contact volume against own baselineTickets in the last 30 days ÷ the account's trailing 12-month monthly average; surges and dropsTicket export; account listAccounts in trouble, and accounts gone quiet
First contact resolution per accountTickets resolved without reopen or follow-up ÷ tickets, per account, above a minimum countTicket exportCustomers who have to ask twice
Repeat contact rateTickets reopened, or raised in the same category within 14 days of a closure ÷ ticketsTicket exportFixes that did not fix
SLA attainment, account's own SLATickets meeting the response and resolution times in that account's contract ÷ ticketsTicket export; contract termsBreaches where the terms are strictest
Effort per account against revenueSupport cost or hours ÷ account revenueTime records or handling time; revenueAccounts that cost more than they pay
Silence listAccounts with volume below a stated fraction of own baseline for 6 weeks, by revenueTicket export; account listUsage that has stopped
Deflectable ticketsHow-to tickets answerable by an article ÷ tickets, by categoryTicket export with categoryArticles worth writing
Time to resolution by severityMedian and 90th percentile, by severityTicket exportWhether serious issues close faster than minor ones
Backlog ageingOpen tickets by age band and ownerTicket exportWhere tickets wait
Recurring causesCategories with repeats across many accounts in the periodTicket exportProduct faults, with the account list attached

Each measure is worked through, with the export it comes from and what to drop, in KPIs for B2B customer service teams.

What the AI model does, and what our tools do

The AI model chooses the measure

From your question and the measures declared for customer service, the model picks the one that answers it, and the period and comparison the question implies.

Our tools do the arithmetic

Deterministic code reads the rows, computes the measure, and checks the identities below. The same question on the same data gives the same answer, every time.

The AI model explains, and cites

The AI model writes the sentence around the result, naming the account or account behind it. It states no figure that is not in the result, and every figure links to its rows.

The identities that must hold

TicketsTickets = first-contact resolved + repeat + open
SLATickets = met + breached + not yet due; every account has its SLA terms recorded
AccountsEvery ticket belongs to one account; unmatched tickets are counted, not dropped
BacklogOpening + raised − closed = closing

What it reads

The exports service teams already produce. Column names are mapped once and the mapping is reused. A file is the way in; scheduled delivery and connections to your systems come with the plan, and every source is listed here.

  • Ticket export
  • Account list
  • Contract terms
  • Time records or handling time
  • Revenue

Who owns each answer

An answer is a list with an owner and a cadence, or it is a chart nobody works.

MeasuresOwnerCadence
Surge and silence listsHead of service with account managementWeekly
SLA by account; backlog ageingTeam leadsWeekly
First contact resolution and repeats by account; recurring causesHead of service; productMonthly
Effort against revenue; deflectable ticketsHead of service; financeQuarterly

Questions service teams ask about AI analytics

Why an account's own baseline rather than a team average?

Because a team dashboard at 96 percent SLA and 84 percent resolution hides one account at 45 percent. Each account's normal volume, its own contract terms and its own history are the comparison. A surge is a change for that account, and it appears whether or not the team average moved.

How does it know each account's SLA?

From the contract file: response and resolution times per account. Attainment is computed against those, per account, and the team average is shown beside it. If an account has no contract terms loaded, the tool uses the default and says so.

What is the silence list?

Accounts whose contacts fell to nothing against their baseline, ranked by revenue, with the renewal date. Silence before a renewal is as much a signal as a surge, and it never shows on a ticket dashboard because there are no tickets.

What files does a service team need?

The ticket export with account, category, times and resolution, the contract file with SLA terms and renewal dates, and revenue by account. Hours per ticket, if logged, turn volume into effort against revenue.

See it on your data

Bring a few thousand rows. The data map opens next, every column mapped once, and the first question is answered in minutes. Free, in your browser, no account.