WFM guideOperations

Contact centre volume spikes

Q: What causes sudden volume spikes in a contact centre?

Sudden volume spikes in contact centres are typically caused by one of six event types: (1) Technology or service failure — a system outage, website down, payment processing failure, or service disruption that causes customers to call for status; (2) Media or social media event — a news story, viral social media post, or coverage of a product issue that triggers customer concern; (3) External communication the contact centre was not told about — a marketing email sent to 500,000 customers without WFM notification, a government announcement affecting the company's customers, a policy change communicated by letter without phone lines prepared; (4) Billing or statement event — a billing run that goes out at an unexpected time, an unexpected price increase on statements, or a billing error affecting large numbers of customers simultaneously; (5) Regulatory or compliance announcement — FCA or Ofcom announcement that customers interpret as requiring them to contact their provider; (6) Weather or emergency event — an event that affects the company's operations or its customers' situations simultaneously at scale.

Q: What is the emergency staffing response to a contact centre volume spike?

The emergency staffing response toolkit for a contact centre volume spike operates in tiers: Tier 1 (immediate, 0–30 minutes): cancel or delay breaks for agents currently available; pull cross-skilled agents from lower-priority queues; activate any agents already on-site but in training or meetings that can be released; promote IVR callback option to reduce trunk occupancy. Tier 2 (30 minutes to 2 hours): call-out agents on scheduled days off (requires prior agreement in employment contracts or through volunteer rosters); extend shift end for willing agents; contact agency or flex pool for same-day availability; activate the BPO overflow partner if available under the contract. Tier 3 (if spike will sustain beyond the day): activate the formal contingency staffing plan; notify management of projected SL impact and duration; communicate to customers via digital channels (website, social media, SMS) to set expectations and deflect contacts where possible.

An unplanned volume spike is a test of operations planning. It reveals whether the WFM team has a pre-approved emergency toolkit, whether the intraday controller detects the spike within 15 minutes or notices it at the end of the shift, and whether the communication protocol is documented or improvised. Operations that prepare for spikes perform better in them.

Six causes of unplanned volume spikes

Technology or service failure

Website down, payment processing failure, app outage, or service disruption triggers customers to call for status. The spike begins within minutes of the failure and sustains until the issue is resolved and customer confidence returns.

Early signal: Social media mentions increase sharply. IVR menu selection 'technical issue' or 'service not working' sees sudden volume increase. Competitor monitoring tools may flag the outage.

Typical magnitude: Can multiply contact volume 3×–15× depending on the scale of the failure and the proportion of customers affected.

Communication sent without WFM notification

A marketing email, direct mail campaign, billing statement, or external letter reaches hundreds of thousands of customers and drives a predictable contact surge — that the contact centre was not told about.

Early signal: Contact surge begins 30–90 minutes after a typical email open window (9–11am for most B2C sends). Spike is highly correlated with a specific contact reason (e.g. billing enquiry, promotional offer query).

Typical magnitude: Typically 1.5×–3× volume. Predictable if the contact centre had been told — this spike is a failure of cross-functional communication, not an unpredictable event.

Media or social media event

A news story, viral complaint on social media, or media investigation into the company triggers customer concern. The severity depends on media reach and customer risk perception.

Early signal: Social monitoring tools flag a spike in brand mentions. News alert notifications. Volume to the complaints or 'media enquiry' IVR option increases.

Typical magnitude: Highly variable. Localised social events: 1.2×–2×. National media coverage of a serious issue: 3×–10× over 24–72 hours.

Billing or payment event

A billing run goes out earlier than expected, an unusual charge appears on statements, or a price increase takes effect and appears on bills for the first time. Customers contact to query the charge.

Early signal: Billing team can identify the event. ACD contact reason categorisation shows 'billing query' or 'charge query' rising sharply. Volume peaks 1–4 hours after the statement delivery window.

Typical magnitude: 1.5×–4× depending on whether the charge is unexpected (higher spike) or communicated in advance (lower spike).

Regulatory or policy announcement

A government, regulator, or industry body announces something that customers interpret as requiring them to contact their provider — even if no action is actually needed. Energy customers contact after an Ofgem announcement; financial services customers contact after an FCA regulatory change.

Early signal: Regulatory monitoring. Industry body communications. Often the contact centre hears about it via agents receiving the calls before management receives the alert.

Typical magnitude: 1.3×–3×. Often lasts 2–5 days as media coverage of the announcement sustains customer concern.

Failure demand amplification

A previous service failure (unresolved complaint, billing error, delivery problem) drives repeat contacts at scale — customers who did not receive a satisfactory resolution are calling back. This can be invisible until the repeat-contact data is analysed.

Early signal: Increase in contacts from customers who have contacted within the previous 7–14 days. Wrap code analysis shows 'follow-up on previous contact' or similar increasing. FCR data deteriorates.

Typical magnitude: Depends on scale of original failure. Often 1.2×–1.8× over several weeks before the pattern is identified and root cause addressed.

Emergency staffing response toolkit

Response actions by tier — deploy in sequence as spike sustains

Tier 1: Immediate (0–30 minutes)

Zero cost; zero notice required

→Cancel or delay scheduled breaks for currently-available agents — every available agent on the floor during the first 30 minutes
→Pull cross-skilled agents from lower-priority queues (back-office, email) if the spike is in voice
→Release agents from non-essential internal meetings and training that can be rescheduled
→Activate IVR callback option if not already default — reduces trunk occupancy and shifts demand to a controllable callback queue
→Increase IVR deflection messaging: 'Our website can help with [issue type]' before offering queue option

Tier 2: Short-term (30 minutes – 2 hours)

Low-medium cost; 30-minute to 2-hour response time

→Call-out agents on scheduled days off — requires prior agreement in employment contracts or voluntary overtime register
→Extend shift end for willing agents — requires management authority and compliance with maximum working hours
→Contact agency or flex pool for same-day or next-day availability — requires pre-existing contract and agency readiness
→Activate BPO overflow partner if under contract — requires activation clause in the BPO agreement specifying lead time
→Deploy proactive digital communications (website banner, social media post, email) to deflect contacts where the issue can be resolved without calling

Tier 3: Sustained spike (if duration exceeds business day)

High coordination cost; requires management escalation and formal BCP activation

→Activate formal contingency staffing plan (documented and pre-approved in BCP)
→Notify senior management of projected SL impact, duration estimate, and customer communication plan
→Brief press/communications team if spike is media-driven — uncoordinated agent responses to media-related questions create additional risk
→Implement triage routing: priority access for vulnerable customers and regulatory-priority contact types; managed waits for standard service
→Consider temporary IVR message reducing service hours or directing customers to digital self-service for standard queries

Communication protocol during a spike

Minute	Intraday controller action	Who is notified	Information shared
+0–15min	Volume spike detected on real-time dashboard. Confirm spike is genuine (not a reporting artefact). Identify probable cause from ACD data (contact reason, IVR path, social media).	TLs on the floor	Volume is X% above forecast. Deploying Tier 1 actions. No breaks until further notice.
+15–30min	Assess duration projection. Is this a short burst (minutes) or a sustained event (hours)? Activate Tier 1 actions. Prepare Tier 2 actions for authorisation.	Operations Manager, WFM Manager	Volume at X%. Probable cause: [event type]. Duration estimate: [short/sustained]. SL currently at X%. Tier 1 deployed. Requesting Tier 2 authorisation.
+30–60min	Tier 2 actions authorised and activated. Monitor impact. Update duration estimate. Assess whether Tier 3 is needed.	Operations Manager, Comms/PR team (if media-related), Tech team (if outage-related)	Updated SL position. Tier 2 actions live. Estimated recovery time.
+60–120min	Sustained spike confirmed. Activate Tier 3 if duration extends beyond business day. Brief management on customer communication plan.	Senior Management, Head of Operations	Full incident brief: cause, volume multiple, SL impact, actions taken, customer communication plan, estimated recovery.
+Post-eventmin	Document the spike: cause, detection time, peak volume multiple, SL impact by hour, actions taken and their effectiveness.	WFM team, Operations Manager	Post-event review: was the spike foreseeable? What data source would have predicted it? Update the forecast model and the contingency plan.

Volume spike questions

What causes sudden volume spikes in a contact centre?

Six main causes: (1) technology or service failure (website/app/payment down — spike begins within minutes, magnitude 3×–15×); (2) communication sent without WFM notification (marketing email, billing statement — spike 1.5×–3×, predictable if notified); (3) media or social media event (1.2×–10× depending on reach); (4) billing or payment event (1.5×–4×); (5) regulatory announcement (1.3×–3× over 2–5 days); (6) failure demand amplification (1.2×–1.8× over weeks from unresolved previous contacts).

What is the emergency staffing response to a contact centre volume spike?

Three-tier response: Tier 1 (0–30min, zero cost): cancel breaks, cross-skill redeployment from lower-priority queues, release from meetings, activate callback, increase IVR deflection. Tier 2 (30min–2hrs): overtime call-out, shift extensions, agency/flex pool activation, BPO overflow, proactive digital communications. Tier 3 (sustained spikes): formal BCP activation, senior management notification, press/comms team briefing, triage routing for vulnerable customers.