Five9 Consultancy,
Asked & Answered

Straight answers to the questions we hear most, on what we do, what it costs, and the Five9 questions our clients ask us every week: dialling modes, abandon caps, spam labels, lead waterfalls and more. If yours isn't here, ask us directly.

The Basics

What does a Five9 consultant do?

A Five9 consultant makes your Five9 platform perform: designing and tuning dial strategy, building and auditing your VCC, managing campaigns and IVRs, improving connect rates, integrating Five9 with your CRM, and turning the platform's data into reporting your leadership team can act on. At AL Performance we go a step further than advice, our consultants embed in your operation and do the work, from configuration changes to in-day real-time dialler management.

How much does Five9 consultancy cost?

It depends on the depth of support: strategic advisory costs less than a fully managed dialler operation. Our three packages, Lite, Core and Managed, step up in intensity, and every package includes blended hours that flex between consultation, configuration and real-time monitoring based on your business needs. The honest answer on price is a scoping conversation, we would rather quote your actual requirement than publish a number that fits nobody.

What packages does AL Performance offer?

Three: Lite, Core and Managed. Lite is strategic advisory, weekly reviews, custom reporting, training and campaign planning for teams with internal muscle. Core is hands-on partnership: daily performance monitoring, campaign builds, VCC configuration and audits, with our reporting pack included. Managed is hypercare: hourly monitoring, fully managed dialler operations, a dedicated support resource and our real-time tooling. All packages include blended hours that flex with your needs.

How quickly can an engagement start?

Most engagements kick off within 1-2 weeks of the initial conversation, and for urgent requirements, a departed dialler manager, a struggling campaign, a looming launch, we can often start sooner.

Who actually works on our account?

A dedicated pod, not a generic support queue: an account lead who owns your relationship and roadmap, a tech specialist for configuration, IVR builds and troubleshooting, a data and MI analyst for reporting and insights, and a real-time analyst for in-day dialler management. They know your agents by name and your campaigns inside out.

Do you offer one-off projects as well as ongoing support?

Yes. VCC builds, audits, healthchecks, integrations and migrations onto Five9 all work as defined projects, and our Lite, Core and Managed packages cover ongoing partnership. Many clients start with a project, an audit is a common first step, and continue into a package once they have seen how we work.

Partnership, Certification & Coverage

What is a Five9 Certified Implementation Partner?

A Certified Implementation Partner (CIP) is a partner Five9 has certified to deliver implementations of its platform. AL Performance is both a Five9 Global Partner and a Certified Implementation Partner, so our VCC builds, audits and migrations are delivered to the standard Five9 itself certifies.

How is AL Performance different from Five9 Professional Services?

We are complementary, not competing. Five9 Professional Services delivers the platform; we add the operator's layer on top: dial strategy, campaign structure, connect rate work, reporting and the day-to-day running that turns a correctly installed platform into a high-performing operation. Many of our clients work with both, and as a Five9 partner we are comfortable in that arrangement.

What does the free Five9 Health Check cover?

It is a free 12-question self-assessment covering the areas we audit most often: campaign structure, dialler modes, lead management, disposition rules and reporting. It takes about three minutes and gives you an instant assessment with actionable recommendations. For a deeper look, our full VCC audits and healthchecks go through your actual configuration line by line.

Do you work with contact centre platforms other than Five9?

Five9 is our specialism, and we believe depth on one platform beats shallow coverage of ten, it is why we are a Five9 Certified Implementation Partner. Around Five9 we work across the wider stack: Salesforce, Microsoft Dynamics, Zendesk, HubSpot, Go High Level, Power BI and omnichannel channels. If you are on another dialler and considering a move, we help businesses migrate onto Five9; if you are staying on another platform, we are probably not the right fit, and we will tell you so.

Do you cover both the UK and US?

Yes, we work with clients across both the UK and US, and our service is built around your operating hours, not ours. Real-time monitoring and in-day management follow your dialling day, and our engagement rhythm of daily reports and weekly calls runs on your timezone.

Which industries do you specialise in?

Our deepest verticals are mortgage and lending and home improvement, both high-volume outbound sectors where dial strategy and speed to lead decide the economics. We also work extensively with legal services, recruitment, BPOs and debt and collections, and across insurance, healthcare, education, energy, retail and technology, anywhere the contact centre is the engine of growth.

The Day-to-Day

Can you act as our outsourced dialler manager?

Yes, this is one of the most common reasons clients come to us, often after a dialler manager resigns. We take on the full role: in-day campaign management, list loading and pacing, configuration changes, monitoring and reporting, delivered by a dedicated pod rather than a single person, so cover never depends on one individual.

What reporting do you provide?

Daily end-of-day performance reports, weekly technical calls and performance reviews, monthly strategic reviews and quarterly business alignment, plus Power BI dashboards tailored to your operation. On top of that: agent productivity analysis, leaderboards, 1-2-1 data packs for team leaders and schedule adherence tracking with proactive alerting.

Which CRMs can you integrate with Five9?

Salesforce (API, LSP, connectors and data sync), Microsoft Dynamics, Zendesk, HubSpot and Go High Level as standard, plus workflow automation and custom API development where there is no off-the-shelf route. We also deliver omnichannel setup across WhatsApp, SMS, chat and email.

What results do your clients typically see?

Clients have seen contact rates improve by up to 40%, abandonment fall by up to 35%, conversion improve by up to 25% and handle time drop by up to 45 seconds. Every operation starts from a different baseline, so we measure yours first and report progress against it, we would rather under-promise than inflate a headline.

Modes, Pacing & Abandons

What's the difference between predictive, power, progressive and preview dialling, and when should I use each?

Predictive dials ahead of agent availability using maths on answer rates and talk times: highest throughput, but it will abandon some calls. Power dials a fixed number of lines per ready agent: simpler, still carries abandonment risk above one line each. Progressive dials one call per ready agent, so nobody answers without an agent waiting. Preview shows the agent the record before dialling: right for complex, high-value conversations. The rule of thumb: the more each conversation is worth, the gentler the mode.

Why shouldn't I put fresh web leads on a predictive campaign?

Because predictive, by design, sometimes dials people it cannot connect to an agent, and a fresh lead who answers to dead air is a lead you paid for and then burned. Speed-to-lead matters, but so does the first impression. We run hot leads on progressive or preview, where every answer meets an agent, and keep predictive for the colder, deeper list where a small abandon rate is an acceptable trade.

What causes abandon rate spikes on my campaigns?

The usual suspects: pacing set too aggressively for the list, a sudden jump in answer rate from a fresh list drop or a better time of day, talk times running longer than the algorithm expected, or agents logging off without pacing catching up. Remember abandon rate is measured against live answered calls, not dials, so a cleaner list can spike the percentage even when abandon counts look flat. In the US you have 3% per campaign per 30 days to protect.

How does pacing tuning actually work?

Pacing balances two costs: agents idle waiting for a call, and answered calls with no agent free. We tune aggressiveness, meaning lines per ready agent or the target abandon threshold on predictive, against real list behaviour: answer rate, average handle time, agent count and time of day. Then we watch and adjust, because a list behaves differently across the day and across its own lifetime. Good pacing is reviewed at least weekly, not set once at launch.

Should I use answering machine detection, or does it cost me more than it saves?

AMD is a trade. It saves agents from listening to voicemails, which is real capacity back. But no detection is perfect: some live answers get classified as machines, and under US rules a live answer not passed to an agent within two seconds of the person's completed greeting counts as abandoned. On fresh or high-value lists we often soften or disable AMD, accepting some voicemail time to protect real answers. On deep cold lists, the agent time saved usually wins.

Why are my agents sitting idle when the list still has thousands of records?

Usually the dialler has run out of records it is allowed to dial right now, not records full stop. Common causes: retry timers holding back recently attempted numbers, dialling-hours and timezone rules blocking big chunks of the list, filters excluding more than intended, campaign priorities starving one list, or the fresh numbers genuinely exhausted with everything sitting in redial. We start by checking what share of the list is actually dialable at that moment.

My connect rates are fine in the morning then collapse mid-day. What should I check?

First, list position: if the campaign burns through its freshest records early, by midday you are into retries and weaker numbers. Second, timezone mix: you may be dialling regions in their dead hours. Third, caller ID reputation: carriers can start flagging numbers as spam as volume climbs. Fourth, pacing and staffing dips around lunch. Work through those four in order and the culprit usually shows itself.

Caps, Consent & Calling Hours

What does the 3% abandon rate cap actually mean, and how is it measured?

Abandonment is measured as abandoned calls as a share of live answered calls, not total dials, and the 3% cap applies per campaign over a 30-day period. A call counts as abandoned when a live person answers and no agent is connected within two seconds of their completed greeting. Because the denominator is live answers, a campaign with a poor contact rate can breach the cap on surprisingly few abandons, which is why we tune pacing per campaign, not account-wide.

How does TCPA affect how my dialler should be configured?

Consent decides dialling mode. Records with prior express written consent can typically run on predictive or progressive campaigns, while numbers without it, mobiles especially, generally need a manual or human-initiated approach. Operationally that means segmenting lists by consent status, routing each segment to the right campaign type, capturing consent evidence in your CRM, and honouring revocations fast. Most TCPA trouble we see starts with mixed lists running through a single predictive campaign.

We dial in the UK as well, what does Ofcom expect from us?

Ofcom takes a stricter line than the US. There is no numeric safe harbour any more: the old 3% threshold was scrapped, and any pattern of abandoned or silent calls can count as persistent misuse. In practice that means pacing conservatively, presenting a valid CLI people can return, playing an information message on any abandoned call, and not redialling a number straight after an abandon. For UK campaigns we design to keep abandons near zero, not up to a cap.

How do we stay inside calling-hours rules across US time zones?

The federal window is 8am to 9pm in the called party's local time, and several states are tighter, with earlier cut-offs or weekend restrictions. The operational trap is knowing what local time actually is: area codes no longer reliably indicate location because mobiles move with their owners. So we key restrictions off address or state data where it exists, apply the most conservative rule when signals conflict, and let the dialler enforce windows automatically rather than trusting list pulls.

What does proper DNC scrubbing and suppression look like?

Treat it as layers. The national DNC registry, state registries where applicable, your internal do-not-call list, and any litigator suppression files all need scrubbing before a list ever loads. Then enforce suppression inside the dialler as the final gate, so a bad list pull cannot leak scrubbed numbers back into rotation. Internal opt-outs should apply immediately across every campaign, not just the one that generated the request. And scrub on a schedule, not once at import.

What does a compliant abandoned-call message need to say?

To qualify for safe harbour it must start within two seconds of the person's completed greeting, identify the company the call was made for, give a toll-free number they can call to make a do-not-call request, and contain no sales pitch. We configure the message, the timing and the DNC handling behind it, but we are consultants, not lawyers: we make your dialler enforce the rules operationally, and your counsel should sign off wording and your overall compliance position.

Caller ID & Spam Labels

Why do my outbound numbers keep getting flagged as Spam Likely?

Because the analytics engines behind the US carriers score numbers on behaviour, and each engine scores independently, which is why a number can be clean on one network and flagged on another. High daily volume from a single number, short average call durations, low answer rates and consumer complaint reports all push the score the wrong way. Your dial strategy drives every one of those signals, so lasting fixes are behavioural: the label is a symptom, not the disease.

What does sensible number rotation actually look like?

A stable pool sized to your volume, not a conveyor belt of fresh numbers. Brand-new numbers dialling at high volume are themselves a spam signal, so constant replacement makes things worse. We size the pool so no single number carries suspicious daily volume, register every number with the analytics engines, rest flagged numbers rather than bin them, and rotate on data (a dip in answer rate on one caller ID), not on a fixed calendar.

Is branded calling worth the money?

Branded calling puts your company name, and in some cases a logo and call reason, on the recipient's screen instead of a bare number. For businesses consumers actually recognise, it can lift answer rates, and it protects your identity against spoofing. Two caveats: it is a per-call cost that needs to earn its keep, and it does not override a spam label. Fix reputation first, then brand the call; branding a flagged number is money down the drain.

What is STIR/SHAKEN in plain English?

It is caller ID authentication. When you place a call, the originating carrier digitally signs it to vouch that you are entitled to present that number. Full attestation (A-level) means the carrier knows who you are and that the number is yours; B and C mean progressively weaker vouching, which invites more scrutiny. The important limit: STIR/SHAKEN tackles spoofing, not labelling. A fully attested call can still be marked Spam Likely on behaviour, so treat A-level attestation as the entry ticket, not the fix.

Should we be using local presence dialling?

Sometimes. Matching your caller ID to the prospect's area code can lift answer rates, particularly on first attempts. The costs: a much larger number pool to buy, register and monitor, growing consumer wariness of unfamiliar local numbers, and a duty to handle callbacks properly on every number you present. Our advice is to test it honestly against a consistent, branded number and measure conversion, not just answers. If you cannot staff the callback path, do not present the number.

One of our numbers is burned. Can it be rescued?

Often, yes. Dispute the label with each analytics engine (registration services make this one process rather than several), then rest the number at little or no outbound volume while the score decays. Crucially, fix the dialling behaviour that burned it, otherwise the replacement burns too, just slower. If remediation fails, retire the number from outbound but keep it live for inbound, because people still return calls to it. Releasing it back to the carrier just passes the problem along.

How do we monitor number health on an ongoing basis?

Two layers. Externally, use a reputation monitoring service that queries how each carrier's analytics engine currently labels every number in your pool, because labels change without warning. Internally, watch answer rate by caller ID in your Five9 reporting: a sudden dip on one number is usually the earliest sign of a flag. Make both a weekly habit. If contact rates have slipped and you are not sure why, our free Five9 Health Check is a sensible place to start.

Lists, Waterfalls & Speed to Lead

How should we prioritise fresh leads over aged ones?

Build a priority waterfall: fresh leads sit in a high-priority campaign that always dials first, and aged records backfill agent capacity when the fresh well runs dry. The most common mistake we see is a flat structure where a 30-day-old record competes with a lead that landed 30 seconds ago. Priority should mirror value: a fresh lead is worth many times an aged one, and your dialling order should say so.

Why do some of our leads never get called back?

Every disposition should tell the dialler exactly what to do next: how long to wait, how many more attempts, when to stop. System defaults are generic, so a no answer and a busy signal often get identical treatment, and dispositions nobody mapped simply stop dialling altogether. That is how leads leak: records park silently with no follow-up ever scheduled. We audit every disposition path on every campaign so nothing falls out of the funnel unnoticed.

What does list penetration actually tell me?

Penetration is the share of a list you have actually attempted, and how many times. Low penetration with strong results means you are leaving contactable people untouched. High penetration with fading results means the list is exhausted and you are burning agent hours on it. Read it alongside contact rate by attempt number: together they tell you when to push harder, when to rest records, and when to ask for fresh data.

Why does speed to lead matter so much for contact rate?

Because intent decays fast. Someone who just filled in a form is at their phone, thinking about the problem and expecting a call. Minutes later they are back at work, screening unknown numbers. Dialling within the first few minutes routinely multiplies your chance of a conversation versus dialling hours later, which is why we treat speed to lead as the highest-leverage setting on most consumer campaigns.

Should fresh and aged leads run on separate campaigns?

Yes, almost always. Fresh leads justify aggressive treatment: immediate dialling, tight redial windows, your best agents. Aged records need patience: attempts spread across different days and time blocks, varied calling windows, and acceptance of a lower hit rate. Blend them on one strategy and you either over-dial the aged records or under-serve the fresh ones, and your reporting cannot tell you which list is actually paying. Separate campaigns keep both the treatment and the economics honest.

We have dozens of active campaigns. Is that a problem?

Usually, yes. Campaign sprawl is one of the most common findings in our free Five9 Health Check. Every extra campaign splits your agents, lists and reporting, and each carries its own abandon rate (abandoned calls as a share of live answered calls) to keep under regulatory caps. Dozens of near-identical campaigns usually means old tests and one-off requests nobody retired. We consolidate to a handful of genuinely different treatments, then let priority and list filters do the rest.

Every one of our teams defines success differently. What should we standardise?

Do not force one universal definition of success onto every floor: a mortgage transfer and a home improvement appointment set are genuinely different outcomes. What you standardise is the disposition framework underneath: a consistent, mutually exclusive set of dispositions mapped to the same funnel stages (contact, conversation, outcome) so every campaign reports on the same skeleton. Each floor then defines its own conversion on top of it. Comparable data, local flavour.

Reporting, Admin & Integrations

Which numbers should my outbound floor watch during the day, and which can wait until end of day?

In-day numbers are for steering: contact rate, abandon rate (abandoned calls as a share of live answered calls, against the 3% US cap), agent occupancy and availability, pacing, and speed-to-lead on fresh records. End-of-day numbers are for judging: conversion, cost per sale, list penetration and disposition mix. The classic mistake is mixing them: floors staring at conversion at 10am when the number that actually needs action is pacing. Watch the levers in-day, judge the outcomes at close.

What should actually be on our real-time dashboard?

Only numbers someone will act on within the hour: agents ready versus on calls, calls in queue, live contact rate, abandon rate tracking against the 3% cap, campaign pacing, and whether fresh leads are being dialled quickly. A good wallboard answers one question at a glance: does anything need intervention right now? If a metric would not change what a supervisor does in the next thirty minutes, it belongs on the end-of-day report, not the wall.

What are the most common Five9 configuration mistakes you find in audits?

The same handful, over and over: dialling ratios set at launch and never revisited, list priorities that quietly bury fresh leads under stale data, redial rules hammering the same numbers, dispositions mapped so loosely that reporting becomes fiction, abandon settings drifting towards the compliance line, and IVRs built for a team structure that no longer exists. None are exotic; they accumulate because nobody owns the configuration. Our free Five9 Health Check exists precisely because these are cheap to find and expensive to ignore.

How do we keep control of admin changes in Five9? Things get changed and we only find out when something breaks.

Treat your VCC like production software. Keep admin rights to as few people as possible, record every change (what, who, why and when) in a simple shared log, and review it weekly. Five9 keeps an audit trail, but by the time you are reading it, something has usually already bitten you. The habit that saves floors is a one-line note before the change and a check on the metric it should move afterwards. Most 'the dialler broke overnight' mysteries turn out to be an undocumented Tuesday tweak.

Can you integrate Five9 with Salesforce?

Yes, and done well it is one of the highest-value parts of the platform: screen pops so agents see the record before they say hello, click-to-dial straight from the CRM, and dispositions syncing back so both systems agree on what happened. The problems we find are rarely the connector itself: mismatched fields, duplicate records, and syncs that fail silently are the usual culprits. We design, build and audit these integrations for Salesforce and similar CRMs, and we always test the failure paths, not just the happy ones.

What is good practice for call recordings?

Three disciplines. Retrieval: recordings should be findable in seconds by agent, campaign, number or disposition, not fished out of a weekly export. Retention: set a policy that matches your regulatory obligations (mortgage and insurance clients often need years rather than months) and know exactly where recordings live if you archive off-platform. Coaching: sample deliberately, calibrate your scorers, and review the calls that convert as well as the ones that fail. A recording nobody ever listens to is just storage cost.

Agents are complaining about call quality. What should we check first?

Start by scoping it: one agent, one team, one site, or everyone? One agent is nearly always local: headset, browser, VPN or home wifi. A whole team points at the network path: bandwidth, firewall rules or QoS on voice traffic. Everyone at once points at the carrier or the platform. While you check, capture timestamps and call IDs, because 'the audio is bad' is undiagnosable, but ten timestamped examples are evidence Five9 support can actually act on.

Still weighing it up? Get a free health check

Two minutes, no obligation, and you keep the scorecard either way.

Get Your Free Health Check

Still Have a Question?

Ask us directly, you'll get an operator's answer, not a sales script. And if you want a quick read on your own setup first, ask about our free Five9 Health Check: 12 questions, about three minutes, instant assessment.

Get in Touch