METHODOLOGY

How We Tune VICIdial Campaigns

Two weeks. Week one we measure. Week two we push. We do not guess, we do not copy a config from another client, and we do not tell you what you want to hear.

Week 1 -- Baseline

Before we touch a dial_level we sit on the existing config for five trading days and read the logs. The numbers we take are the ones the SEC and the FTC already asked for, plus the ones that actually move revenue:

• Connect rate by DID, by hour, by list source
• Answering-machine rate and AMD false-positive rate
• Drop rate (abandoned) vs. the 3% Safe Harbor ceiling
• Talk time distribution and agent pause codes
• Cost per qualified lead, split by campaign and list

If a baseline reading looks wrong we say so. If the existing config is already within 10% of what we could get, we tell you that too.

Week 2 -- Push

Week two is one controlled change at a time with the abandoned-call counter open in the other window. The usual order:

1. Fix obvious list hygiene issues -- bad area codes, wrong time zones, stale phones
2. Move AMD thresholds to the point where false positives stop costing us live answers
3. Lift dial ratio in 0.2 steps, holding the drop rate under the campaign group ceiling
4. Rotate DIDs off the worst-performing ASN and onto local-presence numbers that actually connect
5. Tune queue routing so the best closer gets the warmest transfer

Every change has a rollback. We note the before/after numbers in a shared doc. If a change made things worse, we revert and explain why.

Hand-off

At the end of the two weeks you get a written runbook -- every setting, every threshold, every alert -- plus dashboards for the five KPIs we tuned against. Your ops team can run it. We stay on retainer if you want a second pair of eyes, or we walk away.

Deeper technical write-ups on each lever (dial ratio math, AMD tuning, DID reputation) live on our sister site at predictive-dialer-roi.com.