Half my Meta lead phone numbers are disconnected. Are they fake leads?
Rarely fake on purpose. Meta prefills the phone number from the person's profile, and profiles hold old numbers people set up years ago and never updated.…
Straight answers about connecting AI to the systems a business actually runs on — from the people who build these integrations. New answers publish every day.
Rarely fake on purpose. Meta prefills the phone number from the person's profile, and profiles hold old numbers people set up years ago and never updated.…
Brand search sits at the end of a journey other things created, so last-click reporting will always flatter it. The honest test is a restriction test: red…
You need three linked records for every action: the input it saw, the output it produced, and the effect it had on another system, each timestamped and st…
Do not listen to random calls. Random sampling spends your attention on the calls that went fine. Build a stratified sample from the calls most likely to …
Pull a month of jobs, filter to the ones that started with a tracked phone call or a web form, and compare the recorded source against what the tracking s…
Start by deciding what the system must be the source of truth for. If the answer is jobs, dispatch and invoices, you need field service software, not a CR…
Stop comparing feature lists and start testing your worst job type. Run your most complicated real job through each demo — the one with a change order, tw…
ServiceTitan exposes a REST API, but access is not on by default. You request it, register an integration application, scope it to the endpoints you need,…
Measure after-hours demand from your own telephony records, not from the service's report. Then match every message the service says it took against recor…
Check six things before committing: whether a documented API exists, whether your specific plan includes it, the rate limits, whether historical data can …
Derive the threshold from the metric's own normal variation instead of a round number. Look at several weeks of history, find the typical swing, and alert…
Pull the query in Search Console and watch which URL ranks for it over time. If the ranking URL keeps flipping between two pages, or both sit in the teens…
Follow the same jobs, not the same weeks. Pull every job booked in a period and check where each one ended up: completed, still scheduled, rescheduled, ca…
Ask what happens without the model. If removing it leaves nothing, it is a wrapper. Real AI systems carry substantial engineering around the model: data p…
Carry an identifier from the click all the way to the job record, then report booked jobs and completed revenue by campaign instead of conversions by camp…
Retrieval systems pull passages, not pages. They favor text that answers a question directly, states facts plainly, and can be lifted without surrounding …
Separate the audit record from the payload. Log who or what triggered the run, a reference to the input rather than the input itself, the model and prompt…
Booking rate is booked calls divided by bookable calls over the same period. Using total calls as the denominator understates performance and makes the nu…
Divide fully loaded marketing spend by the number of genuinely new customers acquired in the same period — not leads, not jobs, and not repeat callers. Tw…
You cannot compare them on the same metric, because they capture demand at different stages. Compare each against its own affordable ceiling for the job t…
Move budget when a channel's marginal cost per booked job crosses the ceiling for the work it produces — not when its average looks worse than another cha…
You monitor outputs, not uptime. Four checks catch most failures: freshness (did anything arrive in the expected window), volume (is today's count inside …
Do not import the call log. Write the source onto the record at creation — when a call becomes a customer or a request, stamp the campaign, keyword or cha…
Separate the backfill from the live sync and give each its own budget. Walk history in fixed time slices, checkpoint after every slice so a failure resume…
The Answer Hub you're reading is the same content intelligence system Bluefrog builds for clients. It discovers the questions customers actually ask, publishes structured answers, clusters them by topic, links them to the services they relate to, and adds new answers on a daily schedule instead of dumping a thousand pages at once.
That last detail matters more than it sounds. Search engines treat mass-published content as a quality signal in its own right. A hub that grows steadily, answers real questions and links coherently to the rest of the site behaves like a publication — which is the only version of this that holds up over time.
We run the same system on client sites in home services and the trades. See how the Answer Hub works, the automation behind it, or why it's built for AI search as much as for Google.
Bluefrog makes AI useful by integrating it with the way your business actually works — your software, your calls, your customers, your marketing and your revenue.
Technology development since 1997 · AI integration platforms since 2001
A real conversation with the people who build the technology — no obligation, no pressure.