Enough events
Google advises bidding on one funnel stage with a relatively short conversion delay and at least 15 conversions a month (Google Ads Help).
Industry playbook · Manufacturing
An RFQ is a question, not a sale. Teach Google and LinkedIn which questions your estimators can actually quote, and judge every campaign on cost per quote-ready RFQ rather than cost per enquiry.
Short answer
Bid on the words buyers use once they know what they need: process, material, specification, certification and “manufacturer in <city>”. Exclude jobs, student and DIY searches. Then tell Google which RFQs your team marked quote-ready, so bidding learns from genuine requirements rather than every form fill.
This playbook is for OEMs, contract manufacturers and fabricators who sell through RFQs. If you sell through dealers or stock other brands, read the distributors and dealers playbook. If your work is won through consultants, vendor lists and tenders, read the EPC playbook.
Short answer
Buyers are engineers, procurement teams and plant heads, and an RFQ starts a requirements review rather than a sale. Cycles run from weeks for contract jobs to a year or more for OEM qualification, so purchase orders are few. The ad platforms only see the RFQ unless you send the later stages back.
Three features shape every decision in this guide:
Short answer
They search by process, material, specification, certification and location, often with a drawing already in hand. Buyer searches name what they need made; noise searches name a job, a course, a PDF or a price list. Build campaigns around capability families, and exclude the noise before it fills the RFQ inbox.
| Query family | Such as | What it tells you |
|---|---|---|
| Process + material + spec | “investment casting SS316 valve body” | Requirement is defined; usually the best intent |
| Certification-qualified | “AS9100 CNC machining”, “ISO 13485 injection moulding” | Regulated buyer; the certification doubles as a filter |
| Manufacturer in a place | “hydraulic cylinder manufacturer Ahmedabad”, “steel fabrication company Sharjah” | Supplier search; a mix of end users and traders |
| Contract or OEM | “contract manufacturer for EV battery enclosures” | Longer cycle, larger value, qualification ahead |
| Drawing-led | “custom machined parts from drawing” | Build-to-print; ask for the drawing |
| Noise | jobs, internship, project report, PDF, DIY, used, price list, wholesale market | Add as negatives or separate them |
Wording changes by market. Indian buyers search for “manufacturer”, “exporter” and “supplier in” a city; UAE buyers use “supplier”, “trading” and “LLC”; US buyers use “RFQ”, “contract manufacturer” and “made in USA”. The negative keywords guide covers how to build and maintain exclusion lists.
Positioning
The same factory can sell three different things. Each changes the keywords, the form and what counts as a qualified RFQ, so give each its own campaign and page.
| Own-product OEM | Contract manufacturing | Build-to-print job work | |
|---|---|---|---|
| What the buyer evaluates | Product specification, standards, after-sales support | Capability, capacity, quality system, confidentiality | Tolerances, material, turnaround, price |
| Keywords | Product, model and application terms | “contract manufacturer for” plus product category | Process + material, “from drawing” |
| Form fields | Application, quantity, delivery location | Annual volume, NDA checkbox, target start date | Drawing upload, material, quantity, required-by date |
| Qualified means | A real application and quantity | Volume and timeline justify onboarding | In tolerance and material range |
| Cycle | Weeks to months | Months, with audits and trial orders | Days to weeks |
Be honest about capacity and lead times in ads. Overpromising produces RFQs you have to decline; estimators mark them “not a fit”, and that is exactly the signal you then send back.
Short answer
Google Search captures RFQ intent and should usually lead, with a campaign per capability. LinkedIn reaches named OEM accounts and specifiers when deal sizes justify its cost. Meta supports retargeting and trade-fair promotion. Marketplaces and directories stay or go on cost per quote-ready RFQ, measured with the same definition as your own campaigns.
Original framework
Five yes-or-no checks that sales or estimation completes within 48 hours of an RFQ. Three or more yeses make it quote-ready, and that status is what you send back to the ad platforms.
| Check | Evidence in the RFQ | Common “no” |
|---|---|---|
| In range | Material, process, tolerance and size inside your capability | “Titanium 5-axis” sent to a mild-steel fabricator |
| Real quantity | Annual volume or batch size at or above your minimum | “Need 1 piece, price?” |
| Reachable buyer | Company domain, designation, GSTIN, TRN or registered company | Free email and no company; a student project |
| Timeline | Required-by date or project stage | “Just collecting prices” |
| Commercial path | End user, OEM or approved contractor | Trader re-quoting; competitor benchmarking |
Short answer
A quote-ready RFQ asks for something inside your capability, at a realistic quantity, from a reachable business buyer with a timeline and a direct commercial path. If three of those five checks pass within 48 hours, count it as quote-ready and send that status back to the ad platforms.
Form design trades volume for quality. A drawing upload, quantity and required-by date raise quality and lower volume, so test them one at a time. GSTIN or TRN can be optional fields. Scoring detail lives in the lead scoring framework.
Original framework
Short answer
Usually the quote-ready RFQ, not the purchase order. Google recommends bidding to a stage with a relatively short conversion delay and at least 15 conversions a month, and a manufacturer with four orders a month can’t meet that. Orders also arrive late. Optimise to quote-ready RFQs, then check that they predict orders.
| Stage | Typical lag from click | Volume | Noise | Use as |
|---|---|---|---|---|
| RFQ submitted | Same day | High | High | Early learning only |
| Quote-ready RFQ | 1–2 days | Medium | Low | Primary bidding signal |
| Quote sent | Days to weeks, after estimation | Lower | Low | Secondary; value checks |
| Trial order or first article | Weeks to months | Sparse | Very low | Validation |
| Purchase order | Weeks to a year | Very sparse | Lowest | Validation only |
Google advises bidding on one funnel stage with a relatively short conversion delay and at least 15 conversions a month (Google Ads Help).
Enhanced conversions for leads imports uploads within 63 days of the click, and click-ID imports within 90 days (Google Ads Help). LinkedIn offers 180- and 365-day windows for lead and qualified-lead conversions (LinkedIn Help).
Value-based bidding needs two or more different values reported to your conversion goals (Google Ads Help). Value quote-ready RFQs by capability or deal-size band.
A deeper outcome is not automatically a better signal. The logic is set out in which lead stage to optimise for and the signal gap.
If both opportunities reach Google as the same conversion, bidding learns they are equally good and goes looking for more quote-comparers. Recording the difference in fields sales already fills in, such as “end-buyer access” or “budget approved”, lets each stage carry more information than the one before.
Don’t trust any weighting until later wins and losses across many opportunities confirm it.
Short answer
Capture the click ID and the buyer’s business email on the RFQ form, keep a sheet or CRM with each RFQ’s status, and upload quote-ready and quote-sent statuses every few days. Leave unreviewed RFQs out rather than marking them as failures. No CRM is needed to start; consistent statuses and dates matter more.
Ad platforms execute; the CRM or sheet is the system of record. Sending conversions back helps bidding, but somebody still decides what to change across keywords, pages and budget.
Worked numbers
Broad against technical terms in the US, marketplace against Search in India, and a single out-of-range RFQ. Run the same arithmetic on your own account.
| Broad industrial terms | Technical and part-number terms | |
|---|---|---|
| Spend | $15,000 | $9,000 |
| RFQs | 120 | 45 |
| Cost per RFQ | $125 | $200 |
| Quote-ready RFQs | 6 | 24 |
| Cost per quote-ready RFQ | $2,500 | $375 |
| Proposed budget | $9,000 | $15,000 |
The issue: broad terms look cheaper per RFQ, but only 6 of 120 are quote-ready. The action: move $6,000 to technical terms with total spend unchanged, and add negatives for out-of-range queries, within spend limits and after approval. After it runs: check that quotes and later orders follow before moving more.
Short answer
After the quote, OEM buyers often run trial orders, first article inspection and supplier audits before approving a vendor. That can take months, so revenue within 30 days says nothing about a campaign. Judge campaigns on quote-ready RFQs and quotes, publish the certifications you actually hold, and review orders by RFQ cohort each quarter.
Zephra creates, runs and improves Google, LinkedIn and Meta campaigns, new or existing, using lead quality and business outcomes. It takes RFQ statuses from CRM data or a spreadsheet, so no CRM is needed to start, and sends CRM lead stages to Google through the Data Manager API and to Meta and LinkedIn through their Conversions APIs. Recommendations are labelled act, review, investigate or wait. Low-risk changes can run automatically; everything else follows your permissions or waits for approval, within spend boundaries, cooldowns and learning-phase protection, and every change is recorded in an audit log. See Zephra for industrial and manufacturing and lead quality optimisation.
Questions, answered
Usually not. Most manufacturers win only a few orders a month, and Google recommends bidding to a stage with at least 15 conversions a month and a short delay. Orders also arrive after the upload windows close. Optimise to quote-ready RFQs, then check that campaigns winning more of them also win more orders.
It is worth it when you sell to identifiable companies and roles, such as OEM procurement, plant heads or design engineers, and deal sizes support the cost. LinkedIn needs at least 300 members per ad set and suggests 50,000 or more, so very niche targeting may not deliver. Judge it on cost per quote-ready RFQ, not cost per lead.
Neither by default. Run the same readiness check on both sources for three months and compare cost per quote-ready RFQ, not cost per enquiry. Marketplace enquiries are often shared with several sellers and skew toward price checks. Google leads cost more each, but you own them and can feed the outcome back to bidding.
Quote-ready RFQs show within weeks. Judge orders over at least one full quote cycle, and expect OEM vendor qualification to add months on top.
Enough for each bidding campaign to reach about 15 quote-ready RFQs a month. Multiply your cost per quote-ready RFQ by 15 to get the monthly floor per campaign. If that is out of reach, run fewer campaigns, or bid to RFQs and review quality by hand. See cost per qualified lead.
Offer an optional upload. Drawings raise quality and lower volume, so test the field and watch cost per quote-ready RFQ rather than RFQ count.
No. A sheet with the click ID or business email, the date, the status and a reason is enough to start sending quote-ready statuses back.
Run a campaign per country, use Presence location targeting where interest traffic adds weak RFQs, and put each country’s delivery terms in the ad and landing page.
Your account
The free ad account assessment takes 10–15 minutes, needs no CRM and makes no campaign changes. It reviews wasted spend, campaign structure, performance constraints and measurement gaps.