Should we budget AI-search optimization as its own line item, separate from our existing SEO/content budget?
AI-search optimization earns its own budget line, and that line funds measurement and content rather than tools. Deploi scores AI answer engine visibility CUSTOMIZE: technical hygiene runs free to $69/month, while an owned prompt panel and first citable-content sprint run $10,000 to $30,000 (Deploi estimate, illustrative, September 2026). Folding that into SEO reporting hides both the cost and the result.
Why a separate line, when the work is mostly SEO
Google's own guidance is that the best practices for SEO remain relevant for AI features, and that no additional requirements or special optimizations are necessary (per Google Search Central, September 2026). That argues against a separate discipline. It does not argue against a separate line, because the two are measured differently and a shared line lets the easier metric win.
SEO reports sessions and rankings. AI-search work produces citations, mentions and agentic-channel orders, which Shopify reports per AI channel as sales, orders, online store sessions and conversion (per Shopify's Help Center, September 2026). Put those under one line and quarterly reporting will default to the session number, because it is bigger and older.
What belongs in the line
| Component | Cost signal | Source |
|---|---|---|
| Technical hygiene (schema, discovery files, crawler access) | Free to $69/month | Verified September 2026 |
| Owned prompt panel + first citable-content sprint | $10,000–$30,000 | Deploi estimate, illustrative |
| Theme JSON-LD layer, if not already built | $2,000–$6,000 | Deploi estimate, illustrative |
| Ongoing upkeep | ~15–20% of build cost per year | Deploi estimate |
Thresholds
- Under roughly $20M GMV, or a category nobody researches conversationally: no separate line yet. Fund SEO and product data, revisit next year.
- A category shoppers research by asking (skincare, supplements, gear, home): open the line. Our reference scenario for this decision is exactly that profile.
- Agentic channels already producing orders in the Shopify admin: the line is no longer optional, because you now have a revenue stream nobody owns.
When NOT to open one
- When the line would be spent on tools. A vendor's "AI SEO Score" measures the vendor's checklist; it rises when you complete the checklist, whether or not any engine changes its answer.
- When there is no content budget behind it. A panel with no content program attached is a dashboard of bad news.
- When you need a quarterly ROI number. One quarter is roughly the minimum before a trend means anything (Deploi estimate, illustrative). Committing to a 90-day ROI report guarantees a bad decision at day 91.
The Deploi point of view
Our own position, from building on Shopify. Separate from the facts above.
- Our take: Open the line, and spend it on measurement and writing, not software. The technical layer is nearly free; the expensive part is having something worth citing.
- What we’ve seen: Budgets that stay inside the SEO line get reallocated the first quarter traffic dips, which is precisely when citation work matters most.
- Where we disagree: The GEO tooling market prices AI visibility as a software subscription. We score it CUSTOMIZE and route the money to a prompt panel and content, because no app can promise that ChatGPT, Gemini or Perplexity will recommend you, and several sell as though they can.
- What this page adds: the actual cost components of the line, and the reporting failure that makes a merged line unworkable.
Reviewed by Martin Dejnicki, Director of SEO & AI Search. Facts verified 2026-09-13.