Done-for-you programmatic SEO, built for data scientists
A fixed-scope programmatic SEO build for data scientists: we design the data contract against Python / pandas, ship 5,400 reviewed URLs in the first tranche, and hand you the templates, gates and runbook. Modelled at 18,600 indexed URLs earning roughly 744 conversions a month at data scientists' typical rates.
- Pilot
- from $3,900
- Build
- from $10,200
- Scale
- from $21,100
1,890–5,400 URLs
30,000 URLs
120,000+ URLs
The build, not the theory
The interesting work in programmatic publishing is upstream. Which fields have high enough coverage to render, how do you resolve the same entity across three sources, what happens when a supplier changes a unit, and how do you keep output reproducible when the input drifts. We build the publishing layer to be driven by a validated dataset with a versioned contract, so page generation is deterministic given the data. That is the strategy. This page is the service: what we actually do, in what order, at what price, and the conditions under which we tell you not to buy it. Every engagement starts by pointing at one real system you already run — usually Python / pandas, sometimes dbt — and asking whether its rows can answer a query shape better than a human writing prose. If the answer is no, the engagement stops at the audit and you keep the findings. If the answer is yes, we write the data contract, design one template family against real rows, publish a gated first tranche, and only widen the axis after Search Console proves the pattern indexes. Nothing about this is a retainer: the engine installs on your WordPress, and the conversion path is wired into your funnel, not ours.
- Python / pandas
- dbt
- Great Expectations
- Airflow / Dagster
- BigQuery / Snowflake
- spaCy
- MLflow
Data-heavy catalogue and directory builds with validated pipelines and entity resolution.
Four phases, each with something shipped at the end
Contract definition
Required fields, types, units, allowed ranges and fill-rate floors, versioned in the repository alongside the transformation code.
Versioned data contract with tests.
- A written data contract mapping Python / pandas fields to every rendered block, with null-handling rules per field.
- One production template family with 12 blocks, of which 5 vary per row beyond string substitution.
- 5,400 published URLs with schema, internal links, sitemap entries and IndexNow pings.
- An index-eligibility gate that refuses to publish rows below your evidence threshold.
- Search Console cohort reporting split by template family, not by page.
- A refresh job that rewrites stale rows on a 42-day cycle.
- A data scientists-specific runbook: how to add an axis, how to kill one, how to roll back a bad batch.
- Handover session with dbt and Great Expectations wired into the reporting loop.
- Python / pandas export or API accessRequired
The rows are the product. Without them there is no page family, only prose.
- One named approverRequired
A single person who can sign off a template without a committee round.
- WordPress admin or stagingRequired
We generate into your environment; we do not host your pages.
- A working conversion pathPreferred
Each page hands the visitor to a real conversion flow — form, checkout, booking or trial.
- Search Console property accessPreferred
Cohort-level indexation data is how we decide what widens and what gets cut.
- Review capacity: one page per family, per batchPreferred
A human signs the sample before the batch publishes. That is the scaled-content defence.
Quote and payback estimator
Defaults are this seat's modelled economics. Move them to your own numbers — the build estimate and payback window recalculate live. It is a model, not a quote; the fixed proposal comes after the scoping call.
- Indexed URLs (62% gate)
- 7,440
- Clicks / month
- 29,760
- conversions / month
- 298
- Pipeline value / month
- $62,580
- Indicative build
- $15,000
- Payback
- <1 month
The 62% index rate is the conservative figure we plan against, not a promise. Batches that gate harder index higher.
Three ways to buy this build
Pilot
from $3,900
1,890–5,400 URLs
Proving the the data axis exists before you fund a build.
- Data audit + contract
- One template family
- Gated first tranche
- Indexation read at day 30
Build
Most boughtfrom $10,200
30,000 URLs
Data Scientists with a proven dataset and a live conversion funnel.
- Everything in Pilot
- 2 template families
- Schema + internal link graph
- Refresh job + rollback
- Runbook handover
Scale
from $21,100
120,000+ URLs
Multi-axis surfaces where one bad batch would be expensive to unwind.
- Everything in Build
- Multi-axis expansion plan
- Release drip + quota control
- Quarterly axis pruning
- Named delivery lead
The shape of the surface before a single page is written
/data-scientists/Hub. Human-written, links every axis, never generated.
/data-scientists/{the}/Primary axis — one page per Python / pandas row that clears the gate.
/data-scientists/{the}/{modifier}/Second axis. Only published where the pair has its own evidence.
/data-scientists/compare/{a}-vs-{b}/Comparison family. Capped at pairs with real query volume.
/data-scientists/updates/Change log for the dataset — freshness signal, not filler.
Rules that decide whether a row earns a URL
| Gate | Test | If it fails |
|---|---|---|
| Row completeness | ≥ 65% of contract fields present | Below threshold the row renders as a directory entry, not a URL. |
| Evidence density | ≥ 4 unique data points not shared with a sibling row | Fails → folded into the parent page as a section. |
| Demand | A measurable query shape in Search Console, Semrush or paid data | No demand → the axis is dropped before templates are written. |
| conversion path | A working conversion destination for the row | Missing → the page publishes noindex until the path exists. |
| Human sample | One page per family per batch reviewed and signed | Unsigned batches do not leave staging. |
What the first weeks look like
- Week 1
Scoping call and data review
We look at what data scientists already hold — systems, exports, APIs — and score each axis for demand and defensibility.
- Weeks 3–4
Contract and template design
The data contract is written and the first template is designed against real rows, not placeholders.
- Weeks 5–8
First tranche live
4,500–10,500 URLs published with schema, internal links, sitemap entries and IndexNow.
- Weeks 9–14
Read, widen, hand over
Indexation and impression data decides what widens and what gets cut. Templates, gates and runbook transfer to you.
A supplier unit change caught before publishing
- Setup
- A catalogue pipeline where one supplier silently switched from millimetres to centimetres.
- Mechanism
- A distribution check on the dimension field flagged a 10× shift against the trailing baseline and halted the DAG before generation.
- Result
- Zero incorrect pages published; the previous quarter, an equivalent change had reached 6,000 live URLs before a customer reported it.
Cases where we will not take the build
- Your Python / pandas data has fewer than 320 usable rows — a hand-written hub will outperform a thin page family.
- The query shape is dominated by one brand's own properties and no third party ranks on page one.
- There is no conversion destination, so the pages would earn traffic you cannot bank.
- You want volume as the goal. We size to evidence, and evidence usually caps the count lower than you hoped.
Buy the whole build, or one piece of it
Training data pipelines
Using LLMs in a publishing pipeline where they help, and refusing them where they hallucinate.
Scope this workstreamAnalytics integration
Closing the loop: performance data back into the warehouse so the next generation run is informed by the last.
Scope this workstreamBuying questions data scientists ask us
Pilot engagements start at $3,900 and cover the audit, contract, one template family and a gated first tranche. A full build for the 30,000-URL surface starts at $10,200. Pricing is fixed at proposal — we do not bill by page or by hour.
Other seats we scope builds for
Want the data scientists surface scoped before you build it?
We'll audit the data source, size the first batch, set the performance budget and tell you honestly if programmatic is the wrong tool for your category.
