How We Built 180,000 Search Impressions and 7,350 AI Citations in 5 Months Using Programmatic SEO
A step-by-step account of how Tempus Central went from near-zero organic visibility to 180K+ search impressions, 7,350 Bing AI citations, and 15+ qualified leads using a structured programmatic SEO programme.
Most programmatic SEO content describes the theory. This one describes what actually happened, month by month, with a real client, real numbers, and the specific decisions that moved the results.
The client: Tempus Central, a workforce management software platform with 27 years of history in HR technology, 500+ enterprise customers across India, and, until December 2025, virtually zero organic search presence.
By April 2026, five months into the engagement: 180,000+ search impressions, 585 monthly organic clicks from programmatic pages, 7,350 Bing AI citations, and 15+ qualified leads, including two that converted to product demos within the first eight weeks.
Here is exactly how it was built.
Why Tempus Central Was the Right Candidate
Not every company should run programmatic SEO. The businesses where it works share a specific characteristic: a large surface area of buyer intent mapped to a modular product.
Tempus Central serves manufacturing companies, food processing plants, facilities management firms, and large enterprises across five core modules: attendance tracking, payroll, canteen management, contractor management, and compliance. Each module serves a buyer who searches differently. “Canteen management software for manufacturing” is a different query from “contractor payroll compliance India,” but both are high-intent, commercially specific, and underserved by existing content.
That is the programmatic SEO opportunity. A product with multiple modules, each of which maps to a distinct cluster of high-intent buyer queries, competing in a market where the existing content is thin or generic.
Before we started: approximately 234 monthly search clicks, average position 24, zero programmatic content pages live.
Phase 1: Foundation (December 2025 / January 2026)
The first decision was not what to publish. It was how to structure what would be published.
We ran ICP-aligned keyword research across all five product modules. The goal was not the highest-volume keywords. It was the keywords with the tightest match between search intent and actual product capability. Pages that genuinely answer the query a buyer typed are the pages that index, rank, and convert. Pages that approximate the intent produce impressions without clicks and clicks without conversions.
From that research we built a content framework: a mapping of each query cluster to the relevant product module, the correct page format for that intent, the URL structure, and the heading architecture that would allow internal linking to function as designed.
Then we published 120 commercial pages across the first two months, covering the primary keyword clusters for each module. Not 120 variations of the same template. 120 pages that each answered a distinct buyer question with a specific section of the product offering.
Technical foundations were set simultaneously: standardised URL structure, indexing configuration, and internal link architecture from new pages back to core product and category pages. This work happened before any content went live, and it is the reason pages indexed within 28 days of publication rather than waiting months for crawl equity.
Phase 2: Scale and Momentum (February 2026)
February produced the first clear evidence the programme was working.
We published another 60 pages, bringing the total to 180 live. Month-on-month search clicks grew 88 percent. Impressions grew 77 percent.
Two things happened in February that arrived earlier than expected.
First, two qualified leads came in directly from the programmatic pages: one from an attendance management page, one from canteen management. This validated the intent mapping at the most important level. The pages were not generating impressions from vaguely related queries. They were pulling buyers who were actively evaluating solutions.
Second, Bing AI citation tracking showed Tempus Central pages appearing in Bing’s AI-generated search answers. In January the total was 373 citations. By the end of February: 1,490. A 299 percent increase in a single month.
Bing AI citations matter because they are a leading indicator of topical authority, and they compound. Pages cited in AI answers receive click-through traffic. That traffic generates engagement signals. Those signals reinforce ranking. It is a reinforcing loop that accelerates the later phases of a p-SEO programme rather than running independently of them.
Phase 3: Optimisation and Cluster Development (March / April 2026)
The third phase shifted from publishing new pages to extracting more performance from existing ones.
We updated 180 pages from Phases 1 and 2 with improved formatting, tighter keyword alignment, and revised heading structure based on what the early ranking data showed. We built topic cluster architecture linking pillar pages to cluster content in both directions. We identified pages generating high impression volume but low click-through rates, a signal the page was appearing at too low a position or with a meta description that was not compelling enough to click, and revised those specifically.
In March, Search Console clicks dropped 22 percent from February’s peak while impressions grew another 48 percent. That pattern is normal and expected in large programmatic SEO rollouts. When you publish a large number of pages in a short window, the indexed page count grows faster than the ranking equilibrium settles. Pages appear in search at positions too low to generate clicks. As engagement data accumulates and internal linking distributes authority over time, positions improve and click rates recover.
The metric to watch during this phase is not Search Console clicks. It is organic sessions in Google Analytics, which captures all organic traffic including sessions from queries not individually attributable in GSC. GA organic traffic continued to grow through March and April, confirming the underlying trajectory even while the GSC click figure dipped.
AI citations continued growing through this phase: 2,497 in March, 2,990 in April.
The Full Results
| Metric | Starting Point | End of Engagement |
|---|---|---|
| Search Impressions | ~0 | 180,000+ |
| Organic Clicks (Search Console) | 234 / month | 585 / month |
| Bing AI Citations | 0 | 7,350 cumulative |
| Organic Traffic (Google Analytics) | 1,212 / month | 1,481 / month |
| Qualified Leads | 0 | 15+ direct from p-SEO |
| Pages Live | 0 | 300 |
The strongest single page, /contract-canteen-system/, reached 53,187 impressions and a best position of 3.89 across the engagement. A category most companies in this space would not have thought to target at all.
Three Things This Confirmed About Programmatic SEO
The first pages in an underserved cluster move fastest. Just as the first 10 to 20 referring domains you earn move domain authority the fastest, the first cluster of pages you publish in an underserved topic area claim the available ranking equity before competitors arrive. This argues strongly for targeting the highest-intent, lowest-competition clusters first rather than starting with the highest-volume terms.
Technical foundation determines indexing speed. Most pages reached impressions within 28 days of publication. That speed came directly from the URL structure, indexing configuration, and internal link architecture established in Phase 1. Sites without that foundation in place before publishing will index slower, sometimes significantly slower. The month of infrastructure work before any content went live was not overhead. It was the reason the content compounded as fast as it did.
AI citations are a trackable, optimisable output. Bing’s AI citation tracking is available through Bing Webmaster Tools. The pages generating the most citations shared common characteristics: tight topical focus, FAQ-structured sections, clear entity attribution, and schema markup. These are optimisable inputs, not random outcomes. Treating AI citation as a target metric rather than a side effect changes how you approach page structure from the outset.
Who Programmatic SEO Is and Is Not Right For
Good candidates: Products with multiple modules or distinct use cases that map to separate buyer query clusters. Platforms serving multiple verticals or geographies with the same underlying product. Businesses in markets where keyword-to-feature alignment is high and existing content is generic or thin.
Poor candidates: Niche single-product companies with a limited keyword surface area. Businesses in markets already saturated with high-quality, well-structured content from established players. Companies without the technical infrastructure or editorial oversight the programme requires to stay compliant.
The most common programmatic SEO failure is treating it as a content volume shortcut. Pages generated without genuine keyword-to-feature alignment produce impressions without clicks, clicks without conversions, and eventually invite a manual action. The programme described here worked because every page answered a question a real buyer was typing, matched to a real capability in the product. That constraint is what separates a p-SEO programme that builds pipeline from one that damages the domain.
Engagement period: December 2025 to April 2026. All figures sourced from Google Search Console, Google Analytics, and Bing Webmaster Tools data verified with Tempus Central.
Want to see if programmatic SEO fits your business?
30 minutes. We'll tell you exactly what we'd focus on first.