We’re not going to show you invented case studies. SerpInsight is new, and the client results we’d need to fill this page honestly don’t exist yet. What we can show you is exactly how we plan a campaign — three worked scenarios with the reasoning, the arithmetic and the failure modes all written down.
Why there are no client numbers on this page
Almost every link building agency case studies page you’ll read is unverifiable. A screenshot of a rising line, a first name, an industry, a percentage. You can’t check any of it, and neither can we. We’d rather start from a position we can defend, so here’s the position: SerpInsight has just launched, we don’t have twelve months of permissioned client data, and we’re not going to manufacture some.
What we do have is a method. A link insertion (also called a niche edit) is a link placed inside an article that already exists, is already indexed, and already has readers — and deciding which article, with which anchor, at what pace is the whole job. That reasoning is transferable, so we’ve written three planning scenarios below in the same format we’d use in a kickoff call with you. Every number in them is a plan, not a result.
Illustrative scenario A — B2B SaaS page stuck at position 12
Illustrative scenario — not a client result
The starting situation. A project management tool with a DR in the low 40s. Their comparison page targets a commercial query doing roughly 900 searches a month, and it’s sat between position 11 and 14 for five months. The page itself is fine — long enough, matches intent, converts when people land on it. It has four referring domains. The three pages above it have between 30 and 70 each.
Why we’d choose this tactic mix. The page isn’t losing on content, it’s losing on the link gap, and the gap is roughly an order of magnitude. Guest posts would work but they’re slow and the new post has to earn its own authority first. Insertions into articles that already rank for adjacent queries pass value sooner, so we’d plan about 80% insertions and keep a couple of guest posts in reserve for the two or three sites that won’t sell an edit.
Velocity and anchors. Four to six links a month for four months — call it 18 to 22 placements. Anchor mix roughly 45% branded, 30% partial or topic-based, 15% naked URL and generic, and no more than 10% exact-match, which on 20 links is two. We’d place the exact-match ones last, not first.
Timeline before the data means anything. Nothing. Genuinely nothing for the first six weeks. Links need crawling, the host pages need recrawling, and position 12 wobbles by three places on its own. We’d look at week 10 and take a real read at month four.
How we’d measure it. Average position for the target query in Search Console at 28-day granularity, not daily rank tracker noise. Plus impressions on the page as a whole, because a page climbing from 12 to 7 usually picks up long-tail queries before the head term moves.
What commonly goes wrong. The client sees a two-place dip in week three and asks to change the anchor plan. Or the comparison page gets rewritten mid-campaign and the intent shifts. Or — most often — position 8 turns out to be the ceiling because the four pages above it are all from domains twenty times the size, and the honest answer becomes “this query isn’t winnable this year, here are three you could win instead”.
Illustrative scenario B — local service business in one city
Illustrative scenario — not a client result
The starting situation. An independent electrical contractor covering one metro area. DR 14, eleven referring domains, most of them directories. They rank third or fourth in the map pack on their main term but their service pages barely rank in the blue links at all. Their competitors aren’t strong — the best of them is DR 28.
Why we’d choose this tactic mix. Here’s the part most agencies skip: at DR 14 with eleven referring domains, paid insertions are not the first thing this business needs. Citations, a properly categorized Business Profile and real service pages for each suburb will move more than the first ten links will. So we’d plan a light programme — two links a month, maximum — and spend the first month on the free stuff. If a business below roughly DR 20 comes to us wanting volume, we say so.
Velocity and anchors. Two placements a month for six months, ideally at least half of them on sites with genuine local relevance: the regional news site, a trade association blog, a city guide with actual traffic. Anchors skew hard to branded here — 55% or so brand and brand-plus-city, 25% partial, the rest naked URL. Exact-match commercial anchors on a small local site look exactly like what they are.
Timeline. Local moves slower to start and then jumps. We’d expect nothing readable for two months, then watch months three through six.
How we’d measure it. Non-branded organic clicks to service pages, ranked positions from a grid of points across the service area rather than one centre-point check, and phone calls if they’ll let us see the call log. Rankings are the proxy; calls are the thing.
What commonly goes wrong. The map pack and the organic results move independently, so the client sees no change in the pack for months while the organic pages climb steadily, and concludes it isn’t working. Second failure mode: relevant local hosts with 1,000+ monthly organic visits are genuinely scarce in a small city, and we’d rather deliver one good placement in a month than two bad ones.
Illustrative scenario C — e-commerce category page before a seasonal peak
Illustrative scenario — not a client result
The starting situation. An outdoor equipment retailer, DR 38. Their camping category page ranks around position 9 for a term whose volume triples between March and June. It’s January. The category page has decent internal links and almost no external ones.
Why we’d choose this tactic mix. Timing drives everything. Links placed in June are worthless for a June peak, so the plan is front-loaded: place early, let it settle, hold a small reserve for reactive placements once the season starts and journalists are writing round-ups anyway. Insertions into existing gear guides work well here because the host articles get their own seasonal traffic lift, which means real referral clicks as well as the link.
Velocity and anchors. Six a month in January and February, three a month in March and April, then two held back. Anchors: 40% branded, 35% partial and topic (“this camping gear range”, “their tent selection”), 15% naked URL, 10% exact-match at most, and we’d place exact-match only on the strongest hosts.
A plain timeline, for once:
How we’d measure it. Year-on-year comparison, not month-on-month — the whole category rises seasonally whatever we do, so a March-versus-January chart tells you nothing. We’d compare this season’s non-branded clicks and revenue against the same weeks last year, and against a control category we didn’t touch.
What commonly goes wrong. Someone claims credit for the seasonal curve. That’s the big one, and it’s why the control category matters. The other failure mode is inventory: category pages that go out of stock during the peak lose engagement signals fast, and no amount of link work fixes a page full of unavailable products.
How the three scenarios compare
| Scenario | Planned velocity | Exact-match ceiling | First honest read | Main risk |
|---|---|---|---|---|
| B2B SaaS, position 12 | 4–6 per month, 4 months | 10% | Month 4 | Query may not be winnable |
| Local service, one city | 2 per month, 6 months | Under 5% | Month 3 | Too few relevant local hosts |
| E-commerce seasonal | 6 then 3 then 2 per month | 10% | Year-on-year, next season | Seasonality mistaken for results |
The things all three plans have in common
-
We check whether links are even the constraint
If the page doesn’t match intent, or the site is technically broken, or the domain is below roughly DR 20 with nothing else in place, links are the wrong purchase. We’d tell you that before quoting.
-
Every host is vetted against the same standard
At least 1,000 monthly organic visits, the host page indexed and ranking for a real keyword, shared search intent with your target page, and three or more other articles on the site in the same cluster. The paragraph has to still read correctly with your link removed.
-
You approve the host page before placement
You see the URL, the metrics and the paragraph we’re proposing to edit. If you don’t like it, we find another. Nothing goes live on a guess.
-
The anchor plan is set before the first order
Roughly 40–50% branded, 25–35% partial or topic, 10–20% naked URL and generic, 5–10% exact-match. We write it down at the start so nobody drifts toward exact-match under pressure.
-
We agree what “it worked” means in advance
A metric, a comparison period and a control where one exists. Agreeing this after the fact is how everyone ends up arguing.
What we’ll publish here once we have it
When SerpInsight has client campaigns that have run long enough to say something meaningful, this page changes. Here’s what will be on it, and what won’t.
- Written permission from the client before anything about them is published, including the anonymized versions.
- The methodology attached to every number: which metric, which tool, which date range, which comparison period.
- What else was running at the same time. If they also rewrote the page or launched paid search, that gets stated, because otherwise the link work takes credit it didn’t earn.
- The campaigns that underperformed, written up with the same care as the ones that went well.
Until then, judge us on the standards rather than the stories. The vetting rules are published, the pricing is published, the anchor ratios are published, and the 12-month replacement guarantee is written into the terms rather than implied in marketing copy.
One honest caveat about all of this
Paid link placement is contrary to Google’s published guidelines. Anyone selling you links who doesn’t say that out loud is managing your impression of the risk rather than the risk itself. We plan conservatively — modest velocity, brand-weighted anchors, hosts with real readers — because that’s what reduces exposure. It doesn’t remove it.
Can I see a real client result before I order?
Not yet, and we’d rather say that than fabricate one. What you can see before you spend anything is the specific host page we propose for your link, its traffic and DR figures, and the paragraph we intend to edit — that approval step happens on every single order, including the first.
How long before I should judge a campaign?
Four months in most competitive niches, three in quieter ones. Placements take 7–14 days each, the host pages then need recrawling, and single-week rank movement is mostly noise. If someone promises a read at week two, ask what they’re measuring.
Would you turn down a campaign like scenario B?
We’d scale it right down rather than turn it away — two links a month instead of ten, and a first month spent on citations and service pages that cost nothing. For a DR 14 local business, selling twenty links would be the easier sale and the worse advice.
What happens if a placed link disappears?
We replace it free for 12 months from placement, on a host of equal or better standard. That covers removal, an unagreed nofollow, a 404 on the host page and deindexing. It doesn’t cover you changing your own target URL without telling us — see the refund policy for where the line sits.
Where to go next
How placements actually work
The vetting standard, the outreach process and what you approve before anything goes live.
Learn more
What it costs
$87, $147 and $247 per link by host tier, with bulk discounts at 10 and 25 links.
See pricing
Get a plan for your site
Tell us the target page and the query. We reply within 24 hours, including when the answer is “don’t buy links yet”.
Contact us
If you want the reasoning behind the anchor ratios used in all three scenarios, that’s set out in our guide to anchor text distribution. For velocity, see how many backlinks per month actually makes sense, and for the risk question, our take on whether paid backlinks are safe.