Search & growth

SEO & Digital Marketing Services

SEO is three different jobs sharing one name. Getting a site crawled, indexed and rendered cleanly is an engineering problem. Getting it to answer what people actually type is an editorial problem. Getting other sites and people to reference it is a slow reputation problem. We tell you which of the three is holding you back, and how long the fix realistically takes.

What this service covers

  • On-page & off-page SEO
  • Technical audits: crawling, indexing, canonicals, speed
  • Content strategy built around search intent
  • Social media marketing and distribution
  • Conversion tracking, so the work can be judged

The three layers

Technical, content and authority are separate problems

Most proposals bundle all three together and then report each month on whichever one happens to be moving. They behave differently: one is finite engineering work, one is ongoing editorial work, and one is slow and largely outside your control.

Knowing which of the three layers is your constraint is what decides where the budget should go. Everything below is an attempt to answer that question honestly for your site rather than to sell you all three.

Technical: crawl, index, render, speed

This is the only layer with a defined end state — a page is either reachable, unique and fast, or it is not. It covers one canonical URL per piece of content; consistent handling of trailing slashes, www and non-www, http and https, and tracking parameters that quietly mint thousands of near-duplicate URLs; a robots.txt that does not block the CSS and JavaScript Google needs to render the page; and an XML sitemap listing only live, indexable, canonical URLs.

It also covers structured data that describes what is genuinely on the page, and Core Web Vitals. An audit's job there is narrow: identify which metric is failing, on which templates, and for what share of real visits — the repairs themselves are build-time decisions and belong to web development rather than to a search retainer. Technical fixes take effect within a crawl cycle or two, or they do nothing at all.

Content: intent, not keyword density

A query has a shape behind it. Someone typing a question wants an answer; someone typing a product category wants options and prices; someone typing a brand name plus "login" wants a link, not an essay. Ranking follows from matching that shape better than the pages already there — the right format, the right level of detail, and an answer near the top rather than eight hundred words of preamble.

Repeating the target phrase does nothing; publishing near-identical pages for twenty cities is the doorway pattern Google's own spam policies name. Content work is never finished, because the queries and the competing pages keep changing.

Authority: the slow, hard part

Links from sites that are themselves referenced, plus a consistent brand entity — the same company name, address and profiles wherever you appear — are what let a page compete for terms other people also want. This layer is slow because it depends on other people deciding you are worth citing.

It is the part nobody can shortcut honestly: bought link packages, private blog networks and reciprocal-link schemes are link spam under Google's guidelines, and the downside lands on you, not the vendor. Realistic authority work means things worth referencing, correct profiles and sameAs data, and chasing the unlinked brand mentions you have already earned.

Start here

What a technical audit actually turns up

An audit is not a 90-page PDF of colour-coded warnings. It is a short list of things that are demonstrably wrong, each with the fix and an honest note on whether fixing it will change anything.

An audit is only useful if it separates the findings worth a developer day from the ones worth nothing at all. Both appear in ours, labelled as such.

We are candid about this because our own site made these mistakes. A landing page sat live with no internal links pointing at it from anywhere on the site, so it was effectively invisible to both crawlers and visitors. The fix was the related-services block at the foot of this page, which now keeps every service page linked to its siblings.

Our services page then listed seven service names in a few hundred words and therefore matched almost nothing anyone searches for; the fix was giving each service its own page with something to say, which is why the page you are reading exists. Neither problem was a coding failure, and a clean crawl report would not have flagged either one.

Common findings

  • Duplicate URLs: the same content reachable at several addresses, with no canonical agreeing on which one counts
  • Broken canonicals: tags pointing at the homepage, at a redirect, or at a URL that returns 404
  • Leftover noindex: a staging tag still live in production months later
  • Orphan pages: real content nothing links to internally
  • A dishonest sitemap: advertising redirects, 404s and noindexed URLs
  • Slow mobile rendering: uncompressed hero images and render-blocking scripts wrecking Largest Contentful Paint
  • Invalid structured data: markup that does not match the visible page, or fails validation outright
  • Crawl traps: faceted filters and session parameters generating a near-infinite URL space
  • Headings as styling: multiple h1 elements, or levels chosen for size rather than structure
  • Broken hreflang: missing or contradictory on multi-language and multi-country sites
  • An indexable staging site: a second copy of your content competing with production

Practices

Four decisions that quietly decide whether a site keeps its traffic

These are the rules we hold ourselves to on this site, and the ones we argue for on yours. None of them is clever. All of them are the kind of thing that costs months of traffic when handled casually.

Consolidate domains in the right order

Two domains serving the same content split every signal you have. The fix is to pick one and permanently redirect the other — but the order matters. Capture the Search Console baseline for the domain you are retiring first, then redirect, then file the Change of Address, or you lose the ability to tell whether the dip afterwards was the migration or something else entirely. On this site that redirect sits behind a configuration flag that stays off until the baseline exists, and the target is derived from one canonical URL setting rather than a hostname typed into a template.

One entity, referenced everywhere

Most sites emit a fresh, anonymous Organization block on every template, so a crawler sees dozens of unlinked look-alike entities instead of one company. We define the organization once with a stable identifier and have every page — service pages, FAQ blocks, breadcrumbs — reference that identifier instead of restating it. The verified profiles that feed the same entity are only the ones genuinely owned; padding that list with unverified profiles is a common and pointless habit.

A sitemap that cannot lie

A sitemap listing URLs that redirect, 404 or carry a noindex tag teaches a crawler to trust it less. Ours is generated from the same guard that decides whether a page is routable at all, so a service cannot appear in the sitemap before the page behind it exists. That guard is also what stops thin, near-identical pages shipping just because somebody added a slug — which is precisely the doorway pattern the spam policies describe.

Images sized for the surface that uses them

Share cards and favicons have published requirements and both are usually met with whatever logo file was to hand. A portrait logo used as a large summary card gets letterboxed or dropped, so the share image is a purpose-made 1200 by 630. A favicon has to be square and a multiple of 48 pixels to be used at all, so the icon is a square 512. Neither will move a ranking; both decide whether your link looks like a real result when someone shares it.

Measurement

Impressions are a vanity metric. Clicks and conversions are not.

Almost every argument about whether SEO is working is really an argument about which number is on the first slide. Some of the available numbers describe what buyers did. The rest describe what a search engine displayed, and those are the ones that move most easily without anything improving.

A report leading with impressions and average position is usually a report avoiding the click column.

The table below takes the numbers you are likely to be shown and says what each one honestly reports and where it misleads. One pattern no single metric will show you: clicks rising while enquiries stay flat. That is a real finding rather than a measurement error, and it usually means the traffic is the wrong traffic, or the page ranks for a question your product does not answer.

Instrument the conversions first

If enquiries are not tracked, the work cannot be judged and any claim about it is unfalsifiable. Before campaign work begins we make sure form submissions fire a conversion event, that each enquiry records which page and channel produced it, and that Search Console and analytics are both connected and agreeing on the property.

This site does the same thing: every enquiry form carries a hidden source field, so leads arrive in the admin panel already tagged with the page they came from. It is unglamorous and it is the difference between a report and a story.

What each number is actually telling you

Metric What it genuinely reports Where it misleads
Impressions Your page was shown somewhere in a result set, at any position. Rises when you start appearing at position 40 for hundreds of queries nobody clicks. It costs nothing to inflate.
Average position The mean position across every query a page appears for. Averages a strong ranking for your own brand name with a weak one for the term you actually want, and hides both.
Rank-tracker positions Where one tool, from one location, saw your page on one day. Results are localised and personalised, so a tracked position is a single sample rather than the position your buyers see.
Sessions and users How many visits arrived in total. Blends organic with paid, email, referral and your own staff. It says nothing about search unless it is split by channel.
Bounce rate, time on page How people behaved once they arrived. A page that answers the question in its first paragraph and sends someone away satisfied looks identical to a page that failed them.
Clicks by query How many people chose your result, and what they had typed. Honest, but rare queries are withheld for privacy, so very low-volume terms are under-reported rather than absent.
Click-through rate at position Whether your title and description earn the click at the rank you actually hold. Only comparable between similar positions and similar query types. Brand queries flatter the average badly.
Tracked conversions Enquiries, calls, sign-ups and orders attributed to a page and a channel. Only as good as the tracking behind it, and it breaks silently when a form or a tag changes. It needs checking, not assuming.

Working together

What we need from you, and what usually stalls the work

Engagements rarely fail on strategy. They fail on access and approvals, and both are settled early or not at all. Four things decide whether the work moves at the pace it should.

  • Data access. Read access to Search Console for every property you own, and to your analytics. If a property does not exist yet we create it immediately, because it starts collecting on the day it is verified and cannot be backdated.
  • A way to ship changes. Either access to the site and its repository, or a named developer with time actually booked. A prioritised fix list sitting unread in an inbox is the most common reason a technical audit produces nothing.
  • Someone who knows the subject. Pages that outrank incumbents contain specifics only your team has: how you price, what customers consistently get wrong, why you turn certain work down. We can write and edit; we cannot invent those.
  • Permission to remove things. Merging near-duplicate pages, redirecting dead ones and deleting content that never earned a click is often the highest-value work available, and it needs somebody willing to sign off on losing pages.

The most common cause of a stalled engagement is not disagreement about tactics. It is a fix list everybody agreed with in month one that nobody was given a day to deploy.

Four things that go wrong mid-engagement

All four are avoidable, and all four are easier to head off if they are named at the start rather than discovered in a monthly report.

  • The tracking setup changes without warning. Another vendor edits the tag container or the analytics property, and before-and-after numbers stop being comparable at exactly the moment you need to compare them.
  • Drafts wait on approval. Content sits with a director or a legal reviewer for weeks. The cadence collapses quietly, and unless the delay is written into the report it looks like the work slowed down.
  • A staging or preview copy goes public. A second indexable copy of the site competes with production for the same terms, and it usually surfaces first in a brand search, which is how most people find out.
  • Month two gets judged on rankings. The content commissioned in week three has not been published yet, so the report shows the technical fixes and little else. That is the expected shape of a second month, not a warning sign.

Timelines

How long before anything moves

On a domain with little existing authority, expect one to two quarters before organic performance shifts in a way that is distinguishable from noise, and longer for competitive commercial terms.

The three layers move on different clocks. A robots rule blocking a whole section, or a canonical pointing at the wrong page, can be corrected on a Monday and be visible in index coverage within days, because the only wait is a recrawl. A new comparison article aimed at a term with entrenched pages behaves nothing like that: indexed in days, positioned badly for a while, and only settling after several weeks as the search engine sees how people interact with it. Authority moves on a scale of months.

A short engagement can clean up the technical layer and leave the site measurably better. It cannot buy authority. Seasonality, algorithm updates and competitors publishing against you all sit outside anyone's control.

If a proposal promises rankings inside a month, ask which layer it plans to move and how.

Nobody can guarantee a number one ranking

Search results are ranked by systems nobody outside the search engine controls, personalised by location and history, and reshuffled by updates on no announced schedule. Google's own documentation for site owners says the same thing: no one can guarantee a top ranking.

Anyone selling one is either describing a term nobody searches for, counting a result they can already see for your brand name, or planning something that risks a manual action against your domain. We will tell you what we think is achievable and what we think is not, and we will put the uncertain part in writing as uncertain.

Scope

Local SEO and national ranking are not the same race

Both are search, and almost nothing transfers between them: different surfaces, different ranking factors, different work. Establish which one your enquiries genuinely come from before a budget gets split across the two.

Local: the map pack

Local results are dominated by a Google Business Profile rather than by your website. The work is different in kind: claiming and verifying the profile, choosing accurate primary and secondary categories, keeping the name, address and phone number identical across the site and every directory, publishing correct opening hours, adding real photos, answering questions, and earning reviews steadily rather than in a suspicious burst.

Proximity to the searcher is a heavy factor, so results vary street by street and a single "our ranking" screenshot means very little.

National and international: no proximity to help you

Ranking well in your own city does not transfer. The moment the query drops its location, proximity stops rescuing you and you are competing on content depth and authority against everyone in the country. Businesses that dominate their local map pack are routinely invisible nationally, and this surprises people.

Across markets it gets harder again: currency, spelling, statutory details and search behaviour all differ, and hreflang has to be right or the wrong country's page gets served. Our own market scope covers United States, United Kingdom, Canada, Australia and several others, which is exactly why we scope local and national ambitions as separate pieces of work.

Distribution

Social media marketing, described accurately

Links from social platforms do not pass ranking credit directly; they are almost all nofollowed. What distribution does is put the work in front of the people who might cite it — which is how the authority layer actually gets fed.

It also gives you a channel that does not depend on a search engine's next update. Treated as a publishing habit rather than a growth hack, it is worth doing: a cadence you can sustain, posts written for the platform rather than pasted across all of them, and a link back to something substantial on your own site.

This service is organic work — on-page and off-page SEO, content and social distribution. Where a term is genuinely better bought than earned we will say so, and any advertising budget stays yours, paid directly to the platform and reported separately from organic performance rather than blended into it.

A typical engagement

  • Audit: crawl, index coverage, speed, structured data, analytics
  • Fix and instrument: clear the technical layer, then make conversions countable
  • Map intent: decide what to write, what to merge and what to delete
  • Publish and distribute: ship the content, pursue mentions and citations
  • Report and re-plan: clicks and conversions, then what changes next quarter

An honest limit

When search is not your best channel

SEO can only capture demand that already exists as typed queries. Several common situations do not meet that description, and in each of them a retainer would be the wrong recommendation.

  • Nobody is searching yet. A genuinely new category has no established vocabulary, so there is nothing to rank for. Demand has to be created elsewhere first — outbound, partnerships, communities, or paid tests run purely to find the words people do use.
  • You need enquiries within weeks. Nothing organic works on that timescale. Paid search does, and we would rather say so than sell you a quarter of patience you cannot afford.
  • Marketplaces own the result page. For some queries the entire first page belongs to aggregators and directories. Being well ranked and well reviewed inside the aggregator is often the realistic route to those buyers, not displacing it.
  • The decision runs through a handful of named accounts. Where the buyers are a short, known list, search plays a supporting role at best. A credible site that stands up to scrutiny matters; ranking for the category term does not.

We would rather decline an SEO retainer than take one where the constraint is somewhere else. The audit is the cheapest way to find out which case you are in.

Pricing

Quoted per project, in US dollars

SEO scope varies far too much for a list price. A single-site technical audit and fix is a different job from an ongoing content and outreach programme, so we scope after looking at the site, the analytics and the queries you care about, then quote a fixed figure before any work starts.

Where SEO sits next to build work, those prices are published: website development starts at $399, and ongoing website maintenance — updates, patches, backups and uptime monitoring — starts at $49 per month.

Website build
$399

Starting price for a responsive, indexable site.

Maintenance
$49/mo

Updates, security patches, backups, monitoring.

SEO & marketing
Quoted per project

Fixed quote after the audit, agreed before work begins.

All prices in US dollars (USD). We can invoice in GBP, EUR, AUD or INR on request.

SEO & Digital Marketing — FAQs

On a domain with little existing authority, expect one to two quarters before organic clicks move in a way you can distinguish from normal fluctuation. Technical fixes can register within days once pages are recrawled. New content usually needs several weeks to settle, and link and brand authority move over months, not weeks.

No, and nobody credible can. Rankings are decided by systems outside any agency's control, personalised by location and history, and changed by updates on no announced schedule. Google's own guidance for site owners says the same. A guaranteed number one usually means a term nobody searches, or a tactic that risks a penalty on your domain.

Impressions only mean your pages were shown somewhere in a result set, often at low positions for queries nobody clicks. They are cheap to inflate and mean little on their own. Look instead at clicks by query, click-through rate against your actual position, and tracked conversions. Flat enquiries usually means the traffic is the wrong traffic.

A crawl of the site, plus index coverage, canonical and duplicate-URL handling, redirect chains, robots and sitemap correctness, orphan pages, heading structure, structured data validity, Core Web Vitals on mobile, and hreflang where relevant. You get a short prioritised list of what is genuinely wrong, the fix for each, and an honest note on likely impact.

Local results are largely driven by your Google Business Profile and by how close the searcher is to you. Once a query drops its location, proximity stops helping and you compete on content depth and authority against the whole country. Local strength does not transfer, so national ranking has to be scoped as separate work.

The URL mapping and redirect planning are the same as for any rebuild. What is specific to a domain change is the reporting and the entity. Export the Search Console baseline for the old property first, because it cannot be backdated and without it a normal post-move wobble is indistinguishable from a real loss. File the Change of Address once the permanent redirects are live, then update the profiles and directory listings that still name the old domain.

Yes, otherwise the work cannot be judged. If form submissions, calls and sign-ups are not recorded against the page and channel that produced them, every claim about performance becomes unfalsifiable. We set up conversion events and source tagging first, so later reporting is about enquiries rather than about impressions.

No. Bought links, private blog networks and reciprocal-link schemes are link spam under Google's guidelines, and the consequences land on your domain rather than on the vendor. We work on things worth citing, correct business profiles and consistent brand data, and on converting existing unlinked mentions into links.

Read access to Search Console and analytics for every property you own, a way to get changes deployed — either site and repository access or a developer with time booked — someone internally who knows the subject well enough to answer questions a writer cannot guess, and authority to merge, redirect or delete pages. Missing deployment time is the usual reason an agreed fix list produces nothing.

Rarely by deleting them outright. Pages competing for the same query are better merged into the stronger one and the weaker URL redirected. Pages that earn their keep off-search, such as policies or support notes, can stay and be marked noindex. Check inbound links and conversions first, because low traffic is not the same as no value.

This service is organic: on-page and off-page SEO, content strategy and social distribution. Where a term is realistically better bought than earned we will tell you, and any advertising budget stays with you and is paid directly to the platform. We keep paid and organic reported separately rather than blended, so neither channel can borrow credit from the other.

Ping us for your IT solutions

Start with an audit, not a retainer

Send us the site and the terms you care about. We will tell you which of the three layers is your constraint, what we would fix first, and what we think is realistically achievable — including when the honest answer is that search is not your best channel.

Mon–Fri, 09:00–19:00 IST (UTC+5:30) — 4+ hours of live overlap with US Eastern; full overlap with UK and EU mornings.

Related services

Web Development

Fast, scalable websites and web apps to develop web application solutions with Laravel, React and modern stacks.

Mobile App Development

Native and cross-platform iOS & Android apps built by the best app developers, with React Native and Flutter.

Custom Software

Custom software development for bespoke CRM, ERP and SaaS platforms, engineered around your exact workflow.