Digital Marketing Masterclass

Week 1 — the difference between marketing and posting. One page, one audience, one measurable goal.

1. Who exactly are you talking to

   "Everyone aged 18–35" is not an audience; it is a way of avoiding the decision. Narrow it until your content writes itself.

   The lesson

   "Everyone who needs a job" is not an audience. A usable one is narrow enough to picture: "final-year commerce students in tier-2 cities who want an analyst role and have never used SQL". You can name where they are, what they search for and the sentence that stops them scrolling.

   Write three of them, pick one, and write down what they already believe, what they are afraid of and what they have already tried. Every post, ad and page you make for the next month should be recognisable as being for that one person. One audience done well beats five half-served.

   The mistake is writing for the algorithm instead of a person: hashtag-stuffed captions and headlines that describe your service. Say the thing your reader is thinking — "aptitude tests again?" — and the right people stop because they feel described.

   Plain textCopy

   Narrow enough to picture:
     "Final-year commerce students in tier-2 cities who want an
      analyst role and have never used SQL."
   
   Write down: where they are . what they search . what they believe .
               what they fear . what they already tried.
   One audience done well beats five half-served.

   Example "Final-year engineering students in tier-2 cities applying for their first internship" tells you the platform, the language and the proof they need all at once.

   Practice Write a one-paragraph audience description for a product you actually know, and name where they already spend time online.

2. One goal, one number

   A campaign goal must be countable and dated: 40 sign-ups in 30 days. "Increase awareness" cannot be measured, so it cannot be managed.

   The lesson

   A campaign has one goal and one number attached to it. Not "awareness and engagement and signups" — either 200 newsletter subscribers this month, or 50 enquiries, and you can name the number out loud. Everything else is a supporting metric you read, not a goal you chase.

   Put the number and the date at the top of a one-page plan, then list the three activities that could produce it and how many you will do. Ten posts a week, two email sends, one paid test. When the month ends you will know exactly which of the three moved and by how much.

   The mistake is measuring likes because they are visible. Likes do not pay salaries and they do not correlate reliably with anything you want. Count the action you chose — clicks to the tool, form fills, replies — even when the number is embarrassingly small, because small and real is the only number you can improve.

   Plain textCopy

   Goal:  200 newsletter subscribers
   By:    31 Oct 2026
   Plan:  10 posts/week  -> aim 120
          2 email sends  -> aim  50
          1 paid test    -> aim  30
   
   Count the ACTION (form fills). Not likes. Likes do not pay salaries.

   Example 40 sign-ups at ₹100 each is a ₹4,000 budget you can defend. "Get more users" is a conversation that never ends.

   Practice Set one number and one deadline for your campaign, and write down what would make you cancel it.

3. Content that answers a real question

   Content that ranks answers what people already type into search. Content that does not is advertising nobody asked for.

   The lesson

   Good content answers a question somebody typed into a search box or asked a friend. "Is a BPO job a real career?" is a question. "10 tips to succeed" is not — nobody asks it, so nobody searches for it, so nobody finds it.

   Keep a running list of the questions you are asked in real life — on WhatsApp, in college, in comments — and turn the top one into a post this week. Answer it completely, in the first two lines, before any preamble. The answer-first shape is what gets read on a phone and what search engines lift into results.

   The mistake is a post that builds up to the answer and saves it for the end, or promises "5 secrets" and delivers nothing. If your answer would not help someone reading only the first paragraph, it is not ready to publish.

   Plain textCopy

   Real question:  "Is a BPO job a real career?"   (people search it)
   Not:            "10 tips to succeed"            (nobody asks it)
   
   Answer in the first two lines, then add detail.
   Turn your top WhatsApp / college question into this week's post.

   Example "How to write a resume with no experience" earns traffic for years; "We are the best career platform" earns nothing.

   Practice List ten questions your audience actually asks, in their words, not yours.

Week 2 — analytics. Without this, every later decision is opinion.

1. GA4 basics

   GA4 records events, not just page views: a click, a sign-up and a scroll are all events you can look at afterwards.

   The lesson

   GA4 is the analytics tool bolted onto almost every website, and it answers three questions: how many people came, where from, and what they did. Install the tag once, then look at Reports → Engagement → Pages to see which pages actually hold attention, and Acquisition → Traffic acquisition to see where the visits came from.

   The single most useful feature is an Event: gtag('event', 'score_check_started') marks the moment somebody begins your tool, and score_check_completed marks the moment they finish. Those two events turn a report about visitors into a report about conversion, and the gap between them is where your design loses people.

   The mistake is installing the tag and never checking it. Open the real-time report on your phone while you load the site: if you do not appear, the tag is on a page, not on the site. And resist counting sessions as people — one person refreshing ten times is not ten visitors.

   JavaScriptCopyTry it

   // Fires when someone starts the tool.
   gtag('event', 'score_check_started', { tool: 'altigrad' });
   
   // Fires when they finish it.
   gtag('event', 'score_check_completed', {
     tool: 'altigrad',
     score_bucket: '70-100'
   });
   
   // The gap between started and completed is where your design loses people.

   Example Users, sessions and conversions answer three different questions — mixing them up is how a report ends up claiming growth that did not happen.

   Practice Install a tracking tag on a test page, then confirm your own visit appears in the real-time report on your phone.

2. Sources, mediums and campaigns

   UTM parameters are the labels that record WHERE a visitor came from, so "traffic went up" becomes "traffic from this one post went up".

   The lesson

   Every visit arrives with a label, and the label is what tells you which of your efforts worked. Three parts: source is where they came from (google, instagram), medium is how (organic, paid, email, referral), campaign is which push (altigrad-launch). Untagged links get lumped together as 'direct' and teach you nothing.

   Build the tag into every link you post outside your own site: ?utm_source=instagram&utm_medium=social&utm_campaign=altigrad-launch. Keep a spreadsheet of the exact tags you used, because instagram, Instagram and ig become three different sources and three useless rows.

   The mistake is a link shortener that strips the query string, so every campaign reports as direct traffic. Test the link once in a private window, then check the report the next day that the visit actually landed under the campaign name.

   Plain textCopy

   https://way2fresher.com/tools/?utm_source=instagram&utm_medium=social&utm_campaign=altigrad-launch
   
   Keep a sheet of the exact tags you used.
   instagram / Instagram / ig become THREE different sources and three useless rows.
   Test the link in a private window; confirm it reports the next day.

   Example Adding ?utm_source=linkedin&utm_campaign=august to a link is what makes that traffic attributable — without it, the same visit is lumped into "direct".

   Practice Tag three links to the same page with three different campaign names and confirm all three appear separately in analytics.

3. The report a manager reads

   A weekly report is four lines: what we spent, what we got, what changed, what we do next. Anything longer is not read.

   The lesson

   A marketing report is one page: what we set out to do, what happened, what changed, what we do next. Numbers with a comparison — this month against last, or against the target — because a number on its own tells nobody whether it is good.

   Put the goal at the top, the result immediately under it, then two or three charts that explain the gap, and finish with three sentences of recommendation. If a chart does not change the recommendation, delete it — a report read in ninety seconds is a report that gets acted on.

   The mistake is reporting activity instead of outcome: "posted 20 times" instead of "20 posts brought 180 visits and 12 tool completions". Activity reports feel safe and change nothing, which is exactly why they go unread.

   Plain textCopy

   GOAL 200 subscribers . RESULT 180 (90% of target)
   
     [ 180 / 200  ======  90% ]
     Chart 1: subscribers by source
     Chart 2: this month vs last
   
   Recommendation: double the email cadence, drop the paid test.
   A report read in 90 seconds gets acted on.

   Example "₹1,200 spent, 18 sign-ups, cost per sign-up fell from ₹80 to ₹67, doubling spend on the winning ad" is a complete report.

   Practice Write a weekly report in four lines from your own numbers, with no jargon in it.

Weeks 3–4 — search engine optimisation, without the myth-making.

1. Keyword research honestly

   Target phrases you can actually rank for: long, specific, low competition. A fresher cannot beat a national brand on "jobs".

   The lesson

   Keyword research is finding what people type, how many of them do it, and whether you can realistically rank for it. For a new site the last one matters most: "jobs" is a fight you cannot win, while "fresher jobs in Pune for BCom graduates" is one you can.

   Use the free tools — Google autocomplete, "People also ask", Search Console once you have traffic — and group what you find into themes, one page per theme. A page that answers a whole theme outranks several thin pages about fragments of it, and it reads better for a person too.

   The mistake is chasing a high-volume keyword with a brand-new domain and concluding SEO does not work when it does not rank in three weeks. Pick the long, specific phrase you can actually win, get to page one, then move up the ladder.

   Plain textCopy

   Free tools: Google autocomplete . "People also ask" . Search Console
   
   Impossible:  "jobs"                              (huge, competitive)
   Winnable:    "fresher jobs in Pune for BCom graduates"
   
   One page per theme. Win the long phrase, get to page 1, then climb.

   Example "Fresher resume format for BCom graduates" is winnable in weeks; "jobs" is not winnable at all.

   Practice Find ten keyword phrases for your niche, note which look winnable, and pick one as your first target.

2. On-page SEO: the page that answers

   One page, one topic, one target phrase. Title, headings, description and a genuine answer in the first two paragraphs.

   The lesson

   On-page SEO is mostly the discipline of a good page: one clear topic, a title tag that reads like the search somebody typed, one H1, meaningful H2s, and enough real content that the page is obviously the best answer available. Add a meta description that makes the result worth clicking.

   Speed and mobile layout are ranking factors and UX factors at once. Compress images, avoid heavy scripts above the fold, and check the page on a mid-range Android phone — that is the device 99 per cent of an Indian fresher audience is using, and it is the one that decides whether your ranking turns into reading.

   The mistake is stuffing the keyword into every sentence until the page is unreadable. Write for the person, mention the phrase where it naturally belongs — title, first paragraph, a heading — and internal links from your own related pages, which is the most underrated ranking signal a small site has.

   HTMLCopyTry it

   <title>Fresher Jobs in Pune for BCom Graduates | Way2Fresher</title>
   <meta name="description" content="Verified fresher BCom openings in Pune,
     updated daily. See salaries, skill requirements and how to apply.">
   <h1>Fresher jobs in Pune for BCom graduates</h1>
   <h2>What these roles pay</h2>
   <h2>Skills employers ask for</h2>
   <!-- one topic . one H1 . internal links from related pages -->

   Example A 900-word page that fully answers one question beats five 300-word pages that half-answer five — and it is far easier for a search engine to rank.

   Practice Write one full page targeting your chosen phrase, then read it as a stranger and delete anything that does not answer the question.

3. Content that keeps working

   A published page continues to work while a social post is gone in a day. A realistic mix invests in both, but the durable asset is the page.

   The lesson

   One good evergreen page outperforms thirty posts. A page that answers "how to prepare for a fresher interview" keeps bringing in people for years, while a post about today's result is dead in a week. Choose topics that will still be searched next September.

   Refresh rather than pile up: update the numbers, add a new section, add an internal link from a newer post, and re-share it. Search Console tells you which pages almost rank — positions four to fifteen — and those are the ones where an hour of improvement pays the most.

   The mistake is starting a series of twelve posts you cannot finish. Publish one complete page, get it ranking, then decide about the second. Consistency is a strategy, but only of content good enough to be found.

   Plain textCopy

   Evergreen: "how to prepare for a fresher interview"  (searched for years)
   Ephemeral: "today's result"                          (dead in a week)
   
   Search Console -> pages at positions 4-15 -> improve those first:
     update the numbers, add a section, link from a newer post, re-share.

   Example A single well-answered question page can bring traffic for years — and it is also the page a social post should link to, instead of the homepage.

   Practice Publish your page, then promote it in three places, each with a tagged link.

Weeks 5–6 — spending money small and reading the result. The skill most digital-marketing hires are actually testing for.

1. Ads, budgets and the small-test habit

   Run a small test before committing a budget: two audiences, one message, identical everything else. Spending is easy; knowing what worked is not.

   The lesson

   Paid ads buy you information fast. Start with one campaign, one audience and one ad, on a budget small enough that losing it does not hurt — ₹300 a day for a week tells you something real, and a fresh account rarely converts well in week one anyway.

   One change at a time, then wait for the data. Audience, creative, headline, landing page — change one, note the date, compare seven days later. Two changes at once and you cannot say which one moved the number, which is the entire value of running the test.

   The mistake is boosting a post and calling it a campaign: no landing page, no conversion event, no way to know whether anyone did anything. Decide what counts as success before you spend, or the report at the end can only say "we got some reach".

   Plain textCopy

   Test: 1 campaign . 1 audience . 1 ad . Rs.300/day for 7 days
   Change ONE thing at a time. Note the date. Compare 7 days later.
   
   Before you spend, decide what "success" means:
   a real enquiry? a signup? a tool completion?

   Example ₹500 on two ads with the same text and different images answers a real question. ₹10,000 on one ad answers nothing.

   Practice Draft one small ad test with a fixed budget and a stated decision rule for what you will do with each result.

2. Reading cost per click and conversion

   Clicks are cheap and worthless on their own. Cost per SIGN-UP is the number that decides whether a campaign continues.

   The lesson

   Three numbers tell you whether an ad works. CPC — what a click cost. Conversion rate — of 100 clicks, how many did the thing. Cost per conversion — the spend divided by the conversions, which is the only number that compares against the value of what you are giving away.

   Work backwards from the money: if a genuine enquiry is worth ₹500 to you, then 200 clicks at ₹10 with a 2 per cent conversion rate gives four enquiries at ₹500 — exactly break-even, before any of the value of the fourth. If your ad is at ₹40 a click, it can only work with a much better conversion rate or a much better landing page.

   The mistake is judging an ad by its click-through rate. A beautifully written ad that brings curious clicks and no conversions is worse than a plain one that brings fewer, better people. Judge by cost per conversion and by whether the person stayed.

   Plain textCopy

   CPC Rs.10 . 200 clicks . 2% convert  =  4 enquiries
   Cost per enquiry = (200 x 10) / 4     =  Rs.500
   
   If an enquiry is worth Rs.500, this breaks even - before its extra value.
   At Rs.40/click you need a far better conversion rate or landing page.
   Judge by cost per CONVERSION, not click-through rate.

   Example 100 clicks at ₹5 with no sign-ups is ₹500 wasted; 20 clicks at ₹25 with 4 sign-ups is ₹125 per sign-up and worth scaling.

   Practice Calculate cost per click and cost per conversion for your test, and write one sentence on whether you would scale it.

3. The one-page report

   Spend, results, cost per result, what you learned, what you would do next. One page, no dashboards, no vanity charts.

   The lesson

   Your one-page ad report has five lines: spend, clicks, conversions, cost per conversion, and one sentence on what you would change next week. That is the whole document. Anybody can read it in thirty seconds and decide with you.

   Screenshot the best and worst ad side by side, and keep a note of every test and its result so the second month starts from what you learned rather than from the beginning. This running log is also the portfolio piece that gets a fresher a marketing job — it shows judgement, not tool knowledge.

   The mistake is reporting only the campaigns that did well. The test that failed and the money it cost is the single most credible thing in a junior marketer's report, because it proves you were measuring rather than hoping.

   Plain textCopy

   Spend          Rs.2,100
   Clicks         210
   Conversions    4 (enquiries)
   Cost / conv.   Rs.525
   Next week: new landing page, same audience, same ad.
   
   Keep a log of every test and its result - that log is the portfolio piece.

   Example A four-line report plus one honest failed test is a stronger interview artefact than a screenshot of a platform dashboard.

   Practice Turn your campaign into a one-page report and present it to one person in five minutes.

Tests, an owned email list and a calendar that turns posting into a habit.

1. A/B tests and the one-change rule

   A test compares two versions and changes one thing. If you change the headline and the image together, you learn nothing about either.

   The lesson

   Decide the metric and the minimum sample before you start; stopping the moment one side is ahead is how random noise becomes a confident wrong decision.

   Most tests end in "no difference". Write that down too - knowing what does not matter is as useful as knowing what does.

   Plain textCopy

   Test: CTA wording on the tool page
   
   A (control):  "Check your score - free"
   B (variant):  "See what's holding you back - free"
   Change:       the CTA line only (same page, same traffic)
   Metric:       tool completions per visitor
   Sample:       ~400 visitors per side
   Result:       B +18% completions -> keep B

   Example Version A and version B differ only in the call-to-action line, and the result tells you which line earns more clicks.

   Practice Design one A/B test with a single difference, and write the metric and the sample you would need.

2. Email and retention basics

   Email is the only channel you own. One useful email a week to a list that opted in beats a thousand impressions you rent from a platform.

   The lesson

   Grow the list with something worth trading an address for - a checklist, a template, a tool - not a "subscribe" button nobody presses.

   Watch unsubscribes and opens, not just sends; a falling open rate is the list telling you the emails stopped being useful.

   Plain textCopy

   Subject: 3 real openings for BCom freshers (Pune)
   
   Hi Asha,
   
   This week: an accounts role at a CA firm, a back-office role at a
   bank, and an analyst opening that asks for Excel only.
   
   One tip: mention the tool in line one - screeners search for it.
   
   [See the 3 jobs ->]
   
   You get this because you asked for the Monday digest.
   Unsubscribe: [link]

   Example A weekly digest with three real openings and one tip earns inbox attention; a daily blast of promotions earns the unsubscribe.

   Practice Draft one useful weekly email for your audience with a clear subject and one action.

3. Building a 30-day content calendar

   A calendar turns intention into a habit. Map four weeks of posts to the questions your audience actually asks, and the "what do I post today" problem disappears.

   The lesson

   Batch the work: write and design a month in one sitting, then publish on schedule. Consistency comes from batching, not from daily willpower.

   Leave two empty slots for whatever happens that week - a result, a question in the comments - so the calendar stays alive instead of rigid.

   Plain textCopy

   30-day calendar (one post a week, one theme each)
   
   Wk1  Is a BPO job a real career?       short explainer
   Wk2  Skills they actually screen for   carousel, 5 slides
   Wk3  How to apply in 3 steps           screen recording
   Wk4  What these roles pay in Pune      chart + one line
   
   + 2 open slots for whatever happens that week.

   Example Week 1 answers "is this career real", week 2 "what skills", week 3 "how to apply", week 4 "what it pays" - each week one clear theme.

   Practice Build a four-week calendar with one post a week, each tied to a real question, and a format for each.