Kariyer.net’te günlük iş akışlarımızda; yüzlerce excel, pdf, docx, html veya veri raporların çıktıları arasında dolaşabiliyoruz. Fakat her bir kurumsal şirket gibi, bir rapor (veri), doğru kişiye, doğru zamanda, doğru formatta ve güvenli bir şekilde ulaşmadığında, aslında hiçbir manası/faydası kalmıyor.

Microsoft’un Power Automate aracı, Google’ın Workflows ve diğer otomasyon araçları (n8n, Zapier, Make, UiPath vb.) arasında seçim yaparken; özellikle veri güvenliği, gizlilik ve ekosistem uyumluluğu gibi faktörleri dikkate almak önemlidir. Bu konuda Manu, X hesabinda güzel bir karşılaştırma yapmıştır. Kısaca bu karşılaştırmada, uyum, kurumsal-kullanıma-hazır ve yazılım kolaylıklarını anlatıyor. Örnek eğer microsoft ekosistemini kullanıyorsak, Microsoft Teams ile uygumlu bir otomasyon (en azından yaptığımız işlere göre) çok daha önemli.

Bu yazıda Kariyer.net’in iç süreçlerinde kullandığımız otomasyon yapısını, Adaptive Card ve gizlilik benchmark’lara uyumlu akışlarını adım adım anlatmak istiyoruz.

Kısaca Power Automate nedir?

• Microsoft’un iş otomasyonu aracıdır. (n8n benzeri)
• Kod bilmeden drag and drop (sürükle-bırak) ile akışlar (flow) kurulabilir.
• Office 365 (Teams,Dataverse, Devops Azure, Outlook, Excel, Power BI) ile doğal olarak çalışır.

Power Automate, bildirim göndermek gibi basit görevleri veya birden fazla uygulama ve servis arasında çalışan süreçleri yönetebilir. Bir iş akışı (flow) oluşturmadan önce, Microsoft iş e-posta adresinizle Power Automate’e kaydolmanız gerekir. Erişiminizi aldıktan sonra Power Automate’i kullanmaya başlamaya hazırsınız!. Detaylı olarak microsoft blog kısımında bunu anlatmışlar o yüzden bu yazıda daha çok deneyemilerimizi paylaşıyor oluruz.

Power Automate nedir? - Power Automate

Power Automate ana sayfası web üzerinde: make.powerautomate.com ile ulaşabiliyorsunuz:

Ana sayfada akış oluşturma, template, sizin flow listeniz ve AI hub gibi önemli özelikler bulunuyor.

+Create ile flow (akışları) belirlerken öncesinde templatelere bakıp daha hızlı bir yapı hazırlayabiliriz, fakat copilot ile her zaman istediğiniz sonuç olmayabilir, o yüzden kendinizi bir instant cloud ile kontrol etmelisiniz.

Bir flow (akış), Power Automate içinde oluşturulan otomatik işlem zinciridir. Yani, belirli bir tetikleyici (trigger) olayı izler ve ardından eylemler (actions) gerçekleştirir.

Örnek Excel içeriği (OneDrive veya SharePoint’te):

Dosyayı kaydet → tabloyu seç → Insert → Table → tabloya isim ver. TasksTable

Senaryonun mantığı şu şekilde :

Excel’den Görevleri alır ve Teams’e Adaptive Card Gönderır:

Action 1: Trigger

• Trigger (Tetikleyici): Örneğin, “Manuel başlat” veya “Belirli bir saat” (Scheduled). Excel dosyasını seç → Tablo olarak TasksTable seç. Bu adım, tablodaki tüm satırları getirir.

Action 2: Apply to each

Her satır için işlem yapılır (örneğin, görev adı, son tarih gibi).

Action 3: Post Adaptive Card in Teams

Her görev için bir Adaptive Card oluşturulur ve Teams kanalına gönderilir Kartta görev adı, son tarih, durum gibi bilgiler gösterilir. Kullanıcıya (Takım arkadaşlara) “Tamamlandı” veya “Ertele” gibi butonlar eklenebilir.

Mesaj ve Kart İşlemleri:

• Get message details / Get messages: Mesaj bilgilerini veya mesajları alır.
• List replies of a channel message: Kanal mesajına gelen yanıtları listeler.
• Post message in a chat or channel: Sohbet veya kanala mesaj gönderir.
• Post card in a chat or channel: Kart gönderir (ör. Adaptive Card).
• Post adaptive card and wait for a response: Kart gönderir ve yanıt bekler.
• Reply with a message in a channel: Kanalda mesaja yanıt verir.
• Reply with an adaptive card in a channel: Adaptive Card ile yanıt verir.
• Update an adaptive card in a chat or channel: Gönderilmiş kartı günceller.

Adaptive card bir UI (User Interface )olarak düşünürsek ve mesaj’i bu şekilde doğrudan veya kanal’dan yaptığımızda doğru bir format ve tasarım yapmak lazım. Adaptive Card, (Responsive) düzen, İkon, Rozet (Badge), Carousel, Grafikler (Charts) ve çok daha fazla özellikleri sağladığı için daha doğru bir mesaj içeriyor. Detaylı olarak https://adaptivecards.io/ tasarım ve örnekleri inceleyebilirsiniz, fakat en temel bir tasarım örneği bu şekilde:

{
  "$schema": "https://adaptivecards.io/schemas/adaptive-card.json",
  "type": "AdaptiveCard",
  "version": "1.5",
  "body": [
    {
      "type": "TextBlock",
      "text": "Yeni Özellikler Geldi!",
      "weight": "Bolder",
      "size": "Large"
    },
    {
      "type": "Image",
      "url": "https://adaptivecards.io/content/cats/1.png",
      "size": "Medium",
      "style": "Person"
    },
    {
      "type": "TextBlock",
      "text": "Adaptive Card artık Responsive Layout, Icon, Badge ve daha fazlasını destekliyor.",
      "wrap": true
    },
    {
      "type": "FactSet",
      "facts": [
        { "title": "Responsive:", "value": "Evet" },
        { "title": "Icon:", "value": "Destekleniyor" },
        { "title": "Badge:", "value": "Yeni" }
      ]
    }
  ],
  "actions": [
    {
      "type": "Action.OpenUrl",
      "title": "Detayları Gör",
      "url": "https://adaptivecards.io"
    }
  ]
}

https://adaptivecards.microsoft.com/designer.html içinde çalıştırarak bu çıktığı elde ediyoruz:

Tekrar bunları yaptıktan sonra, flow’u incelersek:

Excel Hazırlığı: Dosyayı OneDrive veya SharePoint’e kaydetip, verileri tabloya dönüştürüz (Insert → Table) ve tabloya bir isim alır (ör. TasksTable).

List rows present in a table: Excel’deki TasksTable verilerini alır ve bunu her satır için Apply to Each ile uygular. Yanı tüm kullanıcılar bitirine kadar flow çalışır, fakat table mode olması çok önemli.

Post Adaptive Card in Teams: İşlem bilgilerini içeren bir Adaptive Card oluşturup ve Teams kanalına/chat gönderim yapılır.

Bu Akış Neden Önemli? (n8n ile Karşılaştırma)

Power Automate avantajları:

• Microsoft 365 entegrasyonu (Teams, Excel, SharePoint, Outlook).
• Hazır konektörler ve güvenli kimlik yönetimi.
• Adaptive Card desteği ile interaktif akışlar.

n8n avantajları:

• Açık kaynak, daha fazla esneklik. (örnek Gemini, Claude için tercih varsa), Power Automate sınırlar.
• Çok sayıda entegrasyon (özellikle bulut dışı servisler).

Fark:

Power Automate, kurumsal Microsoft ekosisteminde en hızlı ve güvenli çözüm, fakat n8n, daha çok geliştirici odaklı ve özelleştirme isteyen senaryolarda güçlü olabilir. Ek olarak mesele veri ve güvenlik ise, Power Automate daha doğru bir tercihdir.

Bu akış, Excel’deki verileri otomatik olarak Teams’e taşıyarak iş süreçlerini hızlandırır ve manuel işleri ortadan kaldırır. Özellikle görev yönetimi, raporlama veya onay süreçlerinde çok etkilidir.

Son söz olmasada, Power Automate, iş süreçlerini Microsoft Ekosisteminde otomatikleştirmek için bir altyapıdır. Power Automate’ın yeni versiyonu, otomasyon deneyimini daha akıllı (copilot desteği ile), daha entegre ve kullanıcı dostu hale getiren önemli yenilikler olabilir.

Power Automate 101 — Kariyer.net’te bir kurumsal Raporlama Otomasyonu nasıl yapılıyor? was originally published in Kariyer.net Tech on Medium, where people are continuing the conversation by highlighting and responding to this story.

İşçi * Yetenek, İşveren*verimlilik.

Türkiye’de, şuandan itibaren belki de kaç hafta öncesinden en yaygın sorulardan biri, “Maaşınız ne kadar?” veya “Zam yapılacak mı?” olabilir. Bu durum, ekonomik soru olsa da daha çok socio-psikoloji bir his aslında. İnsanların maaş ve alım gücüne olan ilgisinin ne kadar ön planda olduğunu gösteriyor. Özellikle artan enflasyon oranları ve yaşam maliyetleri, maaş zamlarını yalnızca bireysel değil, toplumsal bir mesele haline getiriyor.

Ancak maaş artışları sadece bireysel refah açısından değil, işletmelerin sürdürülebilirliği ve ülke ekonomisinin dengesi açısından da karmaşık matematiksel hesaplamaları gerektiriyor.

Maaş zammının gerçek bir refah artışı sağlayıp sağlamadığını anlamak için “Enflasyon Düzeltilmiş Maaş Artışı” modelini çok faydası da olmuyor ve çok kolay değil. Verimlilik, etik, denge kazanç ve asil insan hislerinden birde kıyas bu formülleri bozabilir.

Basit hesaplar:

Bir kişinin maaşı 20.000 TL, yeni maaşı 23.000 TL ve yıllık enflasyon oranı %50 ise:

Gerçek Maaş Artışı (\%) = \frac{23.000 — 20.000}{20.000} — 0.50 = -0.35

Bu durumda kişi %15 nominal bir zam almış olsa da, %35 oranında gerçek bir kayıp yaşıyor.

Maaş Artışlarının İşletme Üzerindeki Etkileri

Maaş zamlarının işverene maliyetini anlamak için “Zam Dengesi Modeli” kullanılabilir.

Gerekli Gelir Artışı (\%) = \frac{Toplam Maaş Artışı}{Mevcut Gelir} \times 100

Örnek:

Bir şirketin toplam geliri 1.000.000 TL ve maaş artışı için bütçesi 100.000 TL ise:

Gerekli Gelir Artışı (\%) = \frac{100.000}{1.000.000} \times 100 = 10\%

Bu durumda, maaş zammını finanse edebilmek için şirketin gelirlerini %10 artırması gerekiyor. Aksi takdirde, kâr marjı ciddi şekilde düşebilir.

Maaş ve Verimlilik İlişkisi

Bir diğer önemli konu ise maaş artışlarının çalışan verimliliğine etkisidir. Bu noktada “Verimlilik/Maaş Oranı” değerlendirilebilir.

Verimlilik/Maaş Oranı = \frac{Üretim Değeri}{Toplam Maaş Maliyeti}

Örnek:

Bir şirketin üretim değeri 5.000.000 TL, toplam maaş maliyeti ise 2.000.000 TL ise:

Verimlilik/Maaş Oranı = \frac{5.000.000}{2.000.000} = 2.5

Eğer maaş maliyetleri artarken bu oran düşerse, maaş zamlarının iş verimliliğine olumlu bir katkı sağlamadığı sonucuna varılabilir.

Maaş artışlarının etkilerini analiz ederken bireylerin yetkinlikleri, işverenlerin maliyet yönetimi ve piyasa koşulları göz önünde bulundurmak keşke kolay olsa!. Modern ekonomilerde maaş artışlarının yalnızca çalışanların mevcut maaşlarına ve enflasyona bağlı olmadığı, aynı zamanda çalışanların beceri seviyelerine, üretkenliğe ve kurumsal hedeflere de dayandığı konuşuluyor. Asil mesele zaten iki taraflı verimlilik.

Beceri faktörleri, karmaşık üretkenlik modelleri, ve maaşın yan unsurlarını içeren maliyet kalemleri düşünelim. Ek olarak, örnek senaryolar ve maaşın tüm bileşenleri üzerinde bi duralım .

Bir çalışanın sahip olduğu beceriler ve bunların iş yerine katkısı, maaş artışlarının temel belirleyicilerindendir. Özellikle yüksek vasıflı çalışanların ücretlerinde artış yaparken işverenler, beceri kazandırma maliyetlerini de hesaba katmalıdır.

örnek ve en temel bir hesaplama:

Beklenen Maaş = Temel Maaş \times (1 + Beceri Katsayısı)

Bu durum, yüksek beceriye sahip bir çalışanın daha fazla kazanç elde etmesini sağlar ama bu indeksin adil olmasını nasıl sağlarız? bu soru zor soru ve İK için bir çıkmaz sokak!

Şirketler, çalışanlarına beceri kazandırmak için eğitim programlarına yatırım yapar. Bu maliyetler uzun vadede verimliliği artırırken kısa vadede mali yük getirebilir.

Eğitim Maliyeti = Eğitim Programı Ücreti + Çalışanın Eğitim Süresindeki Maaşı

• Eğitim programı maliyeti 5.000 TL ve eğitim süresince çalışan maaşı 3.000 TL ise:

Eğitim Maliyeti = 5.000 + 3.000 = 8.000 \, TL

Eğitim yatırımları, üretkenliği artırdığı takdirde maliyetini amorti edebilir. Üretkenlik artışı, maaş artışlarının sürdürülebilirliğini sağlar. Ancak üretkenliğin ölçülmesi, çok boyutlu bir analiz gerektirir.

Üretkenlik Katsayısı = \frac{Toplam Çıktı}{Çalışan Sayısı \times Çalışma Saatleri}

• Toplam çıktı: 1.000.000 TL

• Çalışan sayısı: 50

• Çalışma saatleri: 160 (aylık)

Üretkenlik Katsayısı = \frac{1.000.000}{50 \times 160} = 125 \, TL/saat

Bu katsayı, çalışan başına saatlik verimliliği temsil eder ve zam kararlarını etkileyebilir.

Toplam Üretkenlik Skoru = (Maaş Artışı \times Üretim Değeri Katsayısı) – (Ek Maliyetler \times Yan Haklar)

• Maaş artışı: 2.000 TL

• Üretim katsayısı: 1.5

• Ek maliyetler: 500 TL

• Yan haklar (yemek, ulaşım): 300 TL

Toplam Üretkenlik Skoru = (2.000 \times 1.5) – (500 + 300) = 3.000 – 800 = 2.200 \, TL

Pozitif bir skor, zamların üretkenliğe katkı sağladığını gösterir.

Maaş artışları yalnızca net maaş üzerinden düşünülmemelidir. İşverenler, tüm maliyet kalemlerini dikkate alarak daha geniş bir hesaplama yapmalıdır.

Maaş Bileşenleri

1. Net Maaş: Çalışanın eline geçen tutar.

2. Sigorta Primleri: İşverenin ödediği SGK ve işsizlik sigortası primleri.

3. Vergiler: Gelir vergisi ve damga vergisi.

4. Yan Haklar: Yemek, ulaşım, özel sağlık sigortası gibi ek ödemeler.

5. Primler: Performans ve dönemsel prim ödemeleri.

Tüm Maliyet Kalemlerinin Hesaplanması

Formül:

Toplam Maliyet = Net Maaş + Sigorta Primleri + Vergiler + Yan Haklar + Ek Masraflar

Örnek:

Bir çalışanın net maaşı 20.000 TL ve diğer maliyetleri aşağıdaki gibidir:

Rakamları takılmamak için çok düz hesap yapalım:

• SGK ve işsizlik primi: %20 (5.000 TL)

• Gelir ve damga vergisi: %15 (3.000 TL)

• Yemek ve ulaşım: 5.000 TL

• Performans primi: 1500 TL (yıllık aylıka bölürsek)

Ama temel soru şu, bu maliyetleri şirket nasıl karşılayacak… tabii karşıladı bunu nasıl doğru şekilde yansıtabiliriz?

Peki bunun için şeffaf maaş mümkün mü?

Maaş ne kadar olsa kimler güvenir rakamlara?

standart bir kazanç mi serbest piyasa mı?

Maaşların şeffaf bir şekilde açıklanması, hem çalışanlar hem de işverenler için tartışmalı bir konu olmaya devam etmektedir. Şeffaf maaş, bir organizasyondaki tüm çalışanların maaşlarının açıkça bilinir hale getirilmesi anlamına gelir. Bu uygulama, özellikle sosyal eşitlik ve iş tatmini açısından olumlu veya hatta olumsuz etkiler, yanlış yönetildiğinde çalışanlar arasında huzursuzluk ve verimlilik kaybına da yol açabilir.

ama diyelim bu sisteme geçsek? nasıl yürüyebiliriz?

Şeffaf Maaş,

pros and cons!

a. Eşitlik ve Adaletin Sağlanması

• Şeffaf maaş politikası, çalışanlar arasında ücret eşitsizliklerini önler ve adalet algısını artırır. Ama tabii doğru ve etik bir yönetim şart. Aynı iş için farklı maaş ödenmesi durumlarının önüne geçerek, iş tatminini destekler.

b. Güven Artışı

• Çalışanlar, organizasyonun adil olduğunu hissettiğinde iş yerindeki bağlılık artar. Ama bu durun pek imkansız gibi gelir. Genelde grading veya ölçme insanların kafasında soru işaretleri oluşturur.

• İşe alım sürecinde şirketin etik değerlerini gösterir, yetenekli adayları çekebilir.

c. Üretkenlik ve Performans

Çalışanların performanslarını iyileştirme motivasyonu artabilir. Kıyaslama yapılabilir ve her çalışanın gelişim alanları daha net belirlenebilir. Ama bu konu çok kolay değil, kâğıt üzerinde dökmek pek olmaz ama neyse!

Şeffaf Dezavantajları

a. Rekabet ve Huzursuzluk Riski

Aynı pozisyonda çalışan bireyler arasındaki küçük farklar bile huzursuzluk yaratabilir. Çalışanlar, sadece maaşa odaklanarak diğer yan hakların değerini göz ardı edebilir.

b. Piyasa Dengesizlikleri

• Şeffaf maaş politikası, çalışanların başka iş fırsatlarını değerlendirirken rekabetçi tekliflerle karşılaşmalarına neden olabilir.

c. Maliyet

Eşitsizlikleri düzeltmek için maaş artışları gerekebilir, bu da organizasyonun maliyetlerini artırır.

Şeffaf maaş sisteminin uygulanabilirliğini değerlendirmek için bir ekonomik model geliştirebiliriz. Bu modelde:

1. Maaş Şeffaflığı: Çalışan maaşları açıkça bilinir hale gelir.

2. Beceri ve Performansa Dayalı Artışlar: Ücretler, yalnızca kıdem değil, aynı zamanda beceri ve üretkenlik skorlarına dayanır.

3. Maliyet Analizi: Yan haklar, vergiler, eğitim maliyetleri gibi tüm bileşenler dikkate alınır.

Matematiksel Model

1. Ortalama Maaş Maliyeti:

Toplam Maliyet = (Ortalama Maaş + Ek Yan Haklar) \times Çalışan Sayısı + Düzenleme Maliyetleri

2. Üretkenlik Dengesi:

Net Üretkenlik = (Toplam Çıktı – Toplam Maliyet) / Çalışan Sayısı

3. Adalet Katsayısı (Şeffaflık Etkisi):

Şeffaf maaş sistemi, organizasyona adalet katsayısı (AK) ile katkı sağlar. Bu katsayı, çalışan bağlılığı ve üretkenlik artışını temsil eder:

AK = \frac{Memnuniyet Artışı}{Huzursuzluk Maliyeti + Maliyet Artışı}

4. Uygulanabilirlik Şartı:

Şeffaf maaşın ekonomik olarak sürdürülebilir olması için şu şartın sağlanması gerekir:

Net Üretkenlik + AK > 0

4. Örnek Senaryo: Şeffaf Maaş Sistemi Uygulaması

Senaryo Verileri

Eğer temel maaş 25.000 TL olarak belirlenirse, şeffaf maaş sisteminin etkisini değerlendirmek için aynı modeli güncelleyerek detaylı bir hesaplama yapabiliriz.

1. Güncellenmiş Veriler ve Giriş

Varsayılan Veriler:

• Çalışan sayısı: 100

• Temel maaş: 25.000 TL

• Yan haklar (yemek, ulaşım, sigorta): 5.000 TL

• Eğitim maliyeti: 500 TL/çalışan

• Şeffaflık düzenleme maliyeti: 50.000 TL (veri düzenlemesi ve iletişim maliyeti)

• Üretkenlik artışı beklentisi: %10

• Mevcut üretim değeri: 5.000.000 TL

• Memnuniyet artışı: %5

• Huzursuzluk maliyeti: %2

• Maliyet artışı: %1

2. Hesaplamalar

a. Toplam Maliyet

Maaş ve Yan Haklar:

Toplam Maaş = (Temel Maaş + Yan Haklar) \times Çalışan Sayısı

Toplam Maaş = (25.000 + 5.000) \times 100 = 3.000.000 \, TL/aylık

Eğitim ve Düzenleme Maliyetleri:

Eğitim Maliyeti = Eğitim Maliyeti \times Çalışan Sayısı = 500 \times 100 = 50.000 \, TL

Toplam Maliyet = 3.000.000 + 50.000 + 50.000 = 3.100.000 \, TL/aylık

b. Üretkenlik Artışı

Üretim değeri başlangıçta 5.000.000 TL olarak verilmiştir ve %10 artış beklenmektedir:

Yeni Üretim Değeri = Mevcut Üretim Değeri \times (1 + Üretkenlik Artışı)

Yeni Üretim Değeri = 5.000.000 \times 1.10 = 5.500.000 \, TL

c. Net Üretkenlik

Net üretkenlik, toplam üretim değeri ve toplam maliyet arasındaki farkın çalışan başına düşen kısmıdır:

Net Üretkenlik = \frac{Yeni Üretim Değeri – Toplam Maliyet}{Çalışan Sayısı}

Net Üretkenlik = \frac{5.500.000 – 3.100.000}{100} = \frac{2.400.000}{100} = 24.000 \, TL/çalışan

d. Adalet Katsayısı (AK)

Memnuniyet artışı, huzursuzluk maliyeti ve maliyet artışı oranları şu şekilde kullanılır:

AK = \frac{Memnuniyet Artışı}{Huzursuzluk Maliyeti + Maliyet Artışı}

AK = \frac{5}{2 + 1} = 1.67

e. Toplam Etki

Şeffaf maaş sisteminin ekonomik olarak uygulanabilirliği için Net Üretkenlik ve Adalet Katsayısı toplanır:

Toplam Etki = Net Üretkenlik + AK

Toplam Etki = 24.000 + 1.67 = 25.670 \, TL/çalışan

3. Sonuç ve Yorum

Pozitif Etkiler:

• Net kazanç: Şeffaf maaş sistemi, çalışan başına aylık 25.670 TL’lik bir ekonomik değer yaratmaktadır. Bu, sistemin üretkenlik artışı ile maliyetleri dengede tutabileceğini göstermektedir.

• Memnuniyet artışı: %5’lik bir memnuniyet artışı, çalışanların şirkete bağlılığını artırabilir ve uzun vadede daha düşük işten ayrılma oranlarına yol açabilir.

Negatif Riskler:

• Huzursuzluk: Aynı beceri seviyesindeki çalışanlar arasındaki küçük farklar dahi algılanan adaletsizliğe yol açabilir. Bu, üretkenlik artışını yavaşlatabilir.

• Maliyet baskısı: Yüksek maaş politikası, ekonomik dalgalanmalarda sürdürülebilirliği zorlaştırabilir.

Şeffaf maaş sistemi, temel maaşın 25.000 TL olduğu bir senaryoda, hem ekonomik hem de sosyal açıdan uygulanabilir gözükmektedir. Ancak, üretkenlik artışı ve çalışan memnuniyeti sürdürülebilir olmazsa, maliyetler hızlıca artabilir. Bu nedenle, şeffaf maaş politikası uygulanırken şunlar göz önünde bulundurulmalıdır:

1. Beceri bazlı maaş hesaplama: Adil bir algoritma geliştirilmelidir.

2. Yan hakların vurgulanması: Maaş dışında verilen destekler, çalışanlara açıklanarak maliyetler kontrol altında tutulabilir.

3. Verimlilik ölçümü: Üretkenlik artışını sürekli izlemek, sistemin sürdürülebilirliğini sağlar.

Gizli Maaş Politikası: Avantajlar, Dezavantajlar ve Etkileri

Gizli maaş politikası, bir organizasyonda çalışanların maaşlarının açıkça paylaşılmadığı ve genellikle sadece ilgili kişiler (çalışan, işveren ve muhasebe) tarafından bilindiği bir sistemi ifade eder. Bu yaklaşım, birçok şirkette standart uygulama olsa da, şeffaf maaş politikasının giderek daha fazla tartışıldığı günümüzde eleştirilmekte ve sorgulanmaktadır. Gizli maaş politikası hem organizasyonel kontrol hem de çalışan ilişkileri açısından karmaşık etkiler yaratabilir.

Gizli Maaşın Avantajları

1. Organizasyonel Esneklik

• Şirketler, çalışanlar arasında performans, beceri veya kıdeme göre farklı maaş politikaları uygulayabilir.

• Özel anlaşmalar yapılmasına olanak tanır, böylece üst düzey yetenekler daha kolay elde tutulabilir.

2. Huzursuzluk Riskini Azaltma

• Maaşların gizli olması, çalışanların kendi maaşlarını başkalarıyla kıyaslamasını önleyebilir ve bu durumdan kaynaklanabilecek huzursuzlukları engelleyebilir.

• Şirket içindeki kıskançlık, çatışma ve rekabet riskini düşürür.

3. Rekabet Avantajı

• Şirketler, piyasa rekabeti göz önünde bulundurularak belirli çalışanlar için stratejik maaş politikaları geliştirebilir.

• Dışarıdan gelen rakip şirketlerin maaş stratejilerini öğrenmesi zorlaşır.

Gizli Maaşın Dezavantajları

1. Adalet Algısının Zedelenmesi

• Çalışanlar arasında maaş farklılıkları olduğunda, bilgi eksikliği adaletsizlik algısına yol açabilir.

• Çalışanlar maaşlarının piyasa koşullarına uygun olup olmadığını sorgulamaya başlayabilir.

2. Çalışan Bağlılığının Zayıflaması

• Gizlilik politikası, çalışanların şirkete olan güvenini zedeleyebilir. Maaşın adil olduğuna dair şeffaf bir güvence olmadığı için bağlılık azalabilir.

3. Toplam Maliyetlerin Artması

• Maaş politikalarının gizli olması, her bireyle özel maaş müzakeresi yapma gerekliliğini artırabilir ve bu durum uzun vadede daha yüksek maliyetlere yol açabilir.

• Çalışanlar arasında farklılıkların keşfedilmesi işten ayrılmaları artırabilir ve işe alım maliyetlerini yükseltebilir.

Ekonomik Model: Gizli Maaşın Etkisi

1. İşe Alım ve Tutundurma Maliyeti

Gizli maaş politikası, işe alım sürecinde pazarlık gücünü artırsa da, yüksek yetenekli çalışanları kaybetme riskini de beraberinde getirir.

Formül:

Toplam Maliyet = Mevcut Maaş + İşe Alım Maliyeti + İşten Ayrılma Oranı \times Yeniden Eğitim Maliyeti

Eğer çalışanlar arasında huzursuzluk varsa, işten ayrılma oranı artar ve gizli maaş politikası daha maliyetli hale gelir.

2. Performans ve Motivasyon

Maaşlar gizli olduğunda, çalışanların kendi maaşlarını başkalarıyla kıyaslayamaması olumlu bir motivasyon yaratabilir. Ancak belirsizlik, uzun vadede çalışanların performansını ve iş tatminini düşürebilir.

Örnek Senaryo: Gizli Maaşın Uygulanması

Senaryo Verileri:

• Çalışan sayısı: 100

• Ortalama maaş: 25.000 TL

• İşten ayrılma oranı: %10

• İşe alım maliyeti: 20.000 TL/çalışan

• Eğitim maliyeti: 10.000 TL/çalışan

Maliyet Hesaplaması:

1. İşten Ayrılma Maliyeti:

İşten Ayrılma Maliyeti = İşten Ayrılma Oranı \times Çalışan Sayısı \times (İşe Alım Maliyeti + Eğitim Maliyeti)

İşten Ayrılma Maliyeti = 0.10 \times 100 \times (20.000 + 10.000) = 300.000 \, TL

2. Toplam Maliyet:

Toplam Maliyet = Ortalama Maaş \times Çalışan Sayısı + İşten Ayrılma Maliyeti

Toplam Maliyet = 25.000 \times 100 + 300.000 = 2.800.000 \, TL/aylık

Sonuç:

Gizli maaş politikasının toplam maliyeti 2.800.000 TL’ye ulaşmaktadır. İşten ayrılma oranı düşük tutulursa, bu maliyet kontrol edilebilir. Ancak, ayrılma oranı artarsa, maliyetler hızla yükselir.

Gizli Maaş mı, Şeffaf Maaş mı?

Karşılaştırma:

Kriter Gizli Maaş Şeffaf Maaş

Adalet Algısı Düşük Yüksek

Huzursuzluk Riski Düşük Yüksek (Yanlış yönetilirse)

Maliyet Kontrolü Daha esnek Uzun vadede daha sabit

Çalışan Bağlılığı Orta Yüksek

Gizli maaş sistemi, özellikle dinamik maaş politikalarına ihtiyaç duyan küçük ve orta ölçekli şirketler için uygun olabilir. Ancak, çalışan bağlılığı ve motivasyonunu artırmayı hedefleyen organizasyonlarda şeffaf maaş politikası daha etkili olacaktır. İdeal çözüm, şeffaflık ve gizlilik arasında bir denge kurmak olabilir: Örneğin, maaş aralıklarının açıklandığı ancak bireysel maaşların gizli tutulduğu bir sistem benimsenebilir.

Intro:

After two years at Kariyer.net, I started thinking about how AI and data could shape the future of job platforms. In the next few years, job boards might evolve into systems that not only help people plan their careers but also offer advanced talent intelligence. By analyzing user data, these platforms could suggest courses, predict future job trends, and match individuals with the right opportunities in real-time. This could completely change how we match people with jobs.

Why the trinity;

Türkiye’s job market is at a pivotal moment, driven by rapid digital transformation and a growing demand for smarter, more efficient recruitment solutions. The combination of AI, data, and HR is emerging as a powerful force, reshaping how employers and job seekers connect. As platforms like Kariyer.net lead the way, this “future trinity” has the potential to revolutionize recruitment. AI brings intelligence to decision-making, data provides insights for better strategies, and HR ensures a human-centered approach to hiring. Together, they can go beyond traditional job matching, focusing on talent intelligence — using data to analyze an individual’s skills, potential, and career trajectory to connect them with the right roles and career development paths.

Pillar 1: Data/rich and meta!

In the digital GPT-age, people wants more personalized recommendations. The job board metadata play a crucial role in connecting job seekers with employers. They form the backbone of modern recruitment platforms, enabling efficient matching of skills to opportunities. By organizing and analyzing this data, job boards can provide personalized recommendations, streamline hiring processes, and offer valuable insights into labor market trends.

People Data is gold mine?

People data refers to structured and unstructured information about individuals, such as:

• Profiles: Basic details like name, age, contact information, and location.

• Skills: Technical and soft skills acquired through education, training, or experience.

• Experience: Job history, including previous roles, achievements, and industries worked in.

• Preferences: Desired job type, location, salary expectations, and career goals.

This data helps identify a candidate’s suitability for specific positions and supports dynamic, data-driven decision-making.

1. Position Information:

• Job title

• Role description

• Employment type (full-time, part-time, contract)

2. Skills and Requirements:

• Technical skills (e.g., programming, machine operation)

• Soft skills (e.g., teamwork, communication)

• Education level or certifications required

3. Company Details:

• Employer name and industry

• Location or remote work options

4. Other Metadata:

• Salary range

By standardizing people data and job board metadata, platforms can:

1. Improve Search and Matching: Algorithms can efficiently pair job seekers with relevant opportunities.

2. Provide Insights: Identify demand trends, in-demand skills, and wage benchmarks.

3. Personalize Experiences: Offer tailored job suggestions or candidate recommendations.

4. Automate Hiring: Simplify tasks like shortlisting and interview scheduling.

Pillar 2: AI in Action: What This Means for Turkey

Türkiye presents unique opportunities and challenges for AI-driven job boards.

The agriculture and construction sectors often face labor shortages during peak seasons. By combining people data and job metadata, platforms like İşin Olsun could proactively scan and engage youth workforce. Then can help employers set realistic wage expectations by analyzing regional salary data. As Turkey emerges as a hub for tech startups and digital transformation, AI can support young professionals by:

• Mapping out in-demand skills like Python, AI/ML, and cloud computing.

• Offering targeted training programs based on career goals.

For example, a fresh computer science graduate in Ankara might benefit from AI-powered career guidance, directing them to roles at tech parks or internships with multinational companies.

The job market is becoming increasingly diverse, with more women entering the workforce so reducing bias in hiring algorithms to ensure fairer recruitment practices.

The Future: AI + Data = Better Work for Everyone

Pillar 3: HR mindset and literacy is much more important than data and AI. You have a good car but you should be a good driver.

So what we can do:

The goal isn’t just to match candidates with jobs – it’s to create a labor market where every skill is valued, every opportunity is accessible, and every person finds their purpose.

As we look ahead, the intersection of AI, data, and HR will continue to reshape the workforce, especially in Turkey, where rapid changes in industries and skill sets demand greater flexibility and responsiveness.

So we need to agile and Data-minded HR people!

Dynamic Job Descriptions

Job descriptions will no longer be static; they will evolve in real-time based on changing market demands, skill shifts, and employer needs. Powered by AI and data, dynamic job descriptions will allow companies to adjust the scope and requirements of a position quickly, enabling them to stay competitive and attract the right talent.

Real-Time Role Adjustments: As industries evolve, so too will job requirements. If a new technology emerges or a shift in the market occurs, job descriptions will be automatically updated to include the latest skills or qualifications. For instance, the rise of AI-driven technologies could make data analysis or machine learning a mandatory skill in roles previously not requiring it.

Localized Customization:

Job descriptions will become hyper-localized based on geography, tailoring requirements to the specific needs of different regions in Turkey. For example, Istanbul might see greater demand for digital marketing expertise, while Ankara may prioritize roles in public administration or technology development.

Skill Mapping:

Using data from the job market, AI will adjust job descriptions by suggesting the most relevant skills for candidates. These dynamic updates will not only improve the matching process but also guide candidates to acquire in-demand skills, helping them stay ahead of market trends.

Adjustable Salary Models

Gone will be the days of rigid salary bands. With real-time data and AI, salaries will become more adjustable and dynamic, reflecting individual skills, market conditions, and the specific needs of both the employee and employer.

• Market-Based Adjustments: AI will analyze labor market trends and adjust salary expectations based on supply and demand for specific roles. If a role becomes more in-demand, such as cloud computing specialists or cybersecurity experts, the salary for that position will automatically increase in the system, ensuring it remains competitive.

• Skill-Based Salary Ranges: Salaries will be more flexible, with systems that account for an individual’s unique skill set and experience. AI-driven platforms will match candidates to salary ranges based on their actual competencies and achievements, ensuring fair compensation that reflects the value they bring to an organization.

• Regional Salary Adjustments: Salaries will be adjusted not only by role but also by geographic location. For example, Istanbul may offer higher salaries for tech roles compared to smaller cities in Turkey, reflecting the local cost of living and competition for talent in those areas.

• Performance-Based Flexibility: In addition to static salary structures, AI-powered platforms may suggest salary flexibility based on real-time performance data, suggesting bonuses or pay increases tied to a candidate’s job performance, skills growth, or company profitability.

Multiple Job Roles and Career Path Flexibility

As the workforce becomes more agile and diverse, employees and candidates will no longer be confined to a single job title. The future of HR will be characterized by multiple job roles and career path flexibility, offering workers more choices and opportunities for growth.

• Cross-Functional Roles: Companies will encourage employees to take on multiple roles, combining tasks across different functions (e.g., a data scientist who also handles product management or a developer who leads team training). AI systems will help suggest cross-functional roles that match the individual’s skills, improving employee engagement and career satisfaction.

• Role Evolution: With the rise of dynamic job descriptions, roles will constantly evolve. For example, marketing roles could include more data analytics or tech skills over time. Employees who begin in sales could transition into marketing or product development, all while remaining within the same organization.

• Talent Mobility: AI-driven platforms will make it easier for employees to move between roles, creating a more flexible and dynamic career path. AI will use data to suggest possible transitions or promotions within an organization, allowing workers to explore new opportunities without the need to change employers.

• Career Path Recommendations: AI will recommend specific career paths based on the individual’s experience, market demand, and personal interests. For instance, if an employee in a customer service role demonstrates strong analytical skills, AI might suggest a shift towards a business intelligence or data analysis career.

Job Role Redefinitions

As industries evolve and technology advances, some traditional roles may become obsolete, while new roles emerge. AI and data will play a key role in predicting these changes and suggesting role replacements before they happen.

Role Transitioning: By analyzing labor market trends, AI can predict when certain job functions will become obsolete and suggest replacement roles. For example, data entry jobs may decline due to automation, and AI could guide workers toward roles in data analysis or AI programming.

• Reskilling: AI will help identify which employees may need to upskill or reskill in order to transition into new roles. Based on labor market predictions, employees will be offered targeted training programs to prepare them for emerging job roles, such as AI specialists or blockchain developers.

• Job Role Forecasting: AI can predict the rise of new job titles and create opportunities for workers to shift into roles that don’t yet exist but will become essential in the future. For example, new roles like AI ethics officers or remote work coordinators may emerge as AI technology and flexible work arrangements continue to grow.

Predictions for Blue-Collar Jobs and Logistics

Despite the growing influence of automation and AI, blue-collar jobs – especially in physical labor and logistics – will remain central to Turkey’s economy. AI and automation will transform how blue-collar jobs are performed, especially in logistics and manufacturing. For example: AI will forecast peak demand based on historical data, ensuring companies can deploy workers effectively. This will improve shift planning and workforce utilization, especially during busy seasons or unexpected surges in demand.

Personalized Career Paths: Platforms like Kariyer.net will use AI to recommend training and certifications based on workers’ existing skills and market demand.

Flexibility in Job Roles: Cross-Functional Roles: In logistics, workers may move between roles such as warehouse management, driving, and inventory oversight depending on demand.

Conclusion

As AI, data, and HR continue to evolve, the future of the job market will be far more flexible, responsive, and tailored to both employer needs and employee career goals. Dynamic job descriptions, adjustable salary models, multiple job roles, and job role replacements will redefine how the workforce operates in Türkiye.

At Kariyer.net, we are preparing for this future by incorporating AI-driven platforms that provide real-time updates to job descriptions, adjust salary expectations, suggest career paths, and predict skill requirements. This transformation will empower both employers and employees to thrive in an increasingly competitive and changing market, making recruitment more efficient and career development more personalized than ever before.

What are your thoughts on the future of AI and job boards? Do you think Turkey’s labor market is ready for this transformation? Let me know in the comments below!

The Future Trinity: AI + Data + HR in Türkiye was originally published in Kariyer.net Tech on Medium, where people are continuing the conversation by highlighting and responding to this story.

Intro

OpenAI (GPT models) and other large language models use a sequence of characters called tokens to charge you. These models don’t see text the way we do; instead, they see it as a series of numbers, or tokens.

In very basic terms, the method to convert text (words or tokens) into binary or vice versa is called Byte Pair Encoding (BPE). This method combines or separates tokens to make the processing more efficient.

Token is a basic unit that fall somewhere between characters and words. For instance, the word “tokenizing” might be split into “token” and “izing” as two separate tokens.

Keep in mind aslight change in wording can lead to a different tokenization and, consequently, a different response from the model. Knowing how to structure your prompts to get the desired results involves a good grasp of tokenization principles. OpenAI models learn to understand the statistical relationships between tokens. A sequence of tokens leads to the generation of the next token.

The cornerstone of pricing of the model you use is based on tokens. Clearly, the more tokens you use in your input or produce output, the higher the cost.

This makes understanding tokens crucial not just for using AI effectively, but also for managing costs efficiently. Let’s discuss what is the best toolkit for calculation of the cost of OpenAI. I mean the working with APIs!:

First stratrgy

You have two options based on your production by which you can stream or batch the OpenAI models. Batch is more economical and 50% lower price for tasks like clustering, classification or any thing that you do not need for a real-time responses.

To learn more about Batch Pricing you can visit: https://platform.openai.com/docs/guides/batch/overview

Second: Metrics

You should have some pre-defined metrics to track regularly in time period. For example size of token (context or generated), limits etc. Consider, if you take into regard metrics like

Token Usage:

• Total Tokens Consumed: the total number of tokens used in API calls.
• Tokens per Request: the average number of tokens used per API request.
• Input vs. Output Tokens: Differentiates between tokens sent in the request and tokens received in the response.

API Latency:

• Response Time: Measures the time taken for the API to respond to a request.
• Request Processing Time: Tracks the time taken by the model to generate a response.

Cost Metrics:

• Cost per Token: Calculates the cost incurred for each token used.
• Total API Costs: Tracks the overall expenditure on API usage over a specific period.
• Cost per Request: Measures the average cost per API request.

Usage Patterns:

• Request Frequency: Monitors the number of API requests made over time.
• Peak Usage Times: Identifies the times when API usage is highest.

Performance Metrics:

• Accuracy and Quality of Responses: Evaluates the relevance and correctness of the responses generated by the model.
• Error Rates: Tracks the number of failed or erroneous requests.

Scalability Metrics:

• Concurrent Requests: Monitors the number of simultaneous API requests being handled.
• Throughput: Measures the number of requests processed per unit of time.

These can be some pre-defined metrics you can use for any LLM models.

1. Tokenizer Playground:

Before writing any prompt, check the token size here! This shows how a single character can be counted as a token. However, do not remove delimiters if you need them, as they can be important for guiding your prompt as system or role indicators.

Here, the split of colors shows the unit of tokens. For instance, the “!” symbol can have a different location if we add a single space.

If you read carefully the reference of OpenAI, you can easily know the best tacticts and with error and trial you will have your optimum size of prompt. For output you can play with Max_Token parameters.

2. Tiktoken Library

This is at least my bible to learn token!. A github repo as https://github.com/openai/tiktoken can help you to understand the functionality of tokens. Knowing the number of tokens in a text string helps determine (a) if the string is too lengthy for a text model to handle and (b) the cost of an OpenAI API call, as pricing is based on token usage This library can help you to count tokens or encoding it. You can check libraries based on the stack you use:

Tokenizer libraries by language

For cl100k_base and p50k_base encodings:

• Python: tiktoken
• .NET / C#: SharpToken, TiktokenSharp
• Java: jtokkit
• Golang: tiktoken-go
• Rust: tiktoken-rs

For r50k_base (gpt2) encodings, tokenizers are available in many languages.

• Python: tiktoken (or alternatively GPT2TokenizerFast)
• JavaScript: gpt-3-encoder
• .NET / C#: GPT Tokenizer
• Java: gpt2-tokenizer-java
• PHP: GPT-3-Encoder-PHP
• Golang: tiktoken-go
• Rust: tiktoken-rs

3. openai-cost-tracker 0.5 (Python Library)

You can use the lightweight wrapper that helps you track the cost of each request. This tool not only handles your API interactions but also calculates how much each request will cost you. You need to install package in your IDE.

Installation

pip install openai-cost-tracker

Usage

Import the query_openai function from openai_cost_tracker:

from openai_cost_tracker import query_openai

You can follow the instructions here: https://pypi.org/project/openai-cost-tracker/

4. Dashboard of Platform.OpenAI (Excel, GoogleSheet or Shiny!)

4.1 Logic

The simple way is that you can export your activity logs periodically to track and analyze your costs based on the API keys you use. This allows you to get a comprehensive view of your expenditure. By employing simple calculation functions like count or sumif, you can precisely compute the costs associated with each individual API key. This approach helps in managing and optimizing your API usage efficiently.

4.2 Real Time Dashboarding

The best part is you can create a real-time dashboard with Shiny in Python. If you can fetch with import or maybe directly adding to a Shiny for Python, you can analyze more details.

You can check and learn shiny here: https://shiny.posit.co/

An example of endpoint: Connect (https://api.openai.com/v1/dashboard/billing/usage) to a Shiny application, you can use httr package to make HTTP requests and retrieve the data or in Python as def.

def fetch_api_data(api_key, start_date, end_date):
    url = "https://api.openai.com/v1/dashboard/billing/usage"
    headers = {
        "Authorization": f"Bearer {api_key}"
    }
    params = {
        "start_date": start_date,
        "end_date": end_date
    }

5. OpenAI observability tools (with extra cost for you!)

Not only can you monitor incidents, but you can also use the following tools to analyze more details of cost. Here are some available platforms or open-source tools:

a) Grafana: you can monitor, check and track incidents of openai models here. Check: https://grafana.com/solutions/openai/monitor/

b) new relic : To integrate OpenAI Observability alerts with your favorite company communication tools like Slack, or others, you typically follow incidents and cost volume.

c) whylabs: you can may use more functions to monitor or control.

d)Lakera: has a well-written guide for monitoring and observability of LLM models if you use more than one models.

e)Guardrails: Guardrails is a Python framework designed to enhance the reliability of AI applications. It achieves this by performing two main functions: These guards detect, quantify, and mitigate specific types of risks present in your application. They help ensure that inputs and outputs are managed in a way that minimizes potential vulnerabilities and errors.. This capability helps in organizing and utilizing data effectively within AI applications.

And for sure, if you search you can find more like Langchain or other open-source tools on Huggingface. But my motivation is just help you to find the suitable tools or approach you may want to use.

Final tip: You can use OpenAI CookBook or Forum to learn more. If you keep you updated, I am sure you will learn more tips there!.

More Resource may you want to look at:

https://lilianweng.github.io/

https://github.com/openai/tiktoken/blob/main/README.md

https://help.openai.com/en/articles/4936856-what-are-tokens-and-how-to-count-them

https://neptune.ai/blog/tokenization-in-nlp

https://tiktokenizer.vercel.app/

https://bea.stollnitz.com/blog/how-gpt-works-technical/

• Shiny Templates
• Monitor your OpenAI usage with Grafana Cloud | Grafana Labs

Toolkit to Calculate /Predict the Cost of Using OpenAI’s API Models. was originally published in Kariyer.net Tech on Medium, where people are continuing the conversation by highlighting and responding to this story.

Doing text analysis in Google Sheets for HR data is easy and time-saving, provided you have < 1 million records. While handling large datasets in spreadsheets isn’t ideal, for simplicity and efficiency, preprocessing can be done effectively. You can use a combination of built-in functions, Google Apps Script, or both.

Intro

Here, I’d like to share what I’ve learned through trial and error to make it easier for anyone looking to perform analysis in Google Sheets.

I recommend Google Sheets as it is more versatile than Excel, at least for data cleaning, based on my own experience. However, it’s up to you if you prefer to proceed with Excel.

Built-in-Functions

This link is my bible: https://support.google.com/docs/table/25273?hl=en. You can filter on text funtions to see examples and how to use the built-in functions. Lets get started with what are key text data in HR:

I created a dummy data with ChatGPT with only by my names and ex-startup:

Start with basics:

Before digging into solve the problem, first think how you can address the data and extract meaningful insight. Do not write formulas juts to convince yourself, instead focus on the real problem of your HR department or HR product.

Start with the Basics:

For example, if you want to know what is the overall pattern of job titles or count of any text data in your company, then, focus on utilizing functions and their combinations to extract meaningful insights from the give text. Lets start with alphabets:

Identifying Key Problems and Goals

Examples:

Understanding Job Titles:

• What are the most common job titles in my company?
• How do job titles vary if we consider the word counts?

Count of Specific Job Titles: Identify how many employees hold a specific job title like “Product Manager”.

Pattern Detection in Job Titles: Find common prefixes or suffixes in job titles, such as “Senior” or “Junior”.

Text Functions to Use

Here you go for some text functions that can help you solve these problem sets by built-in functions in Google Sheets. Imagine you have a text in (A2) cell as a simple example.

Functions for Analyzing Job Titles

LEN: It calculates the length of job titles in form of character size. Here product manager has 15 characters.

=LEN(A2)

LEFT, RIGHT, MID: Extract parts of job titles to analyze common prefixes or suffixes. For instance, if you know the patterns and want to extract words like suffix or prefix, you can use LEF function as below.

=LEFT(A2,7)

In a same way you can use MID or RIGHT if you have different pattern or want to do in a different positions.

Word Counting for Job Titles

Now imagine you want to calculate the count of a job titles seperated by spaces as you see here. Then you need to do a simple substraction method:

=IF(A2<>” ”, LEN(A2)-LEN(SUBSTITUTE(A2, “ “, “ ”))+1, 0)

• A2<>"": Checks if cell A2 is not empty. This prevents errors in case A2 is empty.
• LEN(A2): Returns the total number of characters in A2.
• SUBSTITUTE(A2, " ", ""): Removes all spaces from A2.
• LEN(A2) - LEN(SUBSTITUTE(A2, " ", "")): Calculates the difference in length between the original text and the text with spaces removed, which gives the total number of spaces.
• +1: Adds 1 to the count because the number of words is always one more than the number of spaces.

Example:

• If cell A2 contains: “Product Manager” as title, the formula will count 2 words.

You can do more by the other built-in functions as below

FIND, SEARCH: Locate specific words or characters within job titles.

• =FIND("Manager", A2) =SEARCH("Engineer", A2)

SPLIT: Divide job titles into separate words to analyze each part.

• =SPLIT(A2, " ")

CONCATENATE, TEXTJOIN: Combine text from multiple cells for more complex analysis.

=CONCATENATE(B2, " - ", C2) =TEXTJOIN(", ", TRUE, A2:A10)

By using these functions and focusing on your specific goals, you can start to extract meaningful insights from your HR data. For instance, if you need to analyze job titles, you might:

• Use LEFT and RIGHT to identify common prefixes and suffixes.
• Apply SPLIT to break down titles into individual words for pattern analysis.
• Use COUNTIF to see how many times each title appears.

Similarly, for skills analysis:

• Use SPLIT to break down comma-separated skills into individual entries.
• Apply COUNTIF and COUNTUNIQUE to understand the frequency and variety of skills.
• Combine these functions with others like LEN and SEARCH to gain deeper insights.

In the following, I will write more about script as you can play more and more with text or strings!

I hope you enjoy reading this and that it brings something new to you.

Let’s face the reality for a minute. We all know that the job market can be tough, and building a successful career can be even tougher. However, if you’re interested in the people data job family and have a passion for using data to drive decisions, you should understand the differences and similarities between various fields in this domain.

The field of people data is diverse, including not only areas like People (HR) analytics, workforce analytics, employee experience, but also financial, marketing, sales, and operational roles that rely on peole data too. There is a huge demand for skilled professionals who can use data to create insights that inform decision-making about individuals and people. Whether you’re just starting out or looking to advance your career, the opportunities in people data can be broader.

At the intersection of human resources and data-driven decision-making, it’s essential to understand the differences between various sub-fields: People Data, People Analytics, HR Analytics, and other related areas each have their own focus and applications. In this article, we’ll explore the distinctions between these sub-fields and help you determine which one aligns best with your interests and career goals.

People data is a field that involves managing, analyzing, and modeling any information or data related to individuals, such as employees, customers, unemployed job seekers, people who wants to vote, immigrantes etc. This can include a wider range of persona than only professional information. It encompasses more demographic data, education and labour market data or even behavioral (preference) data of any person.

One example of blending people data with financial and economic data is the analysis of the labor market or revenue generation. By combining data on employee performance, skills, and demographics with financial and economic indicators, businesses can better understand the impact of their workforce on company revenue and make more strategic decisions about hiring, sales, growth plan or exectutive plans. Let’s discuss it in more detail at Part II. (next article) and first roadmap the Poeple Analytics (PA) and other sub-fields:

People Analytics, HR Analytics, and other related areas each have their own focus and applications like employee experience, data visualization and reporting in organization cuaght more interest these days. But unfortunalty people misunderstand the real functionality or make a trivial understanding of it.

People Analytics is a sub-discipline within the field of people data that primarily focuses on leveraging data of employees or potential candidates to comprehend and enhance various aspects of the employee or candidate experience. The definition can be interpreted in different words; but one thing is fixed: interpretation of insights (if persona is only employees and candidates). You may hear the definition a lot, but I intentionally insist on this:

You should make a report/data/pattern or any idea to prove evidence-based decisions and optimize the business outcomes. People Analytics is not a novel concept, but its use has grown exponentially in recent 5 years. We are not sure that the term “People Analytics” was coined by Google or Wharton but we should remember that they are pioneers. As an interesting event in the 2011, Kathryn Dekas opens the opportunities of this and since then, it has become a crucial tool for organizations looking to optimize their human capital and drive better business outcomes. The use of People Analytics dates back to at least the 2010s, but its evolution has been significant over the past decade. To this we should add Wharton People Analytics Conference annually held at Upenn (Pensyllvania University) since 2013.

In the Strata Events, Kathryn Dekas presents “People Analytics: Using Data to Drive HR Strategy and Action,” and underscores the significance of using data to inform HR strategy and decision-making. Dekas, who was a People Analytics Manager at Google at the time, elucidates how Google leverages People Analytics to augment organizational efficacy, employee engagement, and overall performance. The presentation encompasses various aspects of People Analytics, such as identifying key metrics, collecting and analyzing data, and using the insights gained to make informed decisions.

This shows how the rise of People Analytics has also given birth to various platforms and tools that assist organizations in better comprehending and managing their workforce. To name a few Visier, Glint (owned by LinkedIn), Peakon (acquired by Workday), Orgnostic, ADP and others provide specialized analytics solutions that enable businesses to collect, analyze, and act upon employee data. These platforms and other aid organizations in gaining insights into areas such as employee engagement, productivity, skill gaps, and more. They also facilitate the development of remote work policies and implementation of strategies to support employee mental health and work-life balance.

As I mentioned People Analytics has evolved significantly over the past decade. It is now an increasingly indispensable sub-discipline of people data, with a growing number of organizations recognizing its value in driving better decision-making and enhancing the overall employee experience. But the level of maturity of People Analytics within an organization varies based on the scope, resources, and leadership support.

It is important to assess the organization’s readiness for People Analytics and determine the appropriate level of investment and development. However, a diverse technical skill set is required to extract insights from HR data. Use cases include machine learning applications such as recommendation systems and predictive models, as well as expertise in statistical analysis, data management, and visualization. Proficiency in programming languages, SQL, business intelligence tools, data warehousing, ONA, and graph analysis also may be crucial for success in this field if you want go further.

Where you should start?

Well, it depends on your actual skill (or intention to learn or team-up) and your company mindset and problems, but generally speaking, in addition to your communicating skills, you should gain and widen your technical background:

— — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — -+
| 1 | HR fundementals: at least main concept to drive problems.
| 2 | Data Management | Ability to collect, clean, and prepare data for analysis |
| 3 | Statistical Analysis | Knowledge of statistical methods and tools, such as regression analysis, hypothesis testing, and ANOVA |
| 4 | Data Visualization |Ability to create meaningful visualizations to effectively communicate data insights |
| 5 | Programming Languages | Proficiency in relevant programming languages, such as R or Python, to perform data analysis and modeling |
| 6 | SQL | Knowledge of SQL to extract, manipulate, and analyze data from databases |
| 7 | Machine Learning | Understanding of machine learning algorithms and techniques to build predictive models, such as: |
| 8 | Recommendation systems to personalize onboarding and training experiences for new hires |
| 9 | Predictive models to identify employees at risk of churning and develop retention strategies |
| 10 | Natural Language Processing (NLP) to analyze employee feedback and sentiment |
| 11 | Classification models to identify high-potential employees for leadership development programs |
| 12 | Spreadsheet |
| 13 | Proficiency in using spreadsheet software, such as Microsoft Excel or Google Sheets, to perform data analysis, create reports, and manipulate data |
| 14 | Business Intelligence Tools |
| 15 | Knowledge of business intelligence tools, such as Tableau or Power BI, to create interactive dashboards and reports |
| 16 | Data Mining |
| 17 | Ability to discover patterns and relationships in large datasets using data mining techniques |
| 18 | Data Warehousing |
| 19 | Understanding of data warehousing concepts and tools to manage large volumes of data |
| 20 | ONA (Organizational Network Analysis) |
| 21 | Ability to use social network analysis to understand the relationships and communication patterns between individuals in an organization |
| 22 | Graph Analysis |
| 23 | Understanding of graph theory and its applications to analyze complex relationships and networks in data |
| 24 | HR Metrics |
| 25 | Knowledge of HR metrics, such as employee turnover, employee engagement, performance appraisal metrics, and OKRs (Objectives and Key Results), to inform HR strategies and improve employee experience. |
| 26 | Financial Metrics |
| 27 | Knowledge of financial metrics, such as revenue per employee, cost per hire, and return on investment (ROI) of HR initiatives, to evaluate the financial impact of HR decisions and optimize resource allocation. |
+ — — + — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — —

How to start and keep practicing!

If you want to become a successful People Analyst, it is essential to avoid relying solely on blog-tailored websites or purely sharing and reading articls,. Instead, build a solid foundation of problem, a team to solve the real problem, break it to actions and find tools to catch data and write insights. You can find lots of GitHub repos, real textbooks or technical docs that cover a range of topics in data and HR.

Moreover, it is crucial to focus on problem-solving rather than memorizing a set of rules or techniques. This involves finding your own data or generating synthetic data using data generators and applying analytical techniques to solve real-world problems. Open AI APIs and ChatGPT can also be valuable tools for generating outstanding problem-solving ideas and modifying proposed code.

Here are some examples of practical People Analytics problems that you can tackle:

• Predicting employee turnover rates and identifying factors that drive employee retention
• Analyzing employee engagement and identifying drivers of engagement
• Identifying skills gaps within the workforce and developing strategies to address them
• Evaluating the competitiveness to redesign EVP (employer branding case)

Becoming a successful People Analytics practitioner requires a combination of technical and HR management skills. By following the tips outlined above and practicing problem-solving with real-world data, you can develop the skills and knowledge needed to succeed in this rapidly growing field.

Is ONA complex?

Employee experience and Organizational Network Analysis (ONA) are two critical areas within the field of People Analytics for understanding and improving the overall employee experience in an organization.

Employee experience encompasses all interactions between an employee and an organization throughout their employment journey, including onboarding and offboarding, learning and development opportunities, and career development. By leveraging People Analytics, organizations can gather and analyze employee data to gain insights into the factors that drive employee engagement, satisfaction, and productivity. This can help identify areas for improvement and develop targeted strategies to enhance the overall employee experience.

ONA is a method of analyzing the relationships and communication patterns between individuals in an organization. By visualizing these relationships, organizations can identify key influencers and communication hubs, as well as potential communication bottlenecks or silos. ONA can help understand the informal networks and communities within an organization, promoting collaboration and knowledge sharing. Ultimately, ONA can optimize social capital and improve the effectiveness of the workforce.

Employee experience can analyze and optimize various aspects of workforce development, such as onboarding and offboarding, learning and development, and skill analysis. By using data-driven insights, organizations can identify gaps in employee development programs and tailor training initiatives to meet specific workforce needs. This helps employees develop new skills and improve their job performance, leading to better business outcomes.

Combining insights from Employee Experience, ONA, and workforce development enables organizations to make data-driven decisions, enhancing the overall employee experience and organizational productivity.

“The Handbook of Regression Modeling in People Analytics: With Examples in R and Python” by Keith McNulty is another resource for those interested in using data analysis techniques in People Analytics. While the book primarily focuses on regression modeling techniques, it also includes a section on Organizational Network Analysis (ONA).

ONA can be a complex topic, and the handbook provides a comprehensive guide on how to apply ONA techniques in the context of People Analytics. The book covers topics such as data preparation, network metrics, and network visualization. Additionally, the book provides practical examples of how to use ONA to address common People Analytics problems, such as identifying key influencers and understanding communication patterns within an organization.

Becoming a successful People Analytics practitioner requires a combination of technical and HR management skills. By following the tips outlined above and practicing problem-solving with real-world data, you can develop the skills and knowledge needed to succeed in this rapidly growing field.

Imagine you have collected an employee_feedback data in csv format. If you challenge yourself by a basic sentiment analysis of feedback, you can advance yourself with a more complex problems than calculation KPIs. (you can even use co-pilot tools like GPT-tools like codex to generate code by prompt). Then, next time you can create a team to advance the level of maturity with reproductitity and upscale.

An (generated) snippet as example of feedback analysis:

library(tidytext)
library(dplyr)
library(ggplot2)

# Import employee feedback data
feedback_data <- read.csv("employee_feedback.csv", header = TRUE, stringsAsFactors = FALSE)

# Clean and process the feedback data
feedback_data <- feedback_data %>%
  unnest_tokens(word, Feedback) %>%
  anti_join(stop_words) %>%
  filter(str_detect(word, "[a-z']")) %>%
  mutate(word = wordStem(word))

# Perform sentiment analysis
sentiments <- get_sentiments("afinn")
feedback_sentiments <- inner_join(feedback_data, sentiments, by = "word")

# Calculate sentiment scores for each feedback entry
feedback_scores <- feedback_sentiments %>%
  group_by(Feedback_ID) %>%
  summarize(sentiment_score = sum(value))

# Visualize the sentiment scores using a histogram
ggplot(feedback_scores, aes(x = sentiment_score)) +
  geom_histogram(binwidth = 1) +
  ggtitle("Sentiment Analysis of Employee Feedback") +
  xlab("Sentiment Score") +
  ylab("Count")

In this example, you can collect and make the employee feedback data and process it using the tidytext package. Then we use the get_sentiments function to import a pre-built sentiment dictionary and join it with the processed feedback data. Finally, you can calculate sentiment scores for each feedback entry and visualize the distribution of scores using a histogram as a very simple example of reproduction of People analytics!

Reproducibility refers to the ability to replicate and verify the results of a project. This is critical to ensuring the accuracy and reliability of the insights gained from the project. To achieve reproducibility, it’s important to clearly document the methods, tools, and data used in the project and make this information accessible to others who may wish to replicate or build upon the project in the future.

Scalability refers to the ability to apply the methods and insights gained from a project to larger or more complex datasets or scenarios. A scalable project should be designed to accommodate larger datasets or more complex scenarios without compromising the accuracy or speed of the analysis.

Privacy is also a critical consideration in People Analytics projects. Any project that involves the collection, analysis, or storage of personal data must comply with relevant laws and regulations governing data privacy and protection. It’s important to ensure that the project is designed with appropriate data security measures and that any personal data is handled in a responsible and ethical manner.

Criteria

Examples

Reproducibility

Use of version control (e.g. Git) for tracking changes in HR data, use of standardized formats for storing and sharing HR data, creation of data dictionaries and metadata to aid in reproducibility

Scalability

Use of distributed computing (e.g. Spark) for handling large-scale HR data, use of cloud computing (e.g. AWS) for scalable data storage and processing, optimization of code to handle large datasets efficiently

Privacy

Adherence to data privacy regulations (e.g. GDPR, CCPA) when collecting and storing HR data, use of anonymization techniques (such as de-identification, aggregation, or masking) to protect the privacy of HR data, use of secure data storage and transmission protocols to prevent unauthorized access to HR data.

References:

If you want to ask me what books worth to start; I can list these three:

People Analytics for HR Decisions: A Guide for HR and Management Professionals- Rahul Ghatak

Modern Survey Analysis- Walter R. Paczkowski

Advancing into Analytics: From Excel to Python and Machine Learning, George Mount

The Fundamentals of People Analytics: With Applications in R

Additionally, it’s worth noting that while People Analytics and HR Analytics share some commonalities, there are also important differences between the two fields. In some cases, the use of HR Analytics may overlap with People Analytics, but the latter is typically broader in scope and more focused on using data-driven insights to optimize the employee experience and drive better business outcomes. As such, we have primarily focused on People Analytics in this article.

In the next article we will back to People data role, Stay tuned!

The brief of history 1894-2024.

(Updating regularly!, last: 23/12/2023)

The salary data, rates, or any information, news, estimation and scenarios are among the economic buzzwords of the end of 2023 and beginning of 2024 in the world and specifically in Türkiye, followed by anyone older than 16 (will explain why!), or perhaps in any location/economic situation. The actual min wage is 11402 TL== 393 USD (if 29 TL = 1 USD).

Queries like ‘What will be the minimum wage for the next mid-period of adjustment (the end of 2023 and beyond)’ are hard to answer with acute accuracy, but a projection of possible increase (%x) is attainable if we follow trends. Is salary a right or a preference? I would like to challenge this question. Can we obtain any freedom or power mechanism of preference to adjust our wage/salary/compensation, or is it not possible by any means?

The answer is not straightforward. We certainly impose a preference, but the output can be leveraged differently. It depends on your experience, your industry or job category, the trend of the market, the company you are in, your image or brand, your footprint, skills, your perceived education, the network you have built, or the fear of the company losing your expertise.

The incremental rates for the new year adjustment are estimated as a range of 35%-45% or even higher. But the higher increase may make happy people only in the short-term! They will pay back more with the multi-effect of inflation! In addition, the high rate of min wage is not good for new entrants of the labour market as employers may prefer experienced to young talent!. The dismissal again can cause higher youth unemployment.

For those working in the government, not all cases are possible and are just bound to years of experience. It’s not enough to claim that ‘salary would be one of the cores of HR products’ this year, but it is at least essential and a ‘please-do-it!’

Let’s dive into the history of minimum wage and then in Türkiye.

Interesting facts/figures:

• New Zealand is the first country to accept and execute the min wage, while UK accepted in 1909.
• “Sweatshops” were one of the ignitors concerning minimum wage in US.
• The difference between “remuneration”, “wage”, “income”, “salary”, and “bonus” is important to understand the definition of minimum wage.

The term “wage” is generally understood to be the payment an employer makes to his or her employees.

• To accept and implement you should be a member of international labour organization-ILO
• The minimum wage is a mechanism to encourage governments where no agreements or wages are exponentially low. 186 out of all countries (195) are the members. EU is a full member, while Arab states still relatively hold lower rate of participation (70%).

Asgari (in Arabic the minimum or elder kid! ) vs Azami (maximum bundle) can be defined for salaries but in reality, the maximum wage does not reflected so far at least for all job categories (we will discuss the free market in detail too in next updates!).

The eligibility for receiving minimum wage is 16 years old according to labour law.

How many people are over +16 in Turkey and -60 (average pension age):

The first and a must-factor to consider for wage/salary in all-time is Population!

Age structure projection in Türkiye (the ageing trend is very visible!)

If we exclude the -16 and + 60 as non-eligible age boundaries for working, then the total population to receive a possible salary (for simplicity we nullify now the economical factors), the total size of the population goes around 60 M. Young and Elderly age structure are going to intersect in the 2040s. This is not far and we are closing 15 years later!. But the problem is not the working population, instead, I feel the skilled, agile and productive working structure- age groups.

The trend shows Türkiye becomes slightly older next to +20 years later on, but keeps the working age around (16–63) as the working age bundles. In total 10 M + 40 M but in 2043 it simply will be intersected as lower than 20 M and we have 10 years older Türkiye!

Inflation vs. CPI and Exchange rate (USD)

To be honest, Türkiye tries to impose a higher adjustment than the announced inflation rate, but the problem is the trust in rates!. Can we trust to TÜİK (Turkstat) or basket of CPI or even the YoY inflation rates?

Let’s look at different angles: Why Türkiye is forced to increase twice a year.

• The net minimum wage has increased substantially over the years, indicating potential adjustments to cope with inflation but the is behind what happens in real inflation

• The inflation rates are relatively high, suggesting economic challenges and the need for adjustments in wages to maintain purchasing power, but a consumption-addicted country like Türkiye can not balance the purchasing power and inflation.

But all time we discuss the increase of minimum wage in one way in details, how about employer’s concern. The cost is increasing hugely!

Imagine a company only paying minimum wages and considering the size. You can see the cost of HR is doubled only in one year!. If it is not the signal of huge unemployment is a signal of preference of experienced people over the new alumni!. This year the cost can be astronomical too as people expected more than last year!

I do not want to advocate employer in any means, but I prefer to have a sustainable employer-employee pay-work relation.

Personally speaking, we should worry about the ever-increasing pay gap. The pay gap in extent of psychological effect, productivity and all-size-fits-all is very dangerous. I will devle more into in series 2!

References:

• Minimum wages (Minimum wages)
• Asgari Ücret

Carr, S. C. (2023). Wage and Well-being: Toward Sustainable Livelihood. Springer Nature.